{"paper_id":"09c09a09-78ca-49b9-9590-6c12c91e04e4","body_text":"R E S E A R C H Open Access\nSerous carcinoma arising from uterine\nadenomyosis/adenomyotic cyst of the\ncervical stump: a report of 3 cases\nBingjian Lu 1*, Qin Chen 1, Xiaofei Zhang 1 and Lili Cheng 2\nAbstract\nBackground: Serous carcinoma arising in adenomyosis and adenomyotic cyst is very rare. Only 3 serous carcinomas\nand 5 serous endometrial intraepithelial carcinomas (EIC) have been reported to date.\nMethods: We reviewed the clinicopathological features of 2 serous carcinoma in uterine adenomyosis and 1 serous\nEIC in adenomyotic cyst of the cervical stump.\nResults: Case 1 had an endometrial serous carcinoma in the uterine myometrium and the left ovary. A minor\ncomponent of benign endometrial glands with minimal endometrial stroma was found in the uterine mass\nand the surrounding myometrium. Case 2 showed 3 small foci of serous carcinoma, serous EIC and endometrial\nglandular dysplasia (EmGD) in the adenomyosis. Scanty serous carcinoma was present in the endometrium without\nevidence of myometrial invasion. The eutopic endometrium in both case 1 and 2 had no evidence of neoplastic\nchanges after complete examination. Case 3 had 3 microscopic serous EICs in the adenomyotic cysts of the\ncervical stump. One EIC lesion coexisted with EmGD. No cancer was found in the endocervical tube although\nthe preoperative endocervical biopsy showed a poorly differentiated endometrioid carcinoma. Immunohistochemistry\ndemonstrated that serous carcinoma in case 1 and EIC in all 3 cases showed a characteristic pattern of p53 and p16\nover expression, high Ki67 index, and lack of WT1, ER and PR staining. EmGD in case 1 and 3 had a similar staining\npattern except a lower Ki67 index and the presence of ER expression.\nConclusions: We believe that this case series may expand our reco g n i t i o no ns e r o u sc a r c i n o m aa r i s i n gi nu t e r i n e\nadenomyosis/adenomyotic cyst including extra-uterine spread and the potential synchronous growth of carcinomas in\neutopic endometrium.\nKeywords: Endometrial serous carcinoma, Serous intraepithe lial carcinoma, Endometrial glandular dysplasia,\nAdenomyosis, Adenomyotic cyst, Immunohistochemistry\nBackground\nCarcinomatous changes from the ectopic endometrial\nglands in endometriosis have been reported in many\nstudies [1 –4]. Most cases occur in patients with ovarian\nendometriosis. Malignant transformation from uterine\nadenomyosis is very unusual. Less than 50 cases have\nbeen reported currently although the first case was\nreported in 1897. The predominant histotypes of\nthese cases are endometrioid carcinoma and clear cell\ncarcinoma [4 –6]. Serous carcinoma arising in uterine\nadenomyosis is extremely rare. Only 3 reports con-\ntaining 3 serous carcinomas and 5 serous endometrial\nintraepithelial carcinomas (EIC) have been reported to\ndate [7 –10]. In this study, we present 2 serous carcin-\nomas in uterine adenomyosis and 1 serous EIC in\nadenomyotic cyst of the cervical stump.\nMethods\nThe 3 cases reported here were retrieved from the ar-\nchives of the Department of Surgical Pathology, the\nAffiliated Women’s Hospital School of Medicine Zhejiang\n* Correspondence: lbj@zju.edu.cn\n1Department of Surgical Pathology, the Affiliated Women ’s Hospital, School\nof Medicine Zhejiang University, Hangzhou, People ’s Republic of China,\n310006\nFull list of author information is available at the end of the article\n© 2016 The Author(s). 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.\nLu et al. Diagnostic Pathology  (2016) 11:46 \nDOI 10.1186/s13000-016-0496-0\n\nUniversity, China, in the last 5 years. Patients 1 and 2 were\nrecent cases. We obtained clinical details and follow-up\ndata from hospital medical records. This study was ap-\nproved by the Institutional Research Ethics Committee\nof the Affiliated Women ’s Hospital School of Medicine\nZhejiang University, China. The patients were well in-\nformed and consent with this study. Archival routine\nhematoxylin and eosin stained slides were assessed by\nthe authors.\nAdditional 4 μm sections were cut for immunohisto-\nchemistry. The sources and dilutions of a group of anti-\nbodies are detailed in Table 1. A two-step En Vision\nimmunostaining procedure (DAKO, Carpentaria, CA,\nUSA) was performed according to the manufacturer ’s\nprotocols. The percentage of positive cells was scored as\nfollows: − for no immunoreactivity; focally + for 1 % to\n5 %; + for 6 % to 25 %; ++ for 26 % to 50 %; +++ for\n51 % to 75 %; ++++ for 76 % to 100 %. Over expression\nof p16 and p53 were defined at least “+++”.\nResults\nClinical findings\nTable 2 summarizes the major clinical features of all 3\ncases.\nCase 1\nA 64-year-old Chinese woman, gravida 2, para 2, com-\nplained of left low abdominal pain for about 6 months.\nHer last menses was more than 10 years ago. She denied\nhistory of hormone replacement therapy. She had a his-\ntory of diabetes mellitus with the treatment of Diamicron\nand Metformin Hydrochloride for more than 10 years.\nShe denied familial history of cancer. Pelvic examination\nrevealed a slightly enlarged and symmetrical uterus. A\npelvic sonogram showed th at the endometrium strip\nmeasured 0.2 mm. An irregular mass of 2.5 × 2.6 ×\n2.6 cm was present in the left parametrium. A cystic\nlesion of 4.0 × 2.1 × 3.2 cm was found in the left\nadnexus. It was adjacent to the mass in the left para-\nmetirum. The serum levels of CA125, CA 199 and\nCA153 were normal. The clinical impression of this\ncase was “adenomyosis and possible left fallopian\ntube-ovarian cyst ”. Total abdominal hysterectomy and\nbilateral salpingo-oophorectmy (TAHBSO) was ini-\ntially performed. A 4.0-cm cystic lesion was found in\nthe left adnexus. The cystic lesion enveloped the left\novary and fallopian tube, and adhered with the uterus\nand intestine. However, the left ovary and fallopian\ntube was not clearly seen. The uterus and the right\nfallopian tube looked normal anatomically. After a\ndiagnosis of “adenocarcinoma in the left adnexus ” was\nrendered on frozen sections, a complete staging sur-\ngery including retroperitoneal lymph node dissection\nand omentectomy was subsequently carried out. A\nsmall focus of endometriosis from the rectal adventi-\ntia was also biopsied. The patient was given one\ncourse of TP (Taxol + Carboplatin) chemotherapy and\nwas unremarkable for one month after surgery at\npresent.\nCase 2\nA 55-year-old Chinese woman gravida 1, para 1,\ncomplained of postmenopausal vagina bleeding for\n8 months. Uterine adenomyosis and multiple leiomyo-\nmas were found 19 years ago. She was regularly\nfollowed-up thereafter. Her serum CA199 was slightly\nelevated (89.8 IU/mL) 4 months ago. Serum CA125\nwas normal. Recent pelvic sonogram indicated that the\nenlarged uterus had a size similar to that of 2 weeks of\ngestation and the endometrium strip measured\n0.22 cm. There were multiple leiomyomas in the uterus.\nThe largest one measured 5.6 × 5.5 × 5.2 cm. Pelvic CT\nscan and MRI had similar findings to that of sonogram.\nHer medical history was significant for diabetes mellitus,\nprimary hypertension, asthma, appendectomy and caesar-\nean. The diagnosis of preoperative endometrial curetting\nwas serous carcinoma. TAHBSO and complete staging\nsurgery was performed 4 weeks after the diagnosis. One\ncourse of adjuvant chemotherapy (TP) was given postop-\neratively by now.\nCase 3\nA 55-year-old Chinese woman, gravida 2, para 1, com-\nplained of vaginal bleeding for 22 days. She had under-\ngone a subtotal hysterectomy for uterine adenomyosis\nTable 1 Antibody clones, sources, and dilutions\nAntibody Clones Dilutions Sources\nER 1D5 1:300 Thermo\nPR 1A6 1:500 Thermo\nCK7 OV-TL 1:100 Thermo\nCK20 KS20.8 1:100 Leica\nPAX8 ZR1 1:50 Invitrogen\nWT1 6 F-H2 1:150 Zeta\nCA125 EPR1020(2) 1:50 Zeta\np53 DO-7 1:600 Thermo\nCD10 56C6 1:100 Leica\nD2-40 D2-40 1:400 Thermo\np16 16P04/JC2 1:100 Zeta\nPTEN 17A 1:50 Zymed\nβ-catenin E247 1:400 Zymed\nHNF1β C-20 1:200 Santa Cruz\nKi67 MIB-1 1:400 DAKO\nLu et al. Diagnostic Pathology  (2016) 11:46 Page 2 of 10\n\n16 years ago. Sonogram revealed a 4.4 × 4.4 × 3.0 cm\nmass in the cervical stump and a 3.3 × 3.5 × 2.8 cm cyst\nin the left ovary. Pelvic CT scan showed an enlarged cer-\nvical stump with a 2.5 × 1.7 × 1.5 cm mass and a 2.8 ×\n2.4 × 3 cm cyst in the left ovary. The pathological\ndiagnosis of her endocervical curetting was poorly dif-\nferentiated endometrioid carcinoma. Her familial his-\ntory was unremarkable except that her father died of\nlung cancer. Radical resection of the cervical stump,\nbilateral salpingo-oophorectmy (BSO) and complete\nstaging surgery were performed two weeks later. The\nrectum was partially removed due to the presence of\nendometriosis resulting in its dense adhesion with the\ncervix stump. She was alive with no evidence of can-\ncer for at least 44 months.\nPathological findings\nThe main pathological and immunohistochmical find-\nings of these cases are given in Table 3 and Table 4,\nrespectively.\nCase 1\nThe uterus measured 6.5 × 3 × 3 cm. A 2.2 × 2.3 × 2.0-cm\nmass was found in the outer two third of the uterine\nmyometrium. It extended from the left side of the cor-\npus to the isthmus. The mass had a white and coarsely\ntrabeculated cutting surface. The texture was similar to\nthat of the surrounding myometrium. The boundary of\nthe mass was not very clear. The myometrium surround-\ning the mass had foci of hemorrhage. A leiomyoma with\na diameter of 0.5 cm was found in the isthmus. The\nendometrium was atrophic and about 0.1 cm thick. The\nuterine cervix was unremarkable. The specimens from\nthe left adnexa ( n = 4), the right fallopian tube ( n = 2),\nthe uterine mass, the endometrium ( n =9 ) a n d t h e\ncervix ( n = 8) were all extensively sectioned for histo-\npathological examination.\nThe broken specimens from the left adnexus measured\n4 × 3 × 2 cm. The fallopian fimbriae and ovary was\ngrossly unrecognizable, but the fallopian tube was par-\ntially recognizable with a diameter of 1.0 cm. The tubal\nTable 2 Clinical findings of the lesions\nCase 1 Case 2 Case 3\nAge (yr) 64 55 55\nHistory of pregnancy G2P2 G1P1 G2P1\nHistory of cancer N N N\nFamilial history of cancer N N Her father died of lung cancer.\nClinical presentation Left low abdominal pain Postmenopausal vagina bleeding Vaginal bleeding\nSerum tumor biomarkers Normal CA125, CA199 & CA153 CA199: 89.8 IU/mL; Normal CA125 ND\nImaging findings Endometrium 0.2 cm; A mass in\nthe left parametrium; A cyst in\nthe left adnexus\nEndometrium 0.22 cm; Uterine\nleiomyomas\nA mass in the cervical stump and a cyst\nin the left ovary\nTreatment TAHBSO + complete staging\nsurgery\nTAHBSO + complete staging surgery Radical resection of the cervical stump + BSO\n+ complete staging surgery\nPostoperative chemotherapy Y Y N\nStatus of follow-up (time) Recent case Recent case Alive with no evidence of cancer (44 months)\nAbbreviations: N no, Y yes, ND not done, BSO bilateral salpingo-oophorectmy, TAHBSO total abdominal hysterectomy and bilateral salpingo-oophorectmy\nTable 3 Pathological results of the lesions\nCase 1 Case 2 Case 3\nFIGO Stage IIIa Ia Ia\nGross uterine mass - - -\nPathological findings in the endometrium - SC EC (G2-3)\nMyometrium invasion NA - -\nLymphovascular tumor emboli + + -\nFindings in the adenomyosis/ endometriotic cyst SC Minimal SC; serous EIC; EmGD Serous EIC; EmGD\nOther findings Spread to the left ovarian;\nUterine leiomyoma;\nEndometriosis in the rectum.\nUterine leiomyoma Glandular cyst in the cervix stump;\nLeft ovarian endometriotic cyst;\nEndometriosis in the rectum.\nAbbreviations: SC serous carcinoma, EIC serous endometrial intraepithelial carcinoma, EmGD endometrial glandular dysplasia, EC endometrioid carcinoma, NA\nnot applicable\nLu et al. Diagnostic Pathology  (2016) 11:46 Page 3 of 10\n\nwall measured 0.3-0.4 cm. Some cystic wall-like tissue\nhad a thickness of 0.1-0.2 cm. Two pieces of grey-to-\nwhite, tender solid tissues, measured 1.5 × 1.0 × 0.8 cm\nand 0.8 × 0.5 × 0.5 cm, separately, were present in the\nbroken tissues. The right fallopian tube and ovary, and\nomentum were grossly normal.\nMicroscopically, the uterine mass showed neoplastic\nglands splitting smooth muscle fibers (Fig. 1a). It displayed\nan expansive growth pattern towards the surrounding\nmyometrium. The neoplastic glands were lined by\nmarkedly atypical, cuboidal to columnar cells with\nhyperchromasia, conspicuous nucleoli, and irregular\nnuclear membrane and numerous mitotic figures\n(>12/10HPFs) (Fig. 1b). The luminal borders were\nsmooth, and some had epithelial tufts. Small clusters of\natypical cells surrounded by a clear space were commonly\nseen. However, desmoplastic stroma was inconspicuous.\nA small amount of benign endometrial glands with min-\nimal endometrial stroma were present in the adjacent\nnormal myometrium surrounding the tumor and within\nthe tumor, implicating the presence of adenomyosis\n(Fig. 1a, c). Tumor emboli within the lymphovascular ves-\nsels were occasionally found within the myometrium. The\natrophic endometrium showed resting changes without\nneoplastic evidence including cancer precursor alter-\nations. The small nodule in the isthmus was a leiomyoma.\nThe solid tissues in the left adnexus showed the same\nmorphological changes as in the uterine mass in general\n(Fig. 1d). However, they showed a vaguely multi-nodular\ngrowth pattern with a prominent desmoplastic reaction.\nThere were no benign-looking endometrioid glands in\nthe tumor. Ovarian stroma was discernable at the per-\niphery of the lesion. The left fallopian tube and the cystic\nwall were in keeping with hydrosalpinx and showed no\nevidence of serous tubal intraepithelial carcinoma (STIC)\nor serous carcinoma. Biopsy from the rectal adventitia\nwas diagnosed as endometriosis.\nTumors from the uterine and ovarian mass showed\nan identical immunohistochemical profile (Table 4;\nFig. 1e-g). Notably, CD10 clearly demonstrated the\npresence of endometrial stroma surrounding the benign\nglands in the uterine tumor (Fig. 1h).\nCase 2\nThe uterus measured 13 × 9 × 6.5 cm. There were\nmultiple leiomyomas in the myometrium. The largest\none had a diameter of 5.0 cm. Adenomyosis was dif-\nfusely found in the myometrium. It was characterized\nby the trabeculated cut surface and small foci of\nhemorrhage. No gross tumor was identified in the\nuterine cavity. The endometrium measured 0.1 cm\nand looked unremarkable. The uterus including endo-\nmetrium and myometroum ( n = 17) were completely\nexamined. The fallopian fimbriae ( n = 2 for each side)\nwere also extensively investigated.\nThe preoperative endometrial curetting consisted of a\nsmall amount of specimen (measured 0.3 × 0.3 × 0.2 cm)\nwith artifact crush. It showed a typical pattern of serous\ncarcinoma including the papillary structure with a slen-\nder fibrous core and the highly atypical lining epithelial\ncells. In the resected uterus, only a small amount of ser-\nous carcinoma was found in 3 foci (each with a diameter\nTable 4 Immunostaining results of the lesions a\nAntibodies Case 1 Case 2 Case 3\nAdenomyosis Left Ovary Curetting Adenomyosis Curetting Adenomyosis\nE R - - -- --\nP R - - -- --\nCK7 +++ +++ ND ND ND ND\nCK20 - - ND ND ND ND\nPAX8 +++ +++ +++ +++ +++ +++\nW T 1 - - -- --\nCA125 ++ ++ ND ND ND ND\np53 +++ +++ +++ +++ +++ +++\nCD10a - - ND ND ND ND\nD2-40a - - - - ND ND\nPTEN ++ ++ ++ ++ ND ND\nβ-catenin Mem+ Mem+ Mem+ Mem+ ND ND\np16 +++ +++ +++ +++ +++ +++\nHNF1β ND ND - - - -\nKi67 >80 % >80 % >70 % >70 % >80 % >80 %\nAbbreviations: ND not done, Mem membrane staining\naCD10 and D2-40 was used to highlight the presence of endometrial stroma and lymphovascular invasion in the adenomyosis, respectively\nLu et al. Diagnostic Pathology  (2016) 11:46 Page 4 of 10\n\nof <0.5 mm) from 17 slides from the whole endomet-\nrium with partial myometrium. They were all located\nwithin the endometrium, and did not show evidence of\nmyometrial invasion. Serous EIC and Endometrial glan-\ndular dysplasia (EmGD) was not found in the eutopic\nendometrium. Three small foci of serous neoplasia\n(2 mm-3 mm in diameters) were found from 2 of 12\nslides from the region of adenomyosis. They were clearly\nseparated from the basalis of the endometrium. Two le-\nsions showed a minimal serous carcinoma and a serous\nEIC (Fig. 2a, 2g). The neoplastic cells showed significant\natypia including nuclear pleomorphism and prominent\nnucleoli (Fig. 2b). The third lesion contained a predom-\ninant component of EmGD (Fig. 2c) and a minor serous\nEIC. All these lesions were surrounded by stromal cells.\nLymphovascular invasion was occasionally found in the\nmyometrium. Immunohistochemistry showed that both\nSC in the endometrial curetting and SC/serous EIC in\nthe adenomyosis had the same immunostaining pattern\n(Table 4). The neoplastic cells were diffusely positive for\np53 and Ki67, and negative for ER, PR and WT1\n(Fig. 2h). The cells in EmGD showed an identical\nFig. 1 Serous carcinoma in uterine adenomyosis (case 1). The uterine mass showed neoplastic glands with marked atypia splitting smooth\nmuscle fibers a, b. A small amount of benign endometrial glands with minimal endometrial stroma were found at the periphery of the tumor\nand within the tumor, implicating the presence of adenomyosis a, c: arrows. The identical morphological changes were seen in the left adnexus\nd. Both uterine and ovarian tumors showed an identical immunohistochemical profile including p53 over expression e, high Ki67 index f, and\nWT1 -ve g. CD10 staining highlights the presence of endometrial stroma surrounding the benign glands in the uterine tumor h.( O r i g i n a l\nmagnifications: A*25; B*400; C-G*200)\nLu et al. Diagnostic Pathology  (2016) 11:46 Page 5 of 10\n\nstaining pattern except for a moderate ER expression\nand a relatively lower Ki67 index (approximately 40 %)\n(Fig. 2d-f ).\nCase 3\nThe cervical stump measured 5 × 5 × 4.5 cm. The mu-\ncosa looked a little bit rough, but no tumor was grossly\nseen in the cervical canal. A 3 × 1.5 × 1.5-cm mass was\nrecognizable in the myometrium. The mass had 3 small\ncysts containing brownish, viscous fluid. The diameters\nof these cysts varied between 0.5-1.0 cm. These cysts\nwere at least 7 mm from the endocervical surface. The\nleft ovarian cyst showed the similar gross features and\nhad a size of 4 × 3.5 × 3 cm. The resected rectum mea-\nsured 5.5 cm long. The intestinal wall was partly scleros-\ning and hemorrhagic. The lumen became narrow, but\nhad no macroscopic tumors or polyps. The cervical\nstump (n = 15) was extensively examined for microscopic\nassessment.\nThe preoperative endocervical biopsy showed a pre-\ndominantly solid pattern and a minor component of\nglandular structures in keeping with a poorly differenti-\nated endometrioid carcinoma (grade 2 –3). However, no\nresidual tumors were found in the removed cervical\nstump. A 0.4-cm lesion of hemorrhage with abundant\nhemosiderin-containing macrophages was noted at the\nFig. 2 Serous EIC in uterine adenomyosis (case 2). A small focus of serous carcinoma and serous EIC (arrows) was present in the adenomyosis\na. The neoplastic cells showed significant atypia b. The pattern of endometrial glandular dysplasia (EmGD) was shown in another field c. Higher\nmagnification (original magnification *400) showed the presence of nuclear atypia in EmGD (inserted in the left lower corner of c). The cells in\nEmGD showed p53 over expression d, a relatively lower Ki67 index (approximately 40 %) e, and a moderate ER expression (++) f. The neoplastic\ncells in serous carcinoma and EIC ( arrows) were diffusely positive for p53 g. (Original magnifications: A*50; B-G*200)\nLu et al. Diagnostic Pathology  (2016) 11:46 Page 6 of 10\n\nsurface of the cervical mucosa. Three microscopic foci\nof serous EIC (each 1 mm in diameter) were noted in 2\nof 3 adenomyotic cysts in the cervical stump. They had\na slightly papillary contour with fibrous cores and the\ncells displayed hobnail cellular morphology and signifi-\ncant atypia (Fig. 3a, b). A focus of EmGD was seen in\nthe adjacent area of one serous EIC lesion (Fig. 3e, f ).\nEndometriosis was present in 4 slides from the sur-\nrounding cervical myometrium and parametrium. The\nleft ovarian lesion was an endometriotic cyst. The rectal\nserosa and muscularis propria were significant for endo-\nmetriosis. The immunostaining results of carcinoma in\nthe endocervical biopsy and serous EIC in the adeno-\nmyotic cyst was shown in Table 4 (Fig. 3g, h). The\nEmGD lesion had an expression pattern identical to\nserous EIC except the lower Ki67 index (30 %) and\npatchy p16 staining (Fig. 3e, f ).\nDiscussion\nCarcinoma arising from adenomyosis was initially re-\nported more than one century ago. Sampson or Colman ’s\ncriteria for the diagnosis of carcinoma arising from adeno-\nmyosis were proposed early in 1950s [11, 12]. These in-\ncludes: “1) the carcinoma must not be present in the\nnormally situated endometrium or elsewhere in the pelvis;\nb) the carcinoma must be seen to arise from the epithe-\nlium within the adenomyosis rather than invasion from\nanother source; and c) endometrial (adenomyotic) stromal\ncells must be seen to support a diagnosis of adenomyosis ”.\nIn fact, these criteria appear to be too stringent. In\nFig. 3 Serous EIC in adenomyotic cyst of the cervical stump (case 3). Serous EIC was present in the adenomyotic cysts by showing a slightly\npapillary contour with fibrous cores and the cells displayed hobnail cellular morphology and significant atypia a, b. The serous EIC showed p53\nover expression c and a high Ki67 index d. A focus of EmGD and serous EIC (arrow) was seen in the adjacent area of one serous EIC lesion e, f.\nThe EmGD lesion showed p53 over expression g and a relatively lower Ki67 index h. (Original magnifications: A, E*50; C, D*100; B, F-H*200)\nLu et al. Diagnostic Pathology  (2016) 11:46 Page 7 of 10\n\nprevious reports, 3 of 5 serous EICs and 1 of 4 car-\ncinomas could not be classified as true tumors arising\nin adenomyosis if all 3 criteria should fully be satis-\nfied [8, 9]. However, the authors provided compelling\nevidence to support that their cases should be consid-\nered as originating from adenomyosis.\nIn this study, we also understand that our cases do not\nmeet the Sampson or Colman ’s criteria completely, but\nwe think that they are most likely from adenomyosis as\nwe discuss below. The point most contradictory to the\ncriteria in case 1 is the coexistence of uterine and ovar-\nian carcinoma with the identical morphological and im-\nmunostaining features. The question arises that both\ncarcinomas are primary or one is metastatic from the\nother. However, we believe that the ovarian lesion is un-\nlikely to be primary according to the following findings.\nFirst, the ovarian carcinoma was grossly unapparent\nwhile the uterine carcinoma was obvious. The uterine\ntumor was deeply located in the myometrium. It had a\nrelatively clear boundary and grossly looked like an ade-\nnomyoma (adenomyosis). Moreover, there was no tumor\nin the pelvis beyond the left ovary and uterus. In particu-\nlar, serous carcinoma and its precursor lesions were not\nseen in the endometrium and cervix even after an exten-\nsive microscopic inspection on these sites. Second, the\novarian carcinoma was characterized by overwhelmingly\nglandular structures with severe cytological atypia, but\nno papillary formation was seen. The great disagreement\nof structure (G1) and nuclear grading (G3) is a classical\ndiagnostic clue for endometrial serous carcinoma [13]. A\ndiffuse glandular pattern is not a common feature for\novarian serous carcinoma. Moreover, no STIC and ovar-\nian endometriosis was found in both fallopian tubes and\novaries. The frequency of STICs was 61 % in pelvic high\ngrade serous carcinoma [14]. Endometriotic cyst can be\nan uncommon source of ovarian or pelvic serous carcin-\noma [2, 15, 16]. In contrast, the uterine carcinoma\nshowed a pattern of neoplastic glands intermingled with\nsmooth muscles, architecturally resembling an adeno-\nmyoma. Endometriosis was found in the tumor and the\nsurrounding myometrium. Third, both uterine and ovar-\nian cancers were totally negative for WT1 and ER, and\nhad p53 over expression and a high Ki67 index. It was\nwell documented that WT1 and ER expression was com-\nmonly present in ovarian serous carcinomas with a fre-\nquency of approximately 90 %, and very unusual in\nendometrial serous carcinoma [17 –19]. Therefore, the\nWT1-ve and ER -ve pattern in both tumors of this case\nindicated that serous carcinoma in the uterine myome-\ntrium was most likely primary from adenomyosis, which\nhad largely been replaced by tumor overgrowth. Accord-\ningly, the ovarian carcinoma was secondary from the dir-\nect extension or lymphovascular spread of carcinoma in\nthe uterine adenomyosis.\nThe presence of carcinoma in the endometrial cav-\nity (case 2) and endocervical tube (case 3) may throw\ndoubt on the origin of the endometriotic carcinoma\nfrom adenomyosis. We think that these cases can be\nexplained by synchronous growth of endometrial and\nendometriotic carcinomas as reported previously [8].\nIn case 2, adenomyosis was diffuse and mostly was\nuninvolved by the serous cancer cells. The 3 involved\nfoci of adenomyosis were separated clearly from the\nbasalis of the endometrium while the small endomet-\nrial carcinoma was completely limited within the\nendometrium. In case 3, there was no residue tumor\nat the surface of the endocervix in the surgical speci-\nmens. The histotypes were different between the ori-\nginal endocervical biopsy and the lesions in the\ncervical adenomyosis. These lesions in the adenomyo-\ntic cyst were at least 7 mm from the endocervical\nsurface. Moreover, all these microscopic lesions in the\nmyometrium from case 2 and case 3 were clearly sur-\nrounded by benign stromal cells. They had foci of\nEmGD and serous EIC. Serous EIC is the established\nprecursor lesion of endometrial serous carcinoma\nwhile EmGD is an emerging early lesion preceding\nEIC [20 –22]. Kumar et al. [23] also emphasized that\nthe presence of either a transition or continuity be-\ntween the benign adenomyotic endometrial glands\nand the carcinomatous glands was essential to prove\nthe adenomyotic origin of the carcinoma and to ex-\nclude the possibility of metastasis or invasion from\nelsewhere. The coexistence of EmGD and EIC in our\ncases indicates the interrelationships between the ec-\ntopic endometrium and serous carcinoma, thus makes\nmetastasis from the eutopic endometrium most unlikely.\nTwo additional points in our cases deserved further\ndiscussion. Lymphovascular invasion is uncommon in\nEIC although EIC can spread in the peritoneum [24].\nHowever, the distinction of extensive EIC and early\ninvasive serous carcinoma has not been well estab-\nlished. According to the criteria suggested by Dr\nWheeler DT, et al. [25], serous carcinoma in the\nendometrium and adenomyosis from case 2 may both\nrepresent for a minimally invasive serous carcinoma\nand can cause lymphovascular invasion. However, we\nthink that the lymphovascular invasion is most likely\nassociated with serous carcinoma from adenomyosis\nbecause it is present in the myometrium. Moreover,\nthe endometrial serous carcinoma did not show evi-\ndence of myometrial invasion. In case 3, the presence\nof hobnail cell changes in EIC showed overlapping\nfeatures with minute clear cell carcinoma or clear cell\ncarcinoma in situ. However, such hobnail cellular\nchanges can also be present in EIC [26]. HNF1 β and\nNapsin A are typically expressed by clear cell carcin-\noma [27, 28], but they are completely negative in this\nLu et al. Diagnostic Pathology  (2016) 11:46 Page 8 of 10\n\ncase (data not shown). Taken the presence of EmGD\ntogether, our case is regarded as EIC most likely.\nTwo broad categories, type I and II, endometrial car-\ncinomas have been well documented [29]. On the basis\nof these two categories, a “dualistic model ” of endomet-\nrial carcinogenesis has been developed recently with\nthe accumulative clinical, morphological and molecular\nevidence [30]. The prototype of type I carcinoma is\nendometrial carcinoma, which is associated with un-\nopposed estrogen stimulus, and favorable prognosis.\nSerous carcinoma is the prototype of type II carcinoma.\nIt is unrelated to the effect of unopposed estrogen. It\ninvariably harbors p53 mutation and has poor progno-\nsis. The type I carcinoma is much more common than\ntype II carcinoma in the uterine endometrium. Carcin-\noma from ectopic endometrium is not very uncommon,\nand most in ovarian endometriosis. The ovarian carcin-\noma associated with endometriosis predominantly rep-\nresents for endometrioid and clear cell types [4 –6]. In\nthe less than 50 well documented carcinomas arising\nfrom adenomyosis, only three were serous carcinoma\nor serous papillary carcinoma [7 –10]. These data sug-\ngest that carcinomas arising from the ectopic endomet-\nrium including adenomyosis are predominantly type I\nover type II. Our report not only adds new cases to this\nrare carcinoma arising from adenomyosis, but also pro-\nvides a characteristic immuno staining profile (PTEN+,\nER-, membranous β-catenin+, p53 and p16 over expres-\nsion, high Ki67 index) to uncover the distinct molecular\nalterations in type II carcinoma. Abushahin et al. [8]\np r o v e dt h ep r e s e n c eo fp 5 3m u t a t i o ni n1o f5s e r o u s\nEICs arising in adenomyosis. That study also inferred\nthat EmGD and serous EIC, arising in adenomyosis\nmay be underdiagnosed because of their small foci in\nadenomyosis. In favor of their opinion, we observed\nthat serous carcinomas in adenomyosis may also be\nunder-recognized particularly when carcinoma occurred\nsimultaneously in the endometrium or outside the uterus.\nCarcinomas of the cervical stump after subtotal\nhysterectomy are not uncommon with an incidence\nof 0.1-3 %. Among these carcinomas, approximately\n88 % were squamous cell carcinoma and the\nremaining were adenocarcinoma [31, 32]. However,\nwe could not find any reports on adenocarcinoma or\nserous EIC arising in endometriosis (adenomyotic\ncyst) of the cervical stump in English literature to\nd a t e .C a s e3d e m o n s t r a t e dt h ep r e s e n c eo fs e r o u s\nEIC in adenomyotic cyst in the cervical stump fol-\nlowing the subtotal hysterectomy for adenomyosis.\nGynecologists might take this rare complication into\naccount in addition to the p reservation of possible\nphysiologic and sexual fu nctions of the cervix when\nthey decided to perform a proper surgery for women\nwith diffuse uterine adenomyosis.\nConclusions\nIn summary, our cases present a rare phenomenon\nthat serous carcinoma and EIC can arise from ectopic\nendometrium including uterine adenomyosis and ade-\nnomyotic cyst in the cervical stump. Their small size\nas well as their inconspicuous histological features\nunder scanning magnification may cause these lesions\nto be overlooked. Such cases may potentially become\nthe unrecognizable source of extra-uterine or pelvic\nspread. These cases highlight the necessity of regular\nclinical follow up in patients with untreated uterine\nadenomyosis or subtotal hysterectomy for adenomyo-\nsis, and more meticulous pathological inspection on\nthe resected uterine specimens. However, more stud-\nies are necessary to investigate the epidemiology and\npathogenesis in serous carcinoma and EIC arising\nfrom ectopic endometrium, and to develop reliable\napproaches to delineate these cases from benign le-\nsions preoperatively.\nAbbreviations\nEIC, endometrial intraepithelial carcinoma; EmGD, endometrial glandular\ndysplasia; TAHBSO, total abdominal hysterectomy and bilateral salpingo-\noophorectmy\nAcknowledgement\nWe especially thank Mrs. Caiyun Zhou and Ying Shao for their technical\nsupport in this manuscript. This work was supported by a grant from the\nNational Natural Science Foundations of China (81372790).\nAuthors’ contributions\nB.L.-manuscript preparation, case collection and pathological examination.\nQ.C.-pathological examination. X.Z.-pathological examination. L.C.-clinical\ndata collection. All authors 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 each patient for publication of\ntheir Case Report and any accompanying images. A copy of the written\nconsent is available for review by the Editor-in-Chief of this journal.\nEthics approval and consent to participate\nThis study was approved by the Institutional Research Ethics Committee of\nthe Affiliated Women ’s Hospital School of Medicine Zhejiang University,\nChina.\nAuthor details\n1Department of Surgical Pathology, the Affiliated Women ’s Hospital, School\nof Medicine Zhejiang University, Hangzhou, People ’s Republic of China,\n310006. 2Department of Gynecologic Oncology, Women ’s Hospital, School of\nMedicine, Zhejiang University, Hangzhou, Zhejiang Province, People ’s\nRepublic of China.\nReceived: 5 January 2016 Accepted: 28 May 2016\nReferences\n1. Pearce CL, Templeman C, Rossing MA, Lee A, Near AM, Webb PM, Ovarian\nCancer Association Consortium, et al. Association between endometriosis\nand risk of histological subtypes of ovarian cancer: a pooled analysis of\ncase–control studies. Lancet Oncol. 2012;13:385 –94.\n2. Boyraz G, Selcuk I, Yaz ıcıoğ lu A, Tuncer ZS. Ovarian carcinoma associated\nwith endometriosis. Eur J Obstet Gynecol Reprod Biol. 2013;170:211 –3.\nLu et al. Diagnostic Pathology  (2016) 11:46 Page 9 of 10\n\n3. Acién P, Velasco I, Acién M, Capello C, Vela P. Epithelial ovarian cancers and\nendometriosis. Gynecol Obstet Invest. 2015;79:126 –35.\n4. Koike N, Tsunemi T, Uekuri C, Akasaka J, Ito F, Shigemitsu A, Kobayashi H.\nPathogenesis and malignant transformation of adenomyosis (review). Oncol\nRep. 2013;29:861–7.\n5. Baba A, Yamazoe S, Dogru M, Ogawa M, Takamatsu K, Miyauchi J. Clear cell\nadenocarcinoma arising from adenomyotic cyst: a case report and literature\nreview. J Obstet Gynaecol Res. 2016;42:217 –23.\n6. Kazandi M, Zeybek B, Terek MC, Zekioglu O, Ozdemir N, Oztekin K. Grade 2\nendometrioid adenocarcinoma arising from adenomyosis of the uterus:\nreport of a case. Eur J Gynaecol Oncol. 2010;31:719 –21.\n7. Izadi-Mood N, Samadi N, Sarmadi S, Eftekhar Z. Papillary serous carcinoma\narising from adenomyosis presenting as intramural leiomyoma. Arch Iran\nMed. 2007;10:258–60.\n8. Abushahin N, Zhang T, Chiang S, Zhang X, Hatch K, Zheng W. Serous\nendometrial intraepithelial carcinoma arising in adenomyosis: a report of 5\ncases. Int J Gynecol Pathol. 2011;30:271 –81.\n9. Koshiyama M, Suzuki A, Ozawa M, Fujita K, Sakakibara A, Kawamura M, et al.\nAdenocarcinomas arising from uterine adenomyosis: a report of four cases.\nInt J Gynecol Pathol. 2002;21:239 –45.\n10. Griffin M, Parai S, Ghatage P, Dimakulangan A. Papillary serous carcinoma of\nthe uterus arising in an adenomyoma. Acta Cytol. 1996;40:849 –50.\n11. Sampson JA. Endometrial carcinoma of the ovary arising in endometrial\ntissue of that organ. Arch Surg. 1925;10:1 –72.\n12. Colman HI, Rosenthal AH. Carcinoma developing in areas of adenomyosis.\nObstet Gynecol. 1959;14:342 –8.\n13. Garg K, Soslow RA. Strategies for distinguishing low-grade endometrioid\nand serous carcinomas of endometrium. Adv Anat Pathol. 2012;19:1 –10.\n14. Przybycin CG, Kurman RJ, Ronnett BM, Shih IM, Vang R. Are all pelvic\n(nonuterine) serous carcinomas of tubal origin? Am J Surg Pathol. 2010;34:\n1407–16.\n15. Takeuchi K, Yamanaka Y, Hamana S, Ohara N, Maruo T. Invasive\nadenocarcinoma arising from uterine adenomyosis involving the\nrectosigmoid colon. Int J Gynecol Cancer. 2004;14:1004 –6.\n16. Slomovitz BM, Soslow RA, Chang RC, Golub R, Kuo DY. Serous\nadenocarcinoma of the inguinal region arising from endometriosis followed\nby a successful pregnancy. Gynecol Oncol. 2002;87:152 –4.\n17. Bárcena C, Oliva E. WT1 expression in the female genital tract. Adv Anat\nPathol. 2011;18(6):454–65.\n18. Hirschowitz L, Ganesan R, McCluggage WG. WT1, p53 and hormone\nreceptor expression in uterine serous carcinoma. Histopathology. 2009;55:\n478–82.\n19. Zhang Y, Garcia-Buitrago MT, Koru-Sengul T, Schuman S, Ganjei-Azar P. An\nimmunohistochemical panel to distinguish ovarian from uterine serous\npapillary carcinomas. Int J Gynecol Pathol. 2013;32:476 –81.\n20. Sherman ME, Bitterman P, Rosenshein NB, Delgado G, Kurman RJ. Uterine\nserous carcinoma. A morphologically diverse neoplasm with unifying\nclinicopathologic features. Am J Surg Pathol. 1992;16:600 –10.\n21. Zheng W, Xiang L, Fadare O, Kong B. A proposed model for endometrial\nserous carcinogenesis. Am J Surg Pathol. 2011;35:e1 –e14.\n22. Zheng W, Liang SX, Yu H, Rutherford T, Chambers SK, Schwartz PE.\nEndometrial glandular dysplasia: a newly defined precursor lesion of uterine\npapillary serous carcinoma. Part I: morphologic features. Int J Surg Pathol.\n2004;12:207–23.\n23. Kumar D, Anderson W. Malignancy in endometriosis interna. J Obstet\nGynaecol Br Emp. 1958;65:435 –7.\n24. Soslow RA, Pirog E, Isacson C. Endometrial intraepithelial carcinoma with\nassociated peritoneal carcinomatosis. Am J Surg Pathol. 2000;24:726 –32.\n25. Wheeler DT, Bell KA, Kurman RJ, Sherman ME. Minimal uterine serous\ncarcinoma: diagnosis and clinicopathologic correlation. Am J Surg Pathol.\n2000;24:797–806.\n26. Ellenson LH, Ronnett BM, Kurman RJ. Precursor lesions of endometrial\ncarcinoma. In: Kurman RJ, Ronnett BM, Ellenson LH, editors. Blaustein ’s\npathology of the female genital tract. 6 th. New York: Springer; 2011. p.\n360–91.\n27. Yamashita Y, Nagasaka T, Naiki-Ito A, et al. Napsin A is a specific marker for\novarian clear cell adenocarcinoma. Mod Pathol. 2015;28:111 –7.\n2 8 . F a d a r eO ,Z h a oC ,K h a b e l eD ,P a r k a s hV ,Q u i c kC M ,G w i nK ,D e s o u k i\nMM. Comparative analysis of Napsin A, alpha-methylacyl-coenzyme A\nracemase (AMACR, P504S), and hepatocyte nuclear factor 1 beta as\ndiagnostic markers of ovarian clear cell carcinoma: an immunohistochemical\nstudy of 279 ovarian tumours. Pathology. 2015;47:105–11.\n29. Bokhman JV. Two pathogenetic types of endometrial carcinoma. Gynecol\nOncol. 1983;15:10–7.\n30. Lax SF, Kurman RJ. A dualistic model for endometrial carcinogenesis based\non immunohistochemical and molecular genetic analyses. Verh Dtsch Ges\nPathol. 1997;81:228 –32.\n31. Hellström AC, Hellman K, Pettersson BF, Andersson S. Carcinoma of the\ncervical stump: fifty years of experience. Oncol Rep. 2011;25:1651 –4.\n32. Hellström AC, Sigurjonson T, Pettersson F. Carcinoma of the cervical stump.\nThe radiumhemmet series 1959 –1987. Treatment and prognosis. Acta\nObstet Gynecol Scand. 2001;80:152 –7.\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:\nLu et al. Diagnostic Pathology  (2016) 11:46 Page 10 of 10","source_license":"CC0","license_restricted":false}