{"paper_id":"abd584e5-a5d4-481d-8881-e6b8737af02c","body_text":"C A S E R E P O R T Open Access\nDedifferentiated endometrioid carcinoma\nof the uterus : report of four cases and\nreview of literature\nJiheun Han 1, Eun Young Ki 2, Sung Eun Rha 3, SooYoung Hur 2 and Ahwon Lee 1*\nAbstract\nBackground: Dedifferentiated endometrioid adenocarcinoma (DEAC) is rare and is known to be more aggressive\nthan high-grade endometrioid carcinoma. Differentiating between the two is important to provide appropriate\ntreatment for patients.\nCase presentation: This is a retrospective study including four cases of DEAC of the uterus, which was diagnosed\nand treated in our Obstetrics and Gynecology department between January 2013 and December 2015. Clinical,\npathological, and immunohistochemical staining features are discussed. Each tumor was composed of\nundifferentiated carcinoma (UC) and low-grade endometrioid carcinoma with abrupt transition between them. Two\npatients showed recurrence or progression within one month postoperatively and died at the last follow-up. An\nimmunohistochemical study showed PAX-8, ER, PR, and E-cadherin expression in UC component.\nConclusions: DEAC should not be underdiagnosed as conventional endometrioid adenocarcinoma due to its\nfulminant clinical course. Therefore, UC, including DEAC, should be further categorized to provide intensive\ntreatment to improve patient survival.\nKeywords: Dedifferentiated endometrioid carcinoma, Undifferentiated carcinoma, Endometrioid carcinoma\nBackground\nEndometrial carcinoma is the most common gynecologic\nmalignancy in developed countries [1]. It comprises several\npathological subtypes, such as endometrioid, mucinous,\nclear cell, mixed cell, undifferentiated, and dedifferentiated\ncarcinoma[2]. The last two entities are recently defined to\ndistinguish them from other less aggressive tumors, there-\nfore, providing proper treatment for patients.\nUndifferentiated carcinoma (UC) of the endometrium is\ndefined as a malignant neoplasm with no differentiation.\nIt displays solid patternless growth and has worse clinical\noutcome than high-grade endometrioid adenocarcinoma.\nIt is often observed admixed with differentiated endome-\ntrioid carcinoma (grade 1/2), which is referred to as dedif-\nferentiated endometrioid adenocarcinoma (DEAC). The\nbiologic features of DEAC are known to be determined by\nUC component even when this component represents\n20% of the entire neoplasms, thus, with aggressive\noutcome [3].\nIn the current International Federation of Obstetrics\nand Gynecology (FIGO) grading system of endometrioid\nadenocarcinoma, tumors are graded by the proportion\nof solid components within a tumor, without further\ndetails on the histologic features of solid areas, resulting\nin misdiagnosis of DEAC as FIGO grade 2 or 3 endome-\ntrioid carcinoma. However, differentiating between the\ntwo is important in providing appropriate treatment\noptions for patients. The present study reports four\ncases diagnosed with DEAC and reviews the literatures\nupdated on the clinical, radiological, and pathological\nDEAC characteristics of the uterus.\nCase presentation\nThis is a retrospective study including four cases of\nDEAC of the uterus, which was diagnosed and treated in\nour Obstetrics and Gynecology department between\nJanuary 2013 and December 2015. During this period,\n* Correspondence: klee@catholic.ac.kr\n1Department of Hospital Pathology, Seoul St. Mary ’s Hospital, College of\nMedicine, The Catholic University of Korea, 222, Banpo-daero, Seocho-gu,\nSeoul 06591, Republic of Korea\nFull list of author information is available at the end of the article\n© The Author(s). 2017 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0\nInternational License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and\nreproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to\nthe Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver\n(http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.\nHan et al. World Journal of Surgical Oncology  (2017) 15:17 \nDOI 10.1186/s12957-016-1093-0\n\n155 cases of endometrial carcinoma were initially diag-\nnosed and of which four patients (2.6%) were diagnosed\nwith DEAC in our hospital. Clinical, pathological, and\nimmunohistochemical staining features are outlined\nbelow and in Table 1, Table 2, and Table 3.\nCase 1\nA 77-year-old woman (body mass index [BMI],\n19.6 kg/m 2) presented with 1 week postmenopausal\nbleeding, which was preceded by vaginal spotting for\n1 year and concomitant genital itching that had\ncontinued for 2 weeks. Transvaginal ultrasound re-\nvealed abnormally thickened endometrium (1.47 cm)\nand biopsy was recommended, but she refused. After\n2 weeks, follow-up ultrasonography showed remark-\nably increased endometrium thickness (2.36 cm) with\na 7 cm mixed-echoic lesion within the endometrial\ncavity. Endometrial curettage revealed poorly differen-\ntiated carcinoma. Magnetic resonance imaging (MRI)\nshowed 8.6 × 3.7 cm heterogeneously enhanced mass\non contrast-enhanced T1-weighted images (CET1WI)\n(Fig. 1a). In addition, tumor involvement of the cer-\nvical stroma was observed on CET1WI. The patient\nsubsequently underwent total hysterectomy (TH) with\nbilateral salpingo-oophorectomy (BSO) and pelvic\nlymphadenectomy. The surgical specimen of the\nuterus showed white gray polypoid mass filling the\nendometrial cavity (Fig. 1b). Upon microscopic exam-\nination, the tumor with high cellularity and pattern-\nless growth without glandular differentiation invaded\nthe full thickness of the lower uterine segment\n(Fig. 1c) extending to the cervical stroma. The tumor\nconsisted of medium –sized monotonous cells with\nbrisk mitosis (5 per high-power field (HPF)) and atyp-\nical mitotic figures (Fig. 1d). Moderately differentiated\nendometrioid adenocarcinoma component with squa-\nmous differentiation comprised the major proportion\nof the tumor mass of the uterine body and fundus,\nwhich invaded more than one half of the myometrial\nthickness. No lymph node (LN) metastasis was found.\nCytokeratin 8/18 (CK8/18) and pancytokeratin (CK\nAE1/AE3) expressed both components but in different\npatterns. In the UC component, they were expressed as\ndot-like patterns, whereas they showed cytoplasmic ex-\npression in differentiated component (Figs. 1e, f ). Epithe-\nlial membrane antigen (EMA), estrogen receptor (ER),\nprogesterone receptor (PR), E-cadherin, and PAX-8 were\nexpressed in the differentiated component alone (T able 2).\nBased on the FIGO system, the patient had stage II. She\nwas recommended adjuvant external beam pelvic RT\n(EBRT), but she refused. A month after the surgery, she\npresented with difficulty in urination. Abdominal pelvic\ncomputed tomography (CT) revealed multiple, large,\nperitoneal seeding masses, and multiple lymph node\nmetastases in the external and internal iliac chains with\nlarge amount of ascites which was positive for malignant\ncells on cytologic examination. She died 7 weeks later after\nsurgery due to tumor lysis syndrome.\nCase 2\nA 54-year-old postmenopausal woman (BMI: 22.5 kg/m 2)\nwas referred to our hospital after being diagnosed with\nendometrioid carcinoma FIGO grade 3 on biopsy at a\nlocal hospital. She experienced vaginal bleeding for\n3 months. She had menopause at 51 year old hypertension\nhistory. MRI showed a 5.0 × 3.6 cm lobulated mass filling\nthe endometrial cavity, which showed slightly high signal\nintensity (SI) on T2-weighted imaging (T2WI) and\nshowed poor enhancement compared with the adjacent\nmyometrium (Fig. 2a). Mildly enlarged pelvic LNs were\nnoted in both external iliac areas. TH with BSO, pelvic\nlymphadenectomy, and para-aortic lymphadenectomy\nwere performed. The mass was bulky and filled the entire\nendometrial cavity on gross examination (Fig. 2b). Micro-\nscopically, well-differentiated endometrioid adenocarcin-\noma and UC with abrupt transition were found between\nthem (Fig. 2c). The UC component showed a solid mono-\nmorphic discohesive cell growth, showing no differenti-\nation except for the presence of some rhabdoid cells\n(Fig. 2d). The invasion depth was less than one half of the\nmyometrium, and LN metastasis was not observed. CK8/\n18, CK AE1/AE3, and EMA showed diffuse cytoplasmic\nexpression in both components, and perinuclear cytokera-\ntin dots were also observed in the UC component. The\nER, PR, E-cadherin, and PAX-8 expressions were recog-\nnized in the differentiated component alone (T able 2). The\nfinal FIGO stage was IA, and adjuvant treatment was not\nperformed. The patient has been disease-free for\n19 months after the initial diagnosis.\nCase 3\nA 60-year-old postmenopausal woman (BMI: 25.3 kg/m 2)\nwith a chief complaint of vaginal spotting for 1 year visited\na local gynecologic clinic. She had menopause at 52 years\nold. Colposcopy showed necrotic tissue with bleeding at\nthe right vaginal wall, and she was referred to our hospital\nfor further evaluation. Ultrasonography showed an echo-\ngenic mass in the endometrial cavity. The MRI revealed a\n7.2 × 3.5 cm lobulated mass extending to the right above\nthe endocervix (Fig 3a). The mass showed high SI on\nT2WI and low SI on T1WI with heterogeneous enhance-\nment. Another 5 cm, elongated mass involving the lower\nvagina was also found. Vaginal biopsy revealed poorly dif-\nferentiated carcinoma, and cytologic test of the endocervix\nrevealed adenocarcinoma. The patient underwent wide\ncuff TH with BSO, and the separated additional vaginal\nmass was not removed. The surgical specimen showed a\npolypoid mass with necrosis and hemorrhage, which\nHan et al. World Journal of Surgical Oncology  (2017) 15:17 Page 2 of 9\n\ninvaded more than one half of the myometrium of the\nuterus (Fig. 3b), extending to the cervix. Most of the\ntumor comprised discohesive cell growth with a solid\nsheet pattern on microscopic examination, suggesting a\nUC component. They were occasionally segregated by\ndelicate fibrovascular septa forming a vague alveolar pat-\ntern (Fig. 3c). The UC component showed small amount\nof myxochondroid stroma, reminiscent of cartilage\n(Fig. 3d). A small proportion (10%) of tumor consisted of\nlow-grade endometrioid adenocarcinoma juxtaposed with\nUC component. The histological finding of the UC com-\nponent was similar to that of the lesion in the vaginal wall,\nsuggesting vaginal metastasis (drop metastasis). ER and\nPR were not expressed in both UC and differentiated\nTable 1 Clinical and pathologic features of four patients with dedifferentiated endometrioid adenocarcinoma (DEAC) of uterus\nCase 1 Case 2 Case 3 Case 4\nClinical features\nAge at diagnosis, years 77 54 60 52\nAge at menopause, years 60 51 52 Not applicable\nPresentation Postmenopausal\nbleeding\nPostmenopausal\nbleeding\nPostmenopausal\nspotting\nPerimenopausal\nbleeding\nInitial diagnosis Poorly differentiated\ncarcinoma\nEndometrioid\nadenocarcinoma,\nFIGO grade 3\nPap smear:\nadenocarcinoma\nvaginal biopsy:\npoorly differentiated\ncarcinoma\nLeiomyoma\nImaging findings MRI: 8.6 × 3.7 cm sized\nheterogeneously\nenhanced mass on\nCET1WI with\ninvolvement of the\ncervical stroma\nMRI: 5.0 × 3.6 cm sized lobulated\nmass filling endometrial cavity\nwith slightly high SI in T2WI\nand poor enhancement than\nadjacent myometrium\nMRI: 7.2 × 3.5 cm sized\nlobulated mass which\nshowed high SI on T2WI\nand low SI on T1WI with\nheterogeneous\nenhancement\nCT: 10 cm sized low\ndensity mass in the\nendometrial cavity\nand cervical canal\nSurgical management TH/BSO/PLND TH/BSO/PLND/PALND Wide cuff TH/BSO TH/BSO/PLND/PALND\nOperative findings Cancer extension\nto cervix\nInvasion to superficial\nmyometrium\nCancer extension to cervix and\nvaginal wall extension\nNecrotic mass filling\nendometrial cavity\nwith protruding\nthrough cervical canal\nPostoperative\nmanagement\nRefused Not needed Chemotherapy (CDDP + ADR + CTX)\nand EBRT(50 Gy) + ICR(20 Gy/4fx)\nEBRT(50 Gy) +\nICR(20 Gy/4fx)\nFirst postoperative\nrecurrence or progression\n1 month None 1 month None\nFIGO surgical stage II IA IIIB II\nClinical history Hypertension Hypertension Hypertension None\nFamily history None None Gastric cancer (mother) None\nStatus at last follow-up DOD (7 weeks) NED (19 months) DOD (10 months) NED (39 months)\nPathologic features\nTumor location Fundus, body,\nlower uterine\nsegment, cervix\nBody Body, lower uterine\nsegments\nBody, lower uterine\nsegment\nTumor grade G2 (80%) + UC (20%) G1 (70%) + UC (30%) G1 (10%) + UC (90%) G2 (40%) + UC (60%)\nMyometrial invasion Full thickness\nof myometrium\n<1/2 of myometrium <1/2 of myometrium >1/2 of myometrium\nLymphovascular space\ninvasion (LSI)\nPresent Present Present Present\nCervical stromal invasion Present Absent Present Present\nOvaries and fallopian\ntubes\nUnremarkable Unremarkable Unremarkable Unremarkable\nCET1WI contrast-enhanced T1-weighted image; SI signal intensity; T2WI T2-weighted image; T1WI T1-weighted image; TH total hysterectomy; BSO bilateral\nsalpingo-oophorectomy; PLND pelvic lymph node dissection; PALND para-aortic lymph node dissection; EBRT external beam radiation therapy; ICR intracavitary\nradiation; DOD die of disease; NED no evidence of disease; UC undifferentiated carcinoma; G1 FIGO grade 1 endometrioid adenocarcinoma; G2 FIGO grade 2\nendometrioid adenocarcinoma; CDDP cisplatin; ADR adriamycin; PTX paclitaxel; CTX cyclophosphamide\nHan et al. World Journal of Surgical Oncology  (2017) 15:17 Page 3 of 9\n\ncomponent on immunohistochemical staining (T able 2).\nOne month after the surgery, postoperative positron emis-\nsion tomography-CT revealed multiple metastatic lymph-\nadenopathies in para-aortic, retrocaval, paracaval, both\ncommon and external iliac chains, and superficial inguinal\nareas with bone metastasis. The patients was treated with\nsequential chemoradiation therapy, which comprised of\nchemotherapy (cisplatin + Adriamycin + cyclophospha-\nmide) followed by EBRT, with a total dose of 50 Gy was\nadministered for 5 weeks to the whole pelvis, vagina, and\nboth inguinal lymph nodes, and high-dose intracavitary\nradiotherapy (ICR) (20 Gy in fractions). She presented\nwith dyspnea 2 months after radiation therapy, and\nclinical examination revealed lung metastasis. She died\n3 weeks later (10 months after surgery).\nCase 4\nA 52-year-old nulligravid, perimenopausal woman (BMI\n22.4 kg/m 2) was admitted to our hospital for treatment\nof leiomyoma, manifested as 10 cm low-density mass in\nthe endometrial cavity and cervical canal on pelvic CT at\na local clinic (Fig. 4a). She experienced intermittent vagi-\nnal bleeding 3 years prior. During surgery, the surgeon\nobserved a 10 cm necrotic mass protruding through the\ncervical canal, which seemed to fill the entire endomet-\nrial cavity. Frozen section revealed malignancy. Surgical\ns t a g ew a sF I G Os t a g eI I ,d u et oc e r v i c a ls t r o m a li n v a s i o n .\nAny enlarged lymph node was not recognized. Hence, the\npatient was treated with bilateral salpingectomy with pelvic\nand para-aortic lymph node dissection. The fungating mass\nfilling the entire endometrial cavity was recognized on gross\nexamination (Fig. 4b). The tumor comprised of conven-\ntional endometrioid adenocarcinoma (FIGO grade 2) and\nUC with sharp border between them (Fig. 4c). CK8/18, CK\nAE1/AE3, and EMA were expressed focally as dot-like and\nc y t o p l a s m i cp a t t e r ni nU Cc o m p o n e n t( F i g .4 d ) .P A X - 8w a s\ndiffusely expressed in both co mponents (T able 2). The pa-\ntient was treated with adjuvant EBRT in the whole pelvis\nwith a total dose of 50 Gy in 28 fractions followed by vagi-\nnal cylinder ICR (20 Gy/4 fractions). No evidence of recur-\nrence or progression of disease was found after 39 months\nafter surgery.\nDiscussion\nBecause DEAC has been recently recognized, the inci-\ndence rate is not established. The incidence of undiffer-\nentiated carcinoma is known to be 1 to 9% [4, 5]. In\nseveral retrospective studies, 37 to 87% of UC were\nadmixed with endometrioid adenocarcinoma [3, 5, 6]. In\nour institution, 2.6% of patients (four out of 155 pa-\ntients) with endometrial carcinoma were diagnosed with\nDEAC between 2013 and 2015.\nA dualistic model, proposed by Bokhman [7] based on\nclinical and epidemiological observations, has been\nknown to be correlated with endometrial carcinoma\nclassification based on histopathological subtypes. Pre-\ndominant form of type 1 endometrial carcinoma is low-\ngrade endometrioid adenocarcinoma, whereas type II\nencompasses high-grade endometrioid adenocarcinoma,\nand most non-endometrioid histologic subtypes. How-\never, increasing evidence supports that the binary classi-\nfication is imperfect [8], and DEAC may be one of the\nrepresentative entities. Type I tumors are known to be\nassociated with unopposed estrogen stimulation,\nwhereas type II tumors are commonly described as es-\ntrogen independent. The established risk factors for type\nTable 2 Results of immunohistochemical stains of the four cases\nIHC stain Case 1 Case 2 Case 3 Case 4\nUC DC UC DC UC DC UC DC\nCK 8/18 Diffuse,\nperinuclear\ndot-like\nDiffuse,\ncytoplasmic\nDiffuse,\nperinuclear\ndot-like,\nand cytoplasmic\nDiffuse,\ncytoplasmic\nNegative Diffuse,\ncytoplasmic\nFocal, perinuclear\ndot-like,\nand cytoplasmic\nDiffuse,\ncytoplasmic\nPancytokeratin Diffuse,\nperinuclear\ndot-like\nDiffuse,\ncytoplasmic\nDiffuse,\nperinuclear\ndot-like\n< cytoplasmic\nDiffuse,\ncytoplasmic\nFocal, perinuclear\ndot-like <\ncytoplasmic\nDiffuse,\ncytoplasmic\nFocal, perinuclear\ndot-like < cytoplasmic\nDiffuse,\ncytoplasmic\nEMA Negative Diffuse,\ncytoplasmic\nDiffuse,\nperinuclear\ndot-like, and\ncytoplasmic\nDiffuse,\ncytoplasmic\nFocal, perinuclear\ndot-like, and\ncytoplasmic\nDiffuse,\ncytoplasmic\nFocal, perinuclear\ndot-like, and\ncytoplasmic\nDiffuse,\ncytoplasmic\nVimentin Focal Negative Diffuse Diffuse Diffuse Diffuse Diffuse Focal\nER Negative Positive Negative Positive Negative Negative Negative Positive\nPR Negative Positive Negative Positive Negative Negative Negative Positive\nE-cadherin Negative Diffuse Negative Focal Negative Diffuse, weakly Negative Diffuse\nPAX-8 Negative Diffuse Negative Diffuse Negative Diffuse Diffuse Diffuse\nProportion of tumor cells are >50%, diffuse; 10 ~ 50%, focal; <10%, negative; EMA epithelial membrane antigen; ER estrogen receptor; PR progesterone receptor\nHan et al. World Journal of Surgical Oncology  (2017) 15:17 Page 4 of 9\n\nTable 3 Cases of dedifferentiated endometrioid adenocarcinoma (DEAC) with clinical and pathologic features\nAuthor Number of\ncases\nAge, years Surgical operation\n(cases)\nStage (cases) Component\nof tumor\nMarker expression of\nUC component\nAdjuvant treatment\n(cases)\nSurvival outcome (cases)\nSilva, EG (2006) [ 3] 25 51 (median)\n(range: 30–82)\nTH+ BSO (24) I (14)\nII (1)\nIII (6)\nIV (4)\nLow grade (10 –80%) +\nUC(20-90%)\nKeratin (13 out of 15,\nfocal or diffuse) EMA (all)\nNeuroendocrine marker\n(4 out of 15)\nChemotherapy (18)\nradiation (4)\nDOD (15) (median:\n7 months)\nAWPD (6)\n(6 ~ 8 months)\nNA(3)\nShen, Y (2012) [ 18] 1 51 TH + BSO + PLND II Low grade (80%)\n+ UC(20%)\nNegative for EMA focally\npositive for CK7, CK18\nVaginal radiation +\nchemotherapy (CDDP+\nDTX + Taxanes)\nNA\nVita, G (2011) [ 19] 1 45 TH + BSO IIIA Low grade (60%)\n+ UC(40%)\nPositive for cytokeratins\nand EMA\nChemotherapy (CDDP\n+ ATC + Taxanes)\nNA\nWu, ES (2013) [ 20] 1 62 NA NA NA NA Radiation + hormone\ntherapy (Megace\nalternating with\nTamoxifen)\nAWPD (3 months)\nBerretta, R (2013) [ 21] 1 67 NA IV NA Positive for keratin and\nnegative for neuronal\nmarkers\nChemotherapy\n(CBDCA + Taxol)\nNA\nPark, SY (2014) [ 22] 1 55 TH + USO + PLND IB Low grade (40%)\n+ UC(60%)\nFocally positive for CK\nand EMA\nChemotherapy\n(PTX + CDDP+DOXO)\nDOD (7 months)\nLi, Z (2016) [ 15] 13 61 (median) NA III/IV (12) NA Pancytokeratin (10 out\nof 13)\nCam5.2 (8 out of 13)\nEMA (8 out of 13, weak\nand patchy)\nPAX-8 (1 out of 13)\nChemotherapy\nand radiation (13)\nRecurrence or\nmetastasis within\n3 years of diagnosis (12)\nDisease-free after 3 years\nof diagnosis (1)\nSoyama, H (2016) [ 23] 1 41 Supravaginal\nhysterectomy +\nUSO+partial\nresection of ileum\nIVB NA NA Chemotherapy alone DOD (7 months)\nRabban (2016) [ 24] 1 50 TH + BSO+PLND IA G1(60%) + UC(40%) Negative for EMA,\nkeratin, PAX-8\nUntreated Progression after\n10 months of surgery\nTH total hysterectomy; BSO bilateral salpingo-oophorectomy; PLND pelvic lymph node dissection; DOD die of disease; AWPD alive with progressive disease; USO unilateral salpingo-oophorectomy; NA not available;\nCDDP cisplatin; DTX docetaxel; ATC anthracycline; CBDCA Carboplatin; PTX paclitaxel; DOXO doxorubicin\nHan et al. World Journal of Surgical Oncology  (2017) 15:17 Page 5 of 9\n\nI endometrial carcinomas are obesity, unopposed estro-\ngen therapy use, nulliparity, early menarche, late meno-\npause, oral contraceptive use, and smoking, whereas\nlittle is known regarding the risk factors for type II tu-\nmors. A recent study [9] showed that most classical\nendometrial cancer risk factors were also associated with\ntype II tumors, suggesting that the etiology of type II tu-\nmors may not be completely estrogen independent. In\nthe same context, three of four patients of our cases\nwere postmenopausal, older ( ≥54 year old) women, and\nthe other perimenopausal woman was nulliparous. Three\npatients of our cases had hypertension as underlying dis-\nease. Giordano et al. [10] showed that the majority of\npatients with malignant endometrial polyps had risk fac-\ntors, such as hypertension, obesity, and unopposed es-\ntrogen therapy, for the development of endometrial\ncarcinoma.\nFig. 1 Case 1. (A1) Sagittal contrast-enhanced T1-weighted MRI shows\nheterogeneously enhanced mass filling the endometrial cavity with\ncervical stromal invasion. (A2) The tumor shows white infiltrative lesion\nwith necrosis involving lower uterine segment and uterine body,\nextending to cervix. The UC component (A3, ×2.5) infiltrates the full\nthickness of the lower uterine segment and arranged in patternless\nwith extensive necrosis and hemorrhage. The tumor (A4,× 4 0 0 )\ncomprises monotonous cells with moderate pleomorphism, prominent\nnucleoli, and high mitotic rates. Cytokeratin 8/18 (CK8/18) expressed as\nperinuclear dots in the UC component (A5, ×400) and showed diffuse\ncytoplasmic expression in differentiated component (A6,× 2 0 0 )\nFig. 2 Case 2. (B1) Sagittal T2-weighted MRI shows a large polypoid\nendometrial mass with superficial infiltration of myometrium. (B2)G r o s s\nphotograph shows a polypoid mass compacting the endometrial cavity.\nWell-circumscribed round mass, submural leiomyoma is seen (arrow).\n(B3, ×100) The tumor comprises moderately differentiated endometrioid\nadenocarcinoma and UC with abrupt transition (arrows) between them.\n(B4, ×400) The UC cell component shows discohesive rhabdoid feature\nFig. 3 Case 3. ( C1) Sagittal contrast-enhanced T1-weighted MRI\nshows a large polypoid mass lesion filling the endometrial cavity.\n(C2) The tan polypoid mass involving more than one half of the\nmyometrium. (C3, ×200) A delicate fibrovascular septa separating\ndiscohesive cells into vague alveolar nests are found. ( C4, ×200)\nFocal areas of UC component shows myxochondroid stroma with\nembedded tumor cells, resembling the cartilage\nHan et al. World Journal of Surgical Oncology  (2017) 15:17 Page 6 of 9\n\nIn all our cases, DEAC could not be diagnosed on bi-\nopsy or curettage and imaging studies, but could only be\ncorrectly diagnosed by pathological examination of the\nsurgical specimen. Two cases were initially diagnosed as\npoorly differentiated carcinoma, and one case was diag-\nnosed as FIGO grade 3 endometrioid adenocarcinoma\nby endometrial biopsy or curettage. A recent study [11]\nshowed that preoperative endometrial biopsy or curet-\ntage is sensitive overall for detecting endometrial cancer,\nbut sensitivity decreases with high-risk histology endo-\nmetrial cancer. In addition, interobserver reproducibility\nin diagnosing high-grade endometrial carcinomas has\nbeen known to be worse than that of low-grade tumors\n[12]. Therefore, if preoperative examination suggests\nhigh-grade endometrial carcinomas, the surgeon should\nproceed to a more thorough surgical staging.\nTo correctly diagnose DEAC, recognizing the UC\ncomponent is important. Recent studies [5, 13] empha-\nsized that there is reproducible, distinctive histological\nfeatures of UC, which are confirmed in all present cases.\nAlthough histological findings of UC can overlap with\nthat of high-grade endometrioid adenocarcinoma, differ-\nences were found between them. In UC, tumor cells lack\nintercellular cohesion and arranged in patternless solid\nsheets without gland formation, which were occasionally\nadmixed with rhabdoid cells. In contrast, high-grade\nendometrioid adenocarcinoma has at least a foci of\ngland formation of cohesive cells with rare or no rhab-\ndoid cells. In addition, the tumor cells of UC in our\ncases also tended to have larger nuclei with more prom-\ninent nucleoli than conventional endometrioid adenocar-\ncinoma. When UC are juxtaposed with low-grade\nendometrioid adenocarcinoma, such as in DEAC, a\nsharp boundary is noted between them, whereas a seam-\nless transition from glandular component to solid area is\nobserved in high-grade endometrioid adenocarcinoma.\nImmunochemical studies are also helpful in making a\ndifferential diagnosis. Among the epithelial markers,\nEMA and CK18 are known to be strongly and diffusely\nstained in the solid area of high-grade endometrioid car-\ncinoma, but are focal or weak in the UC component [6].\nIn the previous case reports, expression of EMA in the\nUC components showed relatively inconsistency\n(Table 3), and our cases also showed variable expression\nof EMA and CK18 (Table 2). However, we found that\nexpression of EMA and CK18 were perinuclear dot-like\npattern in UC component, whereas diffuse cytoplasmic\npattern in differentiated component in all cases. Rama-\nlingam et al. [14] suggested PAX-8 to be the most effect-\nive immunomarker to distinguish UC from the solid\ncomponent of either endometrioid carcinoma or serous\ncarcinoma. In the present study, three of 4 cases\nexpressed PAX-8 in the differentiated component alone,\nwhereas the other one also expressed PAX-8 in the UC\ncomponent. ER, PR, and E-cadherin are known to be\nretained in conventional endometrioid adenocarcinoma\n[6, 15, 16], but not in the UC, which were consistent with\nour cases. Therefore, perinuclear dot-like staining of cyto-\nkeratin and EMA, loss of expression of PAX-8, ER, PR,\nand E-cadherin would be helpful to distinguish UC from\nconventional endometrioid adenocarcinoma.\nDifferentiating between them is important because\nDEACs have fulminant clinical outcomes and poorer\nprognosis than high-grade endometrioid carcinoma. Sev-\neral reports of DEAC cases with clinical feature, are\nsummarized in Table 3.\nIn our cases, the proportion of UC component ranged\nfrom 20 to 90% (Table 1), which did not seem to be as-\nsociated with clinical outcomes. In addition, the involve-\nment to the lower uterine segment and cervix occurred\nin the UC component in all cases.\nThe morphologic appearance of DEACs suggest very\nbroad differential diagnoses, including not only high-\ngrade endometrioid adenocarcinoma, but also unclassi-\nfied sarcomas, malignant mixed Müllerian tumors\n(MMMT), and rhabdoid tumor.\nThe UC component in DEAC may be negative or focal\npositive for keratin in immunohistochemical staining,\nand it can be misdiagnosed as sarcoma. Most sarcomas\nin the uterus comprise spindle cells with muscular dif-\nferentiation and seldom comprise epithelioid cells only.\nImmunostaining with markers for muscular differenti-\nation, such as desmin, caldesmon, and SMA may help to\nFig. 4 Case 4. ( D1) Sagittal contrast-enhanced CT shows a bulky\nhypodense mass filling the endometrial cavity. (D2) The large outbulging\nmass with friable surface. (D3, ×2.5) A sharp border between differentiated\nand UC component and EMA ( D4, ×2.5 and ×400) shows strong\ncytoplasmic expression in the former and focal dot and cytoplasmic\nexpression in the latter\nHan et al. World Journal of Surgical Oncology  (2017) 15:17 Page 7 of 9\n\ndistinguish between them because DEACs are negative\nfor these markers.\nMMMT is a biphasic tumor that usually contains car-\ncinoma and sarcomatous components. However, both\ncomponents are high-grade in MMMT, whereas the\ngland-forming component is low-grade (grade 1/2), and\nno mesenchymal component is found in DEACs. Case 3\nhad focal areas with myxoid stroma mimicking heterol-\nogous component of MMMT. MMMTs typically occur\nin older women, whereas DEACs may occur in young\npatients. Moreover, these two entities may have distinct\nbiology; MMMT is usually not associated with microsat-\nellite instability, however, a proportion of DEACs appear\nto be associated with microsatellite instability [17].\nOur cases had various proportions of rhabdoid cells,\nand if these are prominent, DEACs need to be differenti-\nated from extrarenal malignant rhabdoid tumor. Unlike\nDEACs, rhabdoid tumors are not associated with a well-\ndifferentiated endometrioid carcinoma.\nConclusion\nIn summary, DEAC is a rare and recently defined entity,\nwhich has distinct histological and immunohistochemi-\ncal features and should not be underdiagnosed as con-\nventional endometrioid adenocarcinoma. Furthermore,\namong the endometrioid type 2 tumors, UC including\nDEAC should be further categorized because of their\nworse clinical and biological behavior.\nAbbreviations\nBMI: Body mass index; BSO: Bilateral salpingo-oophorectomy;\nCET1WI: Contrast-enhanced T1-weighted images; CK AE1/\nAE3: Pancytokeratin; CK8/18: Cytokeratin 8/18; DEAC: Dedifferentiated\nendometrioid adenocarcinoma; EBRT: External-beam pelvic RT;\nEMA: Epithelial membrane antigen; ER: Estrogen receptor; FIGO: Federation\nof Obstetrics and Gynecology; HPF: High-power field; ICR: Intracavitary\nradiotherapy; LN: Lymph node; PR: Progesterone receptor; TH: Total\nhysterectomy; UC: Undifferentiated carcinoma\nAcknowledgements\nThe authors declare that there are no acknowledgements.\nAvailability of data and materials\nAll the data on which the conclusions of this case report are based are\nincluded in this manuscript.\nAuthor’s contributions\nJH collected the data and authored the manuscript. EY and SE contributed to\nthe data analysis and shaping of the manuscript. SY performed the surgical\nprocedure and also contributed the data analysis. AW conceived the given\ninstructions for acquisition of data, outlined, and contributed to the shaping of\nthe manuscript. All authors have read and approved the final manuscript.\nCompeting interests\nThe authors declare that they have no competing interests.\nConsent for publication\nWritten informed consent was obtained from the patient for the publication\nof this case report and any accompanying images.\nEthics approval and consent to participate\nEthical approval is obtained from the Institutional Review Boards of St. Mary ’s\nHospital (Application number: KC16ZISE0659). Consents of the patients were\nobtained to participate to the presentation of this report.\nAuthor details\n1Department of Hospital Pathology, Seoul St. Mary ’s Hospital, College of\nMedicine, The Catholic University of Korea, 222, Banpo-daero, Seocho-gu,\nSeoul 06591, Republic of Korea. 2Department of Obstetrics and Gynecology,\nSeoul St. Mary ’s Hospital, College of Medicine, The Catholic University of\nKorea, Seoul, Republic of Korea. 3Department of Radiology, Seoul St. Mary ’s\nHospital, College of Medicine, The Catholic University of Korea, Seoul,\nRepublic of Korea.\nReceived: 23 September 2016 Accepted: 23 December 2016\nReferences\n1. Siegel RL, Miller KD, Jemal A. Cancer statistics, 2016. CA Cancer J Clin. 2016;66:7–30.\n2. Kurman RJ, Carcangiu ML, Herrington CS, Young RH. WHO classification of\ntumours of female reproductive organs. 4th ed. Lyon: International Agency\nfor Research on Cancer; 2014.\n3. Silva EG, Deavers MT, Bodurka DC, Malpica A. Association of low-grade\nendometrioid carcinoma of the uterus and ovary with undifferentiated carcinoma:\na new type of dedifferentiated carcinoma? Int J Gynecol Pathol. 2006;25:52–8.\n4. Ronnett BM, Zaino RJ, Ellenson LH, Kurman RJ. Endometrial carcinoma. In:\nKurman RJ, editor. Blaustein ’s pathology of the female genital tract. 5th ed.\nNew York: Springer-Verlag; 2002. p. 501 –59.\n5. Altrabulsi B, Malpica A, Deavers MT, Bodurka DC, Broaddus R, Silva EG.\nUndifferentiated carcinoma of the endometrium. Am J Surg Pathol. 2005;29:\n1316–21.\n6. Tafe LJ, Garg K, Chew I, Tornos C, Soslow RA. Endometrial and ovarian\ncarcinomas with undifferentiated components: clinically aggressive and\nfrequently underrecognized neoplasms. Mod Pathol. 2010;23:781 –9.\n7. Bokhman JV. Two pathogenetic types of endometrial carcinoma. Gynecol\nOncol. 1983;15:10–7.\n8. Murali R, Soslow RA, Weigelt B. Classification of endometrial carcinoma:\nmore than two types. Lancet Oncol. 2014;15:e268 –78.\n9. Setiawan VW, Yang HP, Pike MC, McCann SE, Yu H, Xiang YB, et al. Type I\nand II endometrial cancers: have they different risk factors? J Clin Oncol.\n2013;31:2607–18.\n10. Giordano G, Gnetti L, Merisio C, Melpignano M. Postmenopausal status,\nhypertension and obesity as risk factors for malignant transformation in\nendometrial polyps. Maturitas. 2007;56:190 –7.\n11. Kanis MJ, Rahaman J, Moshier EL, Zakashansky K, Chuang L, Kolev V. Detection\nand correlation of pre-operative, frozen section, and final pathology in high-\nrisk endometrial cancer. Eur J Gynaecol Oncol. 2016;37:338–41.\n12. Thomas S, Hussein Y, Bandyopadhyay S, Cote M, Hassan O, Abdulfatah E, et\nal. Interobserver variability in the diagnosis of uterine high-grade\nendometrioid carcinoma. Arch Pathol Lab Med. 2016;140:836 –43.\n13. Silva EG, Deavers MT, Malpica A. Undifferentiated carcinoma of the\nendometrium: a review. Pathology. 2007;39:134 –8.\n14. Ramalingam P, Masand RP, Euscher ED, Malpica A. Undifferentiated\ncarcinoma of the endometrium: an expanded immunohistochemical\nanalysis including PAX-8 and basal-like carcinoma surrogate markers. Int J\nGynecol Pathol. 2016;35:410 –8.\n15. Li Z, Zhao C. Clinicopathologic and immunohistochemical characterization\nof dedifferentiated endometrioid adenocarcinoma. Appl Immunohistochem\nMol Morphol. 2016;24:562 –8.\n16. Romero-Pérez L, López-García MÁ, Díaz-Martín J, Biscuola M, Castilla MÁ,\nTafe LJ, et al. ZEB1 overexpression associated with E-cadherin and\nmicroRNA-200 downregulation is characteristic of undifferentiated\nendometrial carcinoma. Mod Pathol. 2013;26:1514 –24.\n17. Garg K, Leitao Jr MM, Kauff ND, Hansen J, Kosarin K, Shia J, et al. Selection of\nendometrial carcinomas for DNA mismatch repair protein\nimmunohistochemistry using patient age and tumor morphology enhances\ndetection of mismatch repair abnormalities. Am J Surg Pathol. 2009;33:925–33.\n18. Shen Y, Wang Y, Shi Y, Liu J, Liu Y. Clinicopathologic study of endometrial\ndedifferentiated endometrioid adenocarcinoma: a case report. Int J Clin Exp\nPathol. 2012;5:77–82.\nHan et al. World Journal of Surgical Oncology  (2017) 15:17 Page 8 of 9\n\n19. Vita G, Borgia L, Di Giovannantonio L, Bisceglia M. Dedifferentiated\nendometrioid adenocarcinoma of the uterus: a clinicopathologic study of a\ncase. Int J Surg Pathol. 2011;19:649 –52.\n20. Wu ES, Shih Ie M, Diaz-Montes TP. Dedifferentiated endometrioid\nadenocarcinoma: An under-recognized but aggressive tumor? Gynecol\nOncol Case Rep. 2013;5:25 –7.\n21. Berretta R, Patrelli TS, Faioli R, Mautone D, Gizzo S, Mezzogiorno A, et al.\nDedifferentiated endometrial cancer: an atypical case diagnosed from\ncerebellar and adrenal metastasis: case presentation and review of literature.\nInt J Clin Exp Pathol. 2013;6:1652 –7.\n22. Park SY, Park MH, Ko HS, Cha EJ, Sohn JS, Jung US, et al. Dedifferentiated\nendometrioid adenocarcinoma of the uterus: highly aggressive and poor\nprognostic tumor. Korean J Pathol. 2014;48:327 –30.\n23. Soyama H, Takano M, Miyamoto M, Kato M, Goto T, Furuya K.\nDedifferentiated endometrioid adenocarcinoma of the uterus: a case report.\nEur J Gynaecol Oncol. 2016;37:426 –9.\n24. Rabban JT. Dedifferentiated endometrial carcinoma: An aggressive tumor\nthat may morphologically mimic lower-grade endometrioid\nadenocarcinoma. AJSP: Reviews & Reports. 2016;21:57 –72.\n•  We accept pre-submission inquiries \n  Our selector tool helps you to find the most relevant journal\n  We provide round the clock customer support \n  Convenient online submission\n  Thorough peer review\n  Inclusion in PubMed and all major indexing services \n  Maximum visibility for your research\nSubmit your manuscript at\nwww.biomedcentral.com/submit\nSubmit your next manuscript to BioMed Central \nand we will help you at every step:\nHan et al. World Journal of Surgical Oncology  (2017) 15:17 Page 9 of 9","source_license":"CC0","license_restricted":false}