Endometrial carcinosarcoma is a rare and aggressive high-grade endometrial carcinoma with secondary sarcomatous trans-differentiation (conversion theory). The clinical presentation and diagnostic work-up roughly align with those of the more common endometrioid counterpart, although endometrial carcinosarcoma is more frequently diagnosed at an advanced stage. Endometrial carcinosarcoma is not a single entity but encompasses different histological subtypes, depending on the type of carcinomatous and sarcomatous elements. The majority of endometrial carcinosarcomas are characterized by p53 abnormalities. The proportion of POLE and microsatellite instablity-high (MSI-H) is directly related to the epithelial component, being approximately 25% and 3% in endometrioid and non-endometrioid components.
The management of non-metastatic disease is based on a multimodal approach with optimal surgery followed by (concomitant or sequential) chemotherapy and radiotherapy, even for early stages. Palliative chemotherapy is recommended in the metastatic or recurrent setting, with carboplatin/paclitaxel doublet being the first-line regimen. Although the introduction of immunotherapy plus/minus a tyrosine kinase inhibitor shifted the paradigm of treatment of patients with recurrent endometrial cancer, patients with endometrial carcinosarcoma were excluded from most studies evaluating single-agent immunotherapy or the combination. However, the US Food and Drug Administration (FDA) and the European Medicines Agency (EMA) approved the use of pembrolizumab and lenvatinib in endometrial cancer (all histotypes) after progression on chemotherapy and single-agent immunotherapy in MSI-H cancers. In the era of precision medicine, emerging knowledge on molecular endometrial carcinosarcoma is opening new promising therapeutic options for more personalized treatment. The present review outlines state-of-the-art knowledge and future directions for patients with endometrial carcinosarcoma.
- uterine cancer
- genital neoplasms, female
Data availability statement
No data are available.
Statistics from Altmetric.com
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.
Endometrial carcinosarcoma is a high-grade endometrial carcinoma with secondary sarcomatous de-differentiation.
Approximately 70% of cases of endometrial carcinosarcoma are classified as p53 abnormal, being characterized by an aggressive nature.
Endometrial carcinosarcoma with endometrioid component are more likely to exhibit ultra- and hypermutator (POLE and MSI-H) subtypes.
MSI-H/dMMR (accounting for 7% in endometrial carcinosarcoma) represents an emerging biomarker, suggesting the efficacy of immunotherapy, even in endometrial carcinosarcoma.
Novel molecular-targeted therapies are emerging that could potentially improve care.
Endometrial carcinosarcoma is a rare and aggressive high-grade endometrial carcinoma, accounting for about 5% of all uterine malignancies and nearly 20% of non-endometrioid endometrial cancer.1 2 Although non-endometrioid histotypes account for 10–20% of all endometrial cancers, they are responsible for more than 40% of endometrial cancer-related deaths. In particular, endometrial carcinosarcoma is responsible for 15% of deaths from uterine malignancies.1 2
Endometrial carcinosarcoma is an intriguing entity as it is a biphasic tumor characterized by coexisting carcinomatous (epithelial) and sarcomatous (mesenchymal) elements.3 It is diagnosed at an advanced stage more often than other endometrial cancers. The stage at diagnosis follows a bimodal distribution: 40–50% of cases are early stage (International Federation of Gynecology and Obstetrics ((FIGO) I–II) and 50–60% are advanced (FIGO III–IV). Up to 30–40% of patients present with lymph node metastases at diagnosis, and 10% have distant metastatic spread, especially in the lungs.2–5 Over 60% of patients with apparently early-stage disease at the time of initial diagnosis are upstaged following comprehensive surgical evaluation due to occult metastatic spread. Despite the multimodal treatment strategy (surgery, platinum-based chemotherapy, radiotherapy), prognosis remains poor. The median overall survival is less than 2 years, and the 5-year overall survival rate is less than 30% (about 50% and 20% in early and advanced stages, respectively). Even patients with early-stage disease have a 5-year recurrence rate of 45% and 5-year related mortality of 50%.2–5
The increasing incidence and poor outcomes of endometrial carcinosarcoma underscore an unmet need for novel therapeutic strategies to treat these challenging patients. Prior to the current millennium endometrial carcinosarcoma was considered a sarcoma, and it was not included in trials on endometrial cancer. Moreover, due to the rarity of endometrial carcinosarcoma, epidemiological studies and high-quality evidence are scarce and future international collaborative projects in this field are warranted. This comprehensive review summarizes the state-of-the-art knowledge on the clinical features of, and treatment options for endometrial carcinosarcoma, with a focus on the most recent molecular updates and promising therapeutic targets on the horizon
Epidemiology and Clinical Characteristics
Endometrial carcinosarcoma is a rare gynecologic cancer, but its incidence has been gradually increasing over the past two decades, with an annual percentage growth rate of nearly 2%.5–7 Accordingly, the proportion of endometrial carcinosarcomas within all endometrial carcinomas has also grown significantly from 1.7% to 5.6%.5–7 Since endometrial carcinosarcoma occurs, almost exclusively, in post-menopausal women (usually over 60–65 years of age), the aging global population may in part explain this increase in incidence, together with a raised awareness of endometrial carcinosarcoma by pathologists. Endometrial carcinosarcoma typically affects the elderly with a peak incidence between 70 and 79 years of age.5–7 However, in recent years the age of affected patients has decreased, with patients aged 60–69 years demonstrating the largest change in incidence rate (annual increase: 2.7%).5–7 The average age of patients with endometrial carcinosarcoma at diagnosis is currently 67 years. Apart from age, other risk factors for endometrial carcinosarcoma include black race, prior pelvic radiotherapy (eg, previous radiotherapy for cervical or rectal cancer), and, for endometrioid endometrial cancers, those factors leading to hyperestrogenism, such as obesity, nulliparity, exposure to exogenous estrogen and tamoxifen.5–7
The clinical presentation and diagnosis of endometrial carcinosarcomas are non-specific and typically similar to those of most endometrial carcinomas.2 7 In clinical practice, it is difficult to distinguish endometrial carcinosarcoma from other uterine neoplasms merely based on clinical features. Typically, symptoms of endometrial carcinosarcomas include persistent or post-menopausal abnormal uterine bleeding, leukorrhea, and/or abdominal pain associated with a rapidly growing fleshy uterine mass (often bulging into the vagina). Other symptoms, such as dysuria, dyspareunia, and bone pain are rarer.6–8 Endometrial biopsy or biopsy of a protruding polypoid mass is a key element for the diagnosis of endometrial carcinosarcoma. It is important to highlight that in a few cases endometrial sampling might reveal only one of the two components (carcinomatous and sarcomatous), and the final diagnosis is obtained only after hysterectomy. Transvaginal ultrasound, pelvic MRI, (thoracic and abdominopelvic) CT, and/or positron emission tomography are beneficial as imaging techniques for diagnostic and staging purposes. The basal level of serum CA125 correlates with an advanced stage and poor prognosis, and it may be useful to guide not so much the diagnosis but the follow-up.8 9 The metastasis pattern of endometrial carcinosarcomas follows the lymphatic and intraperitoneal routes as in epithelial tumors more than the typical hematogenous dissemination of sarcomas, and metastases are usually of epithelial origin. Endometrial carcinosarcomas are characterized by aggressive behavior with a 5-year survival rate of 25–30% (stage I: ~55%; stage II: ~37%; stage III: ~25%; stage IV: ~10%), which has not changed much over the past three decades.2 5–8 The prognosis correlates strongly with the histologic subtype, tumor size (≥5 cm in ~60%), FIGO stage, lymphovascular space involvement (~60%), post-surgical residual disease, malignant peritoneal cytology, and the molecular signature as well as treatment. These survival statistics suggest the need for further research and approaches to the management of endometrial carcinosarcomas.2 5–7
Pathological Features: the Conversion Theory
The pathological classification of endometrial carcinosarcoma has changed over time. Historically, endometrial carcinosarcoma was regarded as a malignant mixed Müllerian tumor and the most common and aggressive type of uterine sarcomas.10 Presently, however, endometrial carcinosarcoma is widely recognized as an epithelial de-differentiated/metaplastic subset of endometrial cancers and it is staged and managed accordingly, as a high-grade endometrial carcinoma.10 Therefore, pathologists should accurately identify the presence of an epithelial component before making the final diagnosis of endometrial carcinosarcoma. Endometrial carcinosarcoma is a biphasic malignant tumor consisting of endometrial adenocarcinomas admixed with a mesenchymal component.10
The epithelial part is the most dominant element and is typically a high-grade (serous, endometrioid, clear cell, mixed, or undifferentiated) histotype, whereas the sarcomatous element can be either homologous (leiomyosarcoma, fibrosarcoma, endometrial stromal sarcoma) or heterologous (rhabdomyosarcoma, chondrosarcoma, osteosarcoma), according to whether the mesenchymal component resembles or not the uterine tissues. Heterologous differentiation is seen in about 40% of endometrial carcinosarcoma and is associated with poorer survival compared with homologous alterations. Moreover, sarcomatous dominance (ie, >50%), is seen in 40% of cases of endometrial carcinosarcoma and is most probably associated with heterologous differentiation and decreased survival. The combination of high-grade carcinoma, and heterologous sarcomatous differentiation and dominance is associated with the worst prognosis.10
Figure 1 shows the pathological characteristics of endometrial carcinosarcoma. Interestingly, few studies reported a higher 5-year overall survival in patients with endometrioid endometrial carcinosarcoma than in those with non-endometrioid (serous, clear cell) subtypes (50–55% vs 30–35%, respectively).6–8 10 Moreover, it is important to point out that the epithelial component is more commonly observed at distant metastatic sites; the sarcomatous component is associated with local tumor extension.10
The epithelial and sarcomatous components were initially thought to develop as a combination of cellular masses secondary to an early divergence from a common precursor cancer stem cell (combination theory) or as a result of the collision between independent but adjacent epithelial and mesenchymal progenitors (collision theory).8 10 Recently, several molecular and clonality studies have suggested that the endometrial carcinosarcoma arises from a single malignant epithelial clone (carcinoma lineage) that subsequently undergoes sarcomatous trans-differentiation, through a process of epithelial-to-mesenchymal transition (conversion theory).10–12 The monoclonal origin of endometrial carcinosarcoma is supported on genetic, molecular, and clinical grounds.10–13 The epithelial and mesenchymal elements share common genetic mutational profiles, and stromal cells often show positive immunohistochemical staining for epithelial markers.10–13
The majority of endometrial carcinosarcomas share molecular and genomic similarities with high-grade serous ovarian carcinoma and serous endometrial carcinoma, while only a minority resembles the endometrioid counterpart.3 4 10 12–15 In particular, TP53 (60–97%) FBXW (10–44%), PPP2R1A (11–30%), HER2 (9–18%) serous-like mutations are common, whereas endometrioid-like mutations such as ARID1A (10–25%), KRAS (8–15%), PTEN (10–50%), and PIK3CA (20–40%) are less frequent.3 4 10 12 The mutational rates vary across the studies depending on the different endometrial carcinosarcoma subtypes included (eg, low-grade vs high-grade carcinomas). Endometrial carcinosarcoma is not a single entity but encompasses different histological subtypes, depending on the type of carcinomatous and sarcomatous elements.10
Thanks to the analyses of The Cancer Genome Atlas (TCGA) Research Network and the Proactive Molecular risk classifier for Endometrial cancer (ProMisE) classification, four novel molecular endometrial cancer subgroups were identified.16 17 The new classification includes: POLE/ultramutated (POLE mutated), microsatellite-instable/hypermutated (MSI-H), copy-number-high/TP53-abnormal (P53-abn), and copy-number-low/TP53-wild-type or non-specific molecular profile endometrial cancers. This classification overcomes the limitation of the dualistic Bokhman model, representing an excellent tool for prognostication and treatment recommendation.10 14 15 However, we have to point out that no prospective randomized studies ave validated the predictive value of adoption of the genomic/molecular profiling (with the except of MIS-H, being an agnostic marker supporting the adoption of immunotherapy).18 Validation studies are still ongoing.14 15 19
Interestingly, the TCGA study included only the endometrioid and serous histotypes while little is known regarding less common endometrial cancer histotypes, such as endometrial carcinosarcoma.16 Recently, a meta-analysis of four studies (231 patients) reported the pooled prevalence of the TCGA groups among endometrial carcinosarcomas: 5.3% POLE, 7.3% MSI-H, 73.9% p53-abnormal, and 13.5% non-specific molecular profile.4 20–23 The vast majority of endometrial carcinosarcoma (73.9%) are classified within the serous-like, p53-abn risk group (which accounts for 5–15% of endometrial cancers and resembles type II endometrial cancers). As aforementioned, those tumors are characterized by advanced stage at diagnosis, late-onset, mutant-like/abnormal p53 immunohistochemical staining, low mutational burden (<10 mutations per megabase), aggressive behavior, high risk of early relapse, and a dismal prognosis.
Another recent meta-analysis of five studies (263 patients) assessed the prognostic value of the TCGA molecular classification in endometrial carcinosarcoma. POLE mutated endometrial carcinosarcoma showed an excellent prognosis similar to that of POLE mutated endometrioid endometrial cancers, supporting their inclusion in the same low-risk category for treatment purposes in the current European Society of Gynecological Oncology (ESGO), the European SocieTy for Radiotherapy and Oncology (ESTRO), and the European Society of Pathology (ESP) guidelines.9 On the other hand, the prognosis of p53-abn and non-specific molecular profile endometrial carcinosarcoma was even worse than that of their endometrioid/serous counterparts, while that of MSI-H/dMMR tumors was unclear and remains to be clarified.4 20–27 Table 1 shows the current evidence regarding the molecular/genomic profiling of endometrial carcinosarcomas.
Due to the rarity of endometrial carcinosarcoma, there is only limited evidence regarding current standard of care, largely from retrospective or non-randomized studies.7 No standard and definitive consensus on the optimal management of endometrial carcinosarcoma exists. Since endometrial carcinosarcoma is now considered a primary endometrial carcinoma, its treatment aligns with that of other non-endometrioid high-grade endometrial cancer, as suggested by the ESGO/ESTRO/ESP and the National Comprehensive Cancer Network (NCCN) guidelines.9 28 A multimodal approach, combining surgery, chemotherapy, and/or radiotherapy is the current mainstay of treatment. Given the absence of solid data, adequate patient counseling should be always offered.
Figure 2 outlines the surgical treatment algorithm for endometrial carcinosarcoma. Complete surgical staging is the standard treatment approach for non-metastatic endometrial carcinosarcoma. Standard surgical procedures include hysterectomy, bilateral salpingo-oophorectomy, infracolic omentectomy, peritoneal biopsies, peritoneal cytology, and only for early stages, retroperitoneal staging (eg, systematic (pelvic and para-aortic) lymphadenectomy or sentinel lymph node biopsy). Peritoneal cytology is not mandatory as it is not a cancer staging factor, but it can be useful as a risk factor for tailoring the adjuvant treatment.7 9
Minimally invasive surgery is the preferred surgical approach in apparently early-stage disease and should be undertaken cautiously, avoiding tumor fragmentation and peritoneal dissemination.9 If vaginal extraction risks uterine rupture, other measures should be taken (eg, mini-laparotomy, use of endobag).8 9 In advanced stages (FIGO III–IV), open abdominal cytoreductive surgery should be considered when complete macroscopic resection is feasible, with an acceptable morbidity and quality of life profile.8 9 Retrospective studies suggest that suboptimal debulking does not confer additional survival benefit over chemotherapy alone in endometrial carcinosarcomas, thus thorough patient selection is key, and surgery should be pursued only if complete macroscopic resection can be achieved.9 28 A few small studies have also investigated the potential role of neoadjuvant protocols, such as platinum-based chemotherapy or concurrent chemoradiotherapy, to increase complete resection rates, reporting interesting results warranting prospective large-scale validation.29 30 Ovarian preservation and fertility-sparing surgery are not recommended for endometrial carcinosarcoma.7 9
Infracolic omentectomy and random peritoneal biopsies are considered part of the surgical staging even for apparent stage I endometrial carcinosarcoma, such as in serous and undifferentiated histotypes.8 9 No specific data regarding the role of peritoneal staging in patients with endometrioid and non-endometroid component exists.
Nodal involvement is not uncommon in endometrial carcinosarcoma (pelvic: 20–25%, para-aortic: 15%), especially in cases of deep myometrial invasion (30–50% of cases) where pelvic and para-aortic nodal metastases are observed in 30–60% and 25–30% of cases, respectively.31–33 Only resection of enlarged lymph nodes, but no systematic lymphadenectomy, is recommended for advanced stages.8 9 On the other hand, systematic lymphadenectomy (up to the level of the left renal vein) has been traditionally recommended in patients with early-stage endometrial carcinosarcoma as a staging procedure because of the high prevalence of occult (no gross) nodal metastases. However, the therapeutic role of systematic lymphadenectomy has been questioned over the past decade.
Sentinel node mapping, which is already accepted in low- and high-risk endometrial cancer, has emerged as an alternative for nodal staging, even in endometrial carcinosarcoma.33 Several studies have demonstrated the efficacy and safety of sentinel node mapping in patients with high-risk endometrial cancers, with an acceptable false-negative rate (less than 1%).31 32 With particular regard to endometrial carcinosarcoma, in 2016, Schiavone et al reported no significant differences in progression-free survival between patients with endometrial carcinosarcoma undergoing sentinel node mapping versus standard lymphadenectomy (23 vs 23.2 months, respectively; p=0.7).34
More recently, Zammarrelli et al, compared the oncologic outcomes of 99 patients with endometrial carcinosarcoma who underwent sentinel node mapping with those of 100 patients receiving systematic lymphadenectomy, thus confirming that sentinel node mapping can detect nodal metastasis without compromising oncologic outcomes.31 However, further prospective studies with longer follow-ups are required to validate these early retrospective results.
The FIRES trial demonstrated that sentinel node mapping with indocyanine green has a high degree of diagnostic accuracy in detecting nodal metastases and can safely replace systematic lymphadenectomy for endometrial cancer staging (any histotype).35 Recently, the multicenter, prospective SENTOR trial showed that sentinel node mapping (using indocyanine green) in high-risk endometrial cancers, such as endometrial carcinosarcoma, had comparable, if not superior, diagnostic accuracy to those of systematic lymphadenectomy.36 However, only a few cases of endometrial carcinosarcoma were included in these two latter trials, thus limiting the applicability to this rare histotype. Additional prospective trials (SNEC, ALICE, ENDO-3, ECLAT) are ongoing and will provide further high-quality evidence on nodal staging in high-risk endometrial cancers.37 To date, there is no consensus on the necessity and extent of lymphadenectomy in patients with early-stage disease who had a hysterectomy without nodal surgical staging and present negative imaging. However, it seems reasonable to avoid another surgery and consider adjuvant radiotherapy in addition to chemotherapy to target the nodal areas at risk.
Adjuvant (Multimodal) Treatment
Owing to the rarity of endometrial carcinosarcoma, no clear consensus exists regarding the optimal adjuvant therapy (after surgery) for patients with endometrial carcinosarcoma. Lacking randomized clinical trials, the benefit of adjuvant therapy is not fully understood and recommendation for subsequent treatment should be considered on a case-by-case basis after multidisciplinary discussion. In general, endometrial carcinosarcoma should be treated as high-risk carcinomas (not as sarcomas).9 Compared with radiotherapy alone, chemotherapy and, even more, (concurrent or sequential) chemoradiation have been proved to reduce the risk of recurrence and improve survival rates at all stages.38–45 Until better treatment options become available, the best approach is multimodal with both chemotherapy and radiotherapy (Table 2).38–70 Notably, as for all high-risk histotypes, chemotherapy plays a central role in the management of patients with endometrial carcinosarcoma.68
The adjuvant treatment is defined based on both traditional and molecular classifications, which identify different prognostic risk groups (Table 3).9 28 FIGO stage IA endometrial carcinosarcoma without myometrial invasion falls within the intermediate-risk group, in the absence of a POLE mutation.9 The presence of a POLE mutation is rare in endometrial carcinosarcoma but, if present, it determines a subclassification in the low-risk group where adjuvant treatment is not mandatory, at least in the early stages. Endometrial carcinosarcomas with myometrial invasion are considered at high risk, irrespective of the stage and the molecular profile.9 28
The adjuvant treatment usually consists of chemotherapy and radiotherapy, although the optimal sequencing (concurrent vs sequential) remains unclear. Some studies revealed an improvement in survival rates at all stages when using a ‘sandwich approach’ (chemotherapy–radiotherapy–further chemotherapy) compared with alternate sequences (radiotherapy–chemotherapy or chemotherapy–radiotherapy).71 The rationale behind the greater benefit of sandwich sequencing can probably be explained through the following considerations: (1) chemotherapy is a priority in endometrial carcinosarcoma and should be administered upfront at the maximum dosage; (2) irradiation leads to vascular dysfunction and may impact the tumor delivery of chemotherapeutic agents, thus reducing the efficacy of following systemic treatments; (3) the sequential approach is more tolerable and allows higher dosages of both therapies to be provided separately; (4) administering all six cycles of chemotherapy upfront may increase the risk of toxicity and cause a delay or dose reduction of subsequent radiotherapy, which is itself important for the locoregional control.72
Chemotherapy alone may be considered in patients with early-stage and locally advanced endometrial carcinosarcoma receiving surgery.73 74 Although patients with endometrial carcinosarcoma were not included in the Gynecologic Oncology Group (GOG) 122 and 258 studies, other experiences reported quite encouraging outcomes following surgery plus chemotherapy.73–75 However, it is important to point out that the omission of radiotherapy might correlate with an increased risk of pelvic recurrence.33 Patients with endometrial carcinosarcoma with residual disease after surgery should be managed with a multimodal approach including chemotherapy and/or radiotherapy. Palliative chemotherapy and the best supportive care should be considered for inoperable advanced endometrial carcinosarcoma.9 28 33 Genetic and molecular differences between non-endometrioid carcinomas and also within every single histotype are gradually emerging and may change therapeutic practices in the future.9
Table 4 outlines published prospective trials on the efficacy of chemotherapeutic and investigational agents both in the adjuvant and advanced/recurrent settings.75–100 Historically, the ifosfamide/paclitaxel doublet was regarded as the category 1 regimen based on several phase III prospective trials and, in particular, the GOG-161 trial, which supported the use of ifosfamide-based combinations over single-agent ifosfamide or cisplatin in patients with endometrial carcinosarcoma.75 83 However, based on the results from the GOG-232B and GOG-261 trials, the carboplatin/paclitaxel doublet has now been recommended as the preferred first-line treatment for endometrial carcinosarcoma, given the non-inferiority and the better toxicity profile, compared with ifosfamide/paclitaxel. Ifosfamide/paclitaxel and cisplatin/paclitaxel regimens remain alternative options (for instance, in cases of hypersensitivity reaction to carboplatin).75 83
Recently, based on the results of several trials, immunotherapy (with or without tyrosine kinase inhibitor) is emerging as the standard treatment modality after the failure of platinum-based chemotherapy. Pembrolizumab plus lenvatinib represents the preferred treatment for non-endometrioid endometrial cancer since they are generally characterized by MSS/pMMR disease.101 Since MSI-H/dMMR represents an agnostic biomarker, also those patients with non-endometrioid endometrial cancer characterized by MSI-H/dMMR are deserving of treatment with single-agent immunotherapy (eg, dostarlimab, pembrolizumab).102 103 Of note, most immunotherapy-based studies (also the KEYNOTE-775) do not include patients with endometrial carcinosarcoma.101 However, the US Food and Drug Administration (FDA) and the European Medicines Agency (EMA) approved the use of pembrolizumab and lenvatinib for endometrial cancer recurring after platinum-based chemotherapy, regardless of the histotype (the FDA just for MSS/pMMR disease and the EMA for all types of endometrial cancer).14 15 A more detailed description of the evidence for the role of immunotherapy is described below.
Before the adoption of immunotherapy (with or without tyrosine kinase inhibitor), no standard chemotherapeutic treatment has been identified as second-line therapy and the prognosis following recurrence is poor. Rechallenge with platinum-based chemotherapy (eg, carboplatin/paclitaxel, carboplatin/pegylated liposomal doxorubicin) can be considered.75–81 Other options include ifosfamide±paclitaxel, weekly paclitaxel, pegylated liposomal doxorubicin, docetaxel, gemcitabine, and topotecan. However, if the median response rate and progression-free survival are 37.5% and 5.9 months, respectively, after first-line therapy, the outcomes are far worse in subsequent lines (5.5% and 1.8 months, respectively). Even some studies investigating the efficacy of targeted therapies, such as tyrosine kinase inhibitors (eg, sunitinib, imatinib, sorafenib, pazopanib) have had disappointing results so far.104 Due to the lack of valid advanced-line therapies, enrollment of patients in future clinical trials is strongly encouraged.
Data on the efficacy of adjuvant radiotherapy (external beam radiation and/or brachytherapy) are limited and mostly retrospective (Table 1). Radiotherapy alone after surgery is not recommended as, despite the reported improvement in local control, it has not demonstrated a clinical benefit for survival compared with observation or chemotherapy alone in all stages.9 105–107 On the other hand, the rationale for a combined modality with both chemotherapy and radiotherapy is stronger.105–107 In particular, the clinical benefit of adding radiotherapy is particularly evident in the case of positive lymph nodes, unknown nodal status (ie, un-staged patients probably harboring occult metastasis), or sarcoma dominance, where it has been shown to both reduce locoregional recurrence and improve oncologic outcomes.9 28 62 The target volumes include the pelvis and nodal areas at risk as well as the vaginal cuff. The most common dose prescription is 45–50 Gy (1.8–2.0 Gy per fraction) external beam radiation treatment plus 6 Gy x three vaginal brachytherapy. When performed, vaginal brachytherapy should cover only the upper third (or the proximal 3–4 cm) of the vagina.61
Endocrine therapy may also be an alternative in the metastatic setting, especially in the case of frail, elderly women not suitable to receive further chemotherapy lines. To date, available data on hormonotherapy for endometrial carcinosarcoma are still scant, although there is anecdotal evidence on the efficacy of systemic progestins in the estrogen receptor/progestin receptor-positive setting, thus further investigation is suggested. However, since endometrial carcinosarcomas are most frequently high-grade undifferentiated tumors, they generally express low levels of hormone receptors (estrogen receptor: 20–30%; progestin receptor: 5–40%), and hormonal therapies should probably be reserved for challenging cases where there are no other valuable alternatives.9 28
The follow-up program for endometrial carcinosarcoma follows that of high-risk endometrial cancers. For the high-risk groups, physical and gynecological examinations are recommended every 3–4 months for the first 2 years, and then every 6 months until 5 years.9 28 108 Physical and radiological assessment are recommended on a regular basis. A CT scan every 12 months for the first 3–5 years (and then on an individual basis) can be considered, particularly if nodal involvement was present at diagnosis.9 28 108 Routine serum CA125 dosage is not recommended, although it can be useful to guide the decision-making process if it was elevated at diagnosis.9 28 Finally, Pap smears have not been shown to be useful for detecting local recurrences. When there are symptoms (eg, vaginal bleeding or discharge), an appropriate investigation should be carried out to exclude a recurrence.108
Despite surgical treatment and timely adjuvant multimodal therapy, more than half of the cases of endometrial carcinosarcoma will recur within the first 2 years.10 The management of the recurrent disease is highly personalized and should consider several factors, such as the performance status of the patient, the size and sites of recurrences, and prior therapies.9 10 28 Importantly, it depends on whether the relapse is locoregional, oligometastatic, or disseminated and, second, on whether the patient has already received radiotherapy, as radiotherapy rechallenge is generally avoided for safety reasons (Table 2). Again, the best treatment approach is multimodal. Patients with recurrent disease (including peritoneal and lymph node relapse) should be considered for surgery only if it is anticipated that complete removal of macroscopic disease can be achieved with acceptable morbidity and be treated in specialized centers.9 External beam radiotherapy can be used in radiotherapy-naïve patients or those who had received only prior vaginal brachytherapy. Immunotherapy (with or without tyrosine kinase inhibitor) is the emerging preferred second-line systemic treatment. After the failure of immunotherapy, chemotherapy alone (generally mono-chemotherapy) is the preferred treatment in cases of disseminated metastases.10 33 Due to the poor outcomes associated with standard treatments for relapses, enrollment in clinical trials is highly recommended.
Novel Therapeutic Agents and Future Perspectives
In this era of precision medicine, there is an unmet clinical need to better understand the pathogenesis and molecular landscape of endometrial carcinosarcoma. It is now clear that not all endometrial carcinosarcomas can be managed in the same way, as endometrial carcinosarcoma is not a single entity but instead can display various genetic, molecular, and histologic profiles. The ability to address the molecular characterization of each singular tumor may open new therapeutic horizons for endometrial carcinosarcoma and help to overcome the poor prognosis. In light of the new molecular classification and raised awareness of the pathogenesis, endometrial carcinosarcoma is being progressively included in endometrial cancer clinical trials after being understudied for several years. However, as we know, conducting prospective clinical trials for rare and heterogeneous tumors is challenging while more concrete support can probably come from multicenter retrospective studies or basket trials. Future trials should focus on the efficacy of pattern-specific treatments, selected based on the specific signatures of endometrial carcinosarcoma. Several molecular studies described mutations or alterations of multiple genes and pathways in endometrial carcinosarcoma, including c-KIT, TKR, VEGF, EGFR, Her2/neu, NTRK, PI3K/AKT/mTOR pathway, WEE1, KRAS, EXP, BRCA1/2, and other genes related to cell-cycle regulation (including homologous recombination deficiency), histone modification, and chromatin remodeling, which may all represent potential targets.21–27
New molecular-targeted therapies may play a pivotal role in the treatment of endometrial carcinosarcoma, especially in the recurrence/metastatic setting, and many of these are currently being investigated in prospective clinical trials (Table 5), with the most promising therapeutic agents being immune checkpoint inhibitors, HER2 targeting agents, and WEE1 inhibitors.104
Immune Checkpoint Inhibitors
Immunotherapy is an emerging area of research and treatment for endometrial cancers, especially for patients with advanced/recurrent disease. There are currently two FDA-approved immune checkpoint inhibitors for the treatment of endometrial cancer, dostarlimab, and pembrolizumab. Both are indicated as single agents for the treatment of advanced/metastatic or recurrent, MSI-H/MMRd endometrial cancer that progresses on or after prior platinum-based chemotherapy in any setting and not amenable to curative surgery or radiotherapy.
Dostarlimab received FDA accelerated approval in April 2021, based on the results from the phase I GARNET trial, which showed an overall response rate of 43.5% with an acceptable safety profile. On May 23, 2017, pembrolizumab was granted accelerated approval for tumors harboring MSI-H/dMMR alterations, which included endometrial cancer.101–103
Furthermore, on March 21, 2022, the FDA approved pembrolizumab for MSI-H/dMMR endometrial cancer based on the results from the KEYNOTE-158 trial, which reported an overall response rate of 46% with a manageable safety profile.101 On July 21, 2021, the FDA granted regular approval to the combination pembrolizumab/lenvatinib in the setting of MSS/MMRp status, based on the results from the KEYNOTE-146 study, which showed an overall response rate of 38% for the endometrial cancer cohort.
In Europe, the EMA approved the adoption of dostarlimab in patients with endometrial cancer with MSI-H/dMMR (progressed after conventional chemotherapy) and pembrolizumab plus lenvatinib for any patients with endometrial cancer (progressed after conventional chemotherapy), regardless of the MMR status.101–103 The FDA and EMA are not restricting the use of immunotherapy (with or without tyrosine kinase inhibitors) in patients with endometrial carcinosarcomas. However, we have to point out that patients with endometrial carcinosarcoma were excluded from these practice-changing trials and this prevents them from translating the results to this rare histotype.101–103 However, given the similarities of endometrial carcinosarcoma with the other high-risk histologies such as uterine serous carcinoma, which instead were included, it can be speculated that immunotherapy plus tyrosine kinase inhibitor might be an attractive option also for endometrial carcinosarcoma patients. Moreover, around 10% of endometrial carcinosarcomas are MSI-H, ~5% are POLE mutated, and~60% show PD-1/PD-L1 immunohistochemical expression; immunotherapy (alone or in combination with tyrosine kinase inihibitor) may be particularly promising in these molecular subgroups.3 4 21–27 Nevertheless, to date, only anecdotal evidence exists.108 109
A small retrospective series including seven patients (14% MSI-H/MMRd) tested the efficacy of the pembrolizumab/lenvatinib doublet in endometrial carcinosarcoma and showed no partial or complete responses.109 One case report described a partial response with pembrolizumab used in a heavily pretreated patient with POLE mutated endometrial carcinosarcoma.110 Another case of a patient with platinum-refractory, PD-L1-positive, MSI-H endometrial carcinosarcoma is described.111 The patient obtained a complete response after receiving pembrolizumab plus pelvic radiotherapy.111 Notably, a subprotocol of the NCI-MATCH (EAY131) basket trial evaluated the efficacy of nivolumab in 42 refractory MSI-H cancers, including four cases of endometrial carcinosarcoma, with promising results.112 The overall response rate, median progression-free, and overall survival of the whole cohort were 36%, 6.3 months, and 17.3 months, respectively.112
More large-scale, research is needed to address the efficacy of immunotherapy in endometrial carcinosarcoma and draw definitive conclusions.104 Several immunotherapic agents are currently under investigation in dedicated clinical trials, alone or in combination (Table 5): pembrolizumab plus lenvatinib (NCT05147558), pembrolizumab plus olaparib (NCT05156268), atezolizumab plus bevacizumab and rucaparib (NCT03694262/EndoBARR), spartalizumab (NCT04802876), nivolumab (NCT03241745), nivolumab plus IDO1 inhibitor (NCT04106414), dostarlimab plus niraparib (NCT03981796/ENGOT-EN6/GOG-3031, NCT03651206/ROCSAN), durvalumab plus tremelimumab (NCT03015129).104
HER2 Targeting Agents
HER2/neu is a tyrosine kinase membrane receptor encoded by the ERBB2 gene and a member of the epidermal growth factor receptor family.111 112 Its role has been well-acknowledged in breast and gastric cancers, where HER2 overexpression has been reported in approximately 15–30% and 7–34% of cases, respectively, leading to the introduction of the monoclonal antibody trastuzumab as a standard treatment in the HER2-positive adjuvant (breast tumors) and metastatic setting (breast and gastric tumors).113 114
To date, the adoption of trastuzumab in uterine serous carcinoma is supported by the NCCN guidelines.28 Trastuzumab can be added (8 mg/kg for the first dose and 6 mg/kg in subsequent cycles) to chemotherapy in HER2-positive uterine serous carcinoma.114 Importantly, a consensus definition for HER overexpression/amplification in endometrial carcinosarcoma is lacking for endometrial cancer. Generally, HER2-positive tumors are defined as 3+ by immunohistochemistry or 2+ by immunohistochemestry with confirmatory in situ hybridization testing.113 114
Several trials are currently investigating the role of HER2 as a molecular target in other solid tumors harboring HER2 amplification, including endometrial carcinosarcoma.104 Pre-clinical research has suggested the efficacy of trastuzumab and another monoclonal antibody, pertuzumab, in endometrial carcinosarcoma.113–115 The rate of HER2 amplification in endometrial carcinosarcoma is 10–20% and reaches nearly 50% in high-grade and recurrent diseases, making it a promising therapeutic target.113 114 Significantly improved survival rates have been recently demonstrated when adding trastuzumab to carboplatin/paclitaxel in recurrent/metastatic HER2-positive uterine serous carcinoma (17.9 vs 9.3 months for advanced-stage disease and 9.2 vs 6.0 months for recurrent disease).113 114 Since serous carcinoma and endometrial carcinosarcoma share a similar molecular background, this treatment strategy has gained increasing attention also in endometrial carcinosarcoma.104
Two clinical studies are currently investigating the intriguing role of HER2 targeting agents specifically in endometrial carcinosarcoma and results are excitedly awaited (Table 5). A multicenter phase II/III trial (NCT05256225) is currently addressing the efficacy of paclitaxel/carboplatin alone or combined with either trastuzumab or pertuzumab/trastuzumab, in newly diagnosed, HER2-positive uterine serous carcinoma and endometrial carcinosarcoma.104 A possible synergistic effect for the combination of the two monoclonal antibodies has been suggested in gynecological malignancies.115 The phase II trial NCT04513665 is investigating the efficacy of another monoclonal antibody, zanidatamab (ZW25), in recurrent or persistent HER2-overexpressed endometrial cancer (including endometrial carcinosarcoma), with one to two prior lines of chemotherapy.104 Moreover, antibody drug conjugates are under evaluation even in HER2-low solid tumors (including endometrial carcinosarcoma).104 The phase II STATICE trial tested trastuzumab deruxtecan (T-DXd) in HER2-positive 1+ by immunohistochemistry endometrial carcinosarcomas. The trial enrolled 34 patients. The preliminary data of this trial is very exciting, with an estimated overall response rate of about 60%.116
WEE1 is a kinase protein that plays a key role in the correct functioning of the G2/M cell-cycle checkpoint, where the cell has the opportunity to further grow and repair the DNA damage before starting the mitotic phase.117 Its role is even more important in the presence of a TP53 mutation, which determines the loss of the G1/S cell-cycle checkpoint and so an increased cell dependency on the regulation of the G2/M checkpoint by WEE1. Since almost all endometrial carcinosarcomas harbor a p53 mutation, these tumors are characterized by cell-cycle dysregulation and high replication stress and so might be particularly vulnerable to WEE1 blockade. Adavosertib is a potent and selective oral WEE1 inhibitor, that has already shown encouraging and durable evidence of activity in women with uterine serous carcinoma, with an overall response rate of 29%, and is now being further investigated in the phase IIb ADAGIO trial.117 A phase II trial (NCT03668340) is currently investigating its efficacy also in the setting of recurrent or persistent endometrial carcinosarcoma after one or more lines of chemotherapy (Table 5).104
Targeting Epithelial–Mesenchymal Transition
The epithelial–mesenchymal transition is the process by which epithelial cells lose their polarity and intercellular junctions and become multipotent mesenchymal cells with invasion and metastatic properties and the ability to differentiate in several cell types. The epithelial–mesenchymal transition is crucial in nearly all cancers and, notably, plays a key role in the pathogenesis of sarcomatous trans-differentiation from carcinomatous elements in endometrial carcinosarcoma.118 119 Endometrial carcinosarcomas with sarcoma dominance and heterologous sarcomatous component display a higher epithelial–mesenchymal transition and have been associated with poorer outcomes than homologous ones. Therefore, the pathogenesis of epithelial–mesenchymal transition has gained increasing attention as its blockage could be a valid therapeutic strategy for patients with endometrial carcinosarcoma. In particular, by blocking epithelial–mesenchymal transition, endometrial carcinosarcomas would maintain their original carcinomatous prevalence and hence be more responsive to the standard treatments used for carcinomas. A clinical trial (the GYNecological Cancers Treated With NETrin mAbs in Combination With Chemotherapy and/or Pembrolizumab (GYNET, NCT04652076)) is ongoing to explore such mechanism of action in gynecological cancers.104 The epithelial–mesenchymal transition is regulated by complex networks involving transcriptional factors, growth factors, and cytokine signaling pathways, such as the transforming growth factor TGF-β1, Wnt, JAK/STAT, and MAPK cascade. In particular, TGF-β regulates a key pathway in the epithelial–mesenchymal transition process and may be a potential target.118 119 A phase IB trial (NCT03206177) is currently investigating the feasibility of combining galunisertib (TGF-β1 inhibitor) with the paclitaxel/carboplatin doublet in patients with newly diagnosed, persistent, or recurrent endometrial carcinosarcoma (Table 5).104
Endometrial carcinosarcoma is now regarded, and consequently staged and treated, as a primary endometrial carcinoma. Despite its rarity, the incidence of endometrial carcinosarcoma is slowly growing while the prognosis has remained extremely poor, despite current available multimodal treatment strategies. Historically, endometrial carcinosarcoma has been underinvestigated and mostly in retrospective series or in unspecific clinical trials designed for uterine sarcomas and endometrial carcinomas. Over the past years, there has been a raised awareness and understanding of endometrial carcinosarcoma pathogenesis and molecular landscape. In the era of tumor-agnostic therapies, researchers should be encouraged to design ad hoc endometrial carcinosarcoma-oriented studies to develop new practice-changing targeted therapies and provide specific guidelines for the management of endometrial carcinosarcoma. In the wake of the new insights in endometrial carcinosarcoma treatment, immunotherapy (plus tyrosine kinase inhibitor) and HER2-targeting antibodies seem to be the most promising agents for the future, and results from ongoing trials are excitedly awaited.
Data availability statement
No data are available.
Patient consent for publication
Twitter @DenysHannelore, @pedroramirezMD, @leitaomd
Contributors Substantial contributions to the conception or design of the work: All authors. Drafting the work or revising it critically for important intellectual content: All authors. Final approval of the version to be published: All authors. Agreement to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved: All authors.
Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.
Competing interests GB: Novartis AG Pharma (C/A, H), Italian Ministry of Health (RG); NC: AstraZeneca (C/A, SH), Seattle Genetics (C/A, SH), MSD (SAB), Mersana (C/A, SH), eTheRNA immunotherapies NV (C/A, SH), Roche (travel expenses), Genmab (travel expenses), Amgen (travel expenses). IR-C: honoraria from AstraZeneca, Clovis, GSK/Tesaro, and PharmaMar; consulting/advisory board fees from AstraZeneca, Roche, Clovis, GSK/Tesaro, Genmab, PharmaMar, MSD, Mersana, Deciphera, OncXea, Esai, BMS, Novartis, and Pfizer; research funding from MSD; travel expenses from AstraZeneca, GSK, and Roche. YS: AstraZeneca (CA), GSK (CA). PW: Amgen (SH, RF, SAB), AstraZeneca (SH, RF, H, SAB), Clovis (SH, RF, SAB), Eisai (SH, SAB), GSK (SH, SAB), Lilly (SH, SAB), MSD (SH, RF, SAB), Novartis (SH, RF, SAB), Pfizer (SH, RF, SAB), Roche (SH, RF, H, SAB), TEVA (SH, SAB). NYLN: AstraZeneca (SH), Janssen (SH). KT: AstraZeneca (SH), Chugai (SH, RF), Eisai (SH), MSD (SH), Mochida (SH), Takeda (SH). TE: Takeda (SH), Astra Zeneca (SH), Eisai (SH), Chugai Pharma (SH, RF), MSD (SH), Mochida (SH). DL: AstraZeneca (H, CA), Clovis (H, CA, RF), GSK/Tesaro (H, CA), Roche (CA), Genmab (CA), PharmaMar (CA, RF), MSD (CA, RF), Esai (CA), Merck Serono (CA), Novartis (CA), and PharmaMar (H); consulting/advisory board fees from AstraZeneca, Roche, Clovis, GSK/Tesaro. DMP: AstraZeneca (CA, SH, SAB), GSK (CA, SH, SAB). NRA-R: NIH/NCI Cancer Center support grant P30 CA008748 (F). HD: Roche (CA, SH, SAB), Pfizer (CA, SH, SAB), AstraZeneca (SH, SAB), Lily (SAB), GSK (SAB), Novartis (SH), Pharmamar (SH); BJM: AstraZeneca (SH, SAB), GSK (SH, SAB), Incyte (SAB), Merck (SH, SAB), Roche/Genentech (SH, SAB), Eisai (SAB), GOG-Foundation (E), US Oncology (E).
Provenance and peer review Not commissioned; externally peer reviewed.