{"paper_id":"7f951e46-df58-4b11-9293-fb1e102ac4f8","body_text":"CASE REPORT\nSmall cell carcinoma in endometrium on the base of extensive\nadenomyosis: differential diagnosis with immunochemistry\nSaba Kiremitci • Korhan Kahraman •\nAyse Sertcelik • Fırat Ortac\nReceived: 7 September 2011 / Accepted: 28 October 2011 / Published online: 20 December 2011\n/C211The Japan Society of Clinical Oncology 2011\nAbstract Small cell carcinomas (SCCs) are rarely seen in\nthe endometrium and usually present as a component of a\ncombined tumor consisting of more differentiated endo-\nmetrial tumors. SCCs commonly represent the deeply\ninvasive and aggressive component of the tumor, and dif-\nferentiating SCC from counterparts has a great importance\nin terms of different treatment modalities and worse\nprognosis. Differential diagnosis includes distinct tumors\nindistinguishable from SCC by histologic and morphologic\nfeatures, and the definite diagnosis requires immunohisto-\nchemical proof of diffuse neuroendocrine differentiation.\nWe present a case of combined endometrial tumor com-\nposed of SCC, endometrioid adenocarcinoma, and serous\ncarcinoma, which was initially misdiagnosed as carcino-\nsarcoma at another institute, and discuss the differential\ndiagnosis in terms of clinicopathologic features.\nKeywords Endometrium /C1Neuroendocrine tumor /C1\nSmall cell carcinoma\nIntroduction\nNeuroendocrine tumors (NETs) are a well-documented\nheterogeneous group of neoplasms originating from NE\ncells, most frequently seen in the gastrointestinal tract and\nbronchopulmonary system [ 1]. The female genital tract is\nnot a common site for NETs and there have been a limited\nnumber of reports in the literature involving the most\ncommon site—the cervix, followed by the ovary [ 2].\nHowever such a tumor occurring in the endometrium is\nvery rare. Poorly differentiated (PD) NETs, i.e., small cell\ncarcinoma (SCC) and large cell neuroendocrine carcinoma\n(LCNEC), are a distinct group with a high proliferation\nindex and poor prognosis, and chemotherapy is the treat-\nment of choice. In the endometrium, the SCC component is\nfrequently combined with more differentiated tumors like\nendometrioid adenocarcinomas and serous carcinomas, and\nmay also represent a component of malignant mixed\nMullerian tumor (MMMT) [ 3]. A review of the literature\nrevealed approximately 49 PD endometrial NETs (41 SCC,\nand 8 LCNEC), 22 of which were combined with endo-\nmetrioid adenocarcinoma [ 4–12], 2 with adenosquamous\ncarcinoma [ 6], 2 with serous carcinoma [ 13, 14], and 3\nwere a component of an MMMT [ 3]. But to the best of our\nknowledge, the concomitant existence of SCC, endome-\ntrioid adenocarcinoma, and serous carcinoma arising in the\nendometrium has not been previously reported in the\nEnglish literature.\nSCCs have a conventional histology of small-sized\ntumor cells with scant cytoplasm, hyperchromatic and\nfinely granular nuclei, inconspicuous nucleoli, crushing\nartifacts, and high mitotic activity. However, definite\ndiagnosis is based on the immunohistochemical (IHC)\ncharacterization of NE markers: chromogranin A, synap-\ntophysin, CD56, and neuron-specific enolase (NSE).\nWe report a case of a combined endometrial tumor\ncomposed of SCC, endometrioid adenocarcinoma, and\nserous carcinoma on the base of extensive adenomyosis.\nWe will discuss distinct entities in differential diagnosis\nwhich share similarities in both clinical presentation and\npathologic findings.\nS. Kiremitci /C1A. Sertcelik\nDepartment of Pathology,\nAnkara University School of Medicine, Ankara, Turkey\nK. Kahraman ( &) /C1F. Ortac\nDepartment of Obstetrics and Gynecology,\nAnkara University School of Medicine, 06100 Ankara, Turkey\ne-mail: korhankahraman@hotmail.com\n123\nInt Canc Conf J (2012) 1:27–32\nDOI 10.1007/s13691-011-0004-z\n\nCase presentation\nClinical presentation\nA 76-year-old, multiparous woman was referred to our\ngynecology department with the diagnosis of uterine\nMMMT from a curettage material. She had been meno-\npausal since the age of 49. Her medical history was sig-\nnificant for hypertension and coronary arterial disease.\nTumor markers were normal except the slightly elevated\nserum CA 125 value: 53.6 U/ml (reference range less than\n35 U/ml). A chest radiograph and computed tomography\nscan of the thorax and upper abdomen were negative for\nprimary or metastatic disease. A magnetic resonance\nimaging was not performed before the operation. An\nexploratory laparotomy was performed. The uterus was\nslightly enlarged and showed minimal irregularity on the\nserosal surface. Minimal serous ascites was observed. Both\novaries were atrophic. Other pelvic and abdominal struc-\ntures and peritoneal surfaces were all grossly normal. A\nstaging procedure was performed including a peritoneal\nwashing for cytological evaluation, total abdominal\nhysterectomy, bilateral salpingo-oophorectomy, bilateral\npelvic and para-aortic lymph node biopsy, infracolic\nomentectomy, and multiple peritoneal biopsies. She\nrecovered uneventfully and was discharged on the eighth\npostoperative day in a stable condition. However the\npathology report of the surgical specimen revealed a\ncombined endometrial tumor with an SCC component,\nrather than an MMMT. She received adjuvant chemother-\napy consisting of cisplatin and etoposide, and radiotherapy\nconsisting of a total dose of 2600 cGY (26 Gy) in 13\nfractions. At the fourth postoperative month, the patient\nwas well and disease-free in terms of uterine carcinoma.\nPathology examination\nGrossly, the uterus measured 5 9 5 9 3.2 cm with a\nslightly irregular serosal surface. When cut open, a broad-\nbased hemorrhagic and partially cystic polypoid lesion\nmeasuring 1 9 0.8 9 0.8 cm was identified located close\nto the uterine fundus. The cut surface of the polypoid lesion\nshowed a fleshy mass deeply invasive into the thickened\nmyometrium and serosa. The endometrium was diffusely\nhemorrhagic with smooth and irregular areas and measured\n13 mm in thickness. The cut surface of the left fallopian\ntube showed nodular fleshy masses in the paraovarian\nregion, whereas both ovaries were atrophic.\nMicroscopic examination revealed a malignant tumor\ncomposed of three distinct components (Fig. 1); the pol-\nypoid lesion corresponded to both diffusely infiltrative PD\nsmall cells and focal areas of well-formed malignant\nglands. The PD component constituted the large and deeply\ninvasive portion of the tumor and was composed of densely\npacked small- to intermediate-sized cells with hyperchro-\nmatic nuclei, inconspicuous nucleoli, and scant cytoplasms\nforming solid sheets and extensive geographic tumor\nnecrosis. Nuclear molding, crushing artifact, and high\nmitotic rate were evident. The endometrioid adenocarci-\nnoma component was composed of glandular and villo-\nglandular foci with no evidence of squamous metaplasia.\nThe third and minor component of the tumor was com-\nposed of cells with a high degree of cytological atypia and\nmicropapillary appearance consistent with serous carci-\nnoma (Fig. 1b). A striking finding was that the uterine wall\nrevealed extensive adenomyosis comprising tubular and\ncystic atrophic glands, and tumor components were inti-\nmately nested within the adenomyotic foci some of which\nexhibited malignant transformation of atrophic epithelium\nto malignant epithelium in the same gland (Fig. 2). Lym-\nphovascular invasion was prominent. The right ovary and\nleft tube were infiltrated by the PD component of the\ntumor. There were only microscopic tumor foci in the\nformer, whereas the wall of the tube was diffusely infil-\ntrated reaching up to the subepithelial region. The tumor\npossessed no sarcomatous or heterologous component upon\nthorough examination. Lymph node involvement was\nconfined to only one (1/6) left pelvic lymph node and the\nmetastasis was in serous carcinoma phenotype. The\nomentum was free of tumor. The cytology of the peritoneal\nfluid was negative for malignant cells.\nA wide IHC panel was applied to identify the nature of\nthe tumor components, particularly the PD one. Antibodies\nand the staining properties of the tumor components are\nsummarized in Table 1. IHC results demonstrated a dif-\nferent pattern of protein expression in the two components\n(Fig. 3); the PD component was diffusely positive with\nsynaptophysin, CD56, and vimentin, and focally with\nchromogranin A, whereas the adenocarcinoma component\nwas positive with only PANCK and vimentin. CD10 was\nexpressed only in endometrial stromal cells in residual\nadenomyosis foci. Ttf-1, caldesmon, SMA, and desmin\nwere all negative. Hematoxylin and eosin (H&E) slides of\nthe previous endometrial curettage obtained from the pre-\nvious institute showed intermingled serous carcinoma and\nundifferentiated areas with no sarcomatous foci. We real-\nized that the undifferentiated component was misinter-\npreted as sarcomatous differentiation and thereby a\ndiagnosis of MMMT was made without the assistance of\nIHC in the previous investigation.\nStrong and consistent expression of NE markers (chro-\nmogranin A, synaptophysin, and CD56) and the cytological\nfeatures proved the PD component to be a SCC, whereas\nother components were consistent with endometrioid ade-\nnocarcinoma and serous carcinoma on the basis of their\ncharacteristic histological appearance. A final diagnosis\n28 Int Canc Conf J (2012) 1:27–32\n123\n\ncompatible with a combined tumor of SCC, endometrioid\nadenocarcinoma, and serous carcinoma of 2009 FIGO\n(Federation Internationale de Gynecologie et d’Obstetri-\nque) stage IIIC was achieved.\nDiscussion\nSurgical material of the present case revealed an aggressive\ntumor with three distinct components: SCC, endometrioid\nadenocarcinoma, and serous carcinoma. The predominant\ncomponent, SCC, fulfilled the three diagnostic criteria\nproposed by Van Hoeven et al. [ 5]: no evidence of primary\norigin from the endometrium, characteristic histology with\nsmall- to intermediate-sized tumor cells growing in a\ndense, sheet-like fashion, and immune reactivity for the NE\nmarkers synaptophysin, CD56, and chromogranin A. As in\nour case, focal chromogranin A positivity was reported\npreviously [10, 12]. SCC was the aggressive component as\nwas evident by its full-thickness myometrial invasion and\ndiffuse endometrial stromal invasion leaving residual\nendometrial glands with or without atypia. Metastatic\nFig. 1 Histologically tumor showed three different components:\na PD component ( green arrow) and well-differentiated endometrioid\nadenocarcinoma component ( yellow arrow) were admixed and grew\nup on the basis of adenomyosis ( red arrows). b The third component\nwas viewed in limited areas with small micropapillary morphology\nand significant cytologic atypia. c PD component was highly cellular\ncomposed of small cells with high nucleocytoplasmic ratio and high\nmitotic activity, and showed extensive geographic tumor necrosis (d).\nH&E 940, 9100, 9400, 9100, respectively\nFig. 2 PD component diffusely infiltrated the uterine wall, and the\nremaining endometrial glands, both in the atrophic endometrium and\nadenomyosis foci, revealed various histomorphology including\natrophic cystic ( a), tubular-proliferative ( b), and atypical ( c) glands\n(transition to malignant epithelium in an atrophic gland; arrow),\nintermingled with the three components of tumor. H&E 9200\nInt Canc Conf J (2012) 1:27–32 29\n123\n\ntumor deposits were almost entirely composed of SCC,\nexcept that lymph node involvement was in serous carci-\nnoma morphology. The present case favors the notion that\nSCC and serous carcinoma are the components with\naggressive behavior in combined tumors of the endome-\ntrium [5, 13].\nDifferential diagnosis of endometrial SCC would com-\nprise various types of tumors derived from primary endo-\nmetrium disease or metastasis from another site.\nCarcinosarcoma (aka MMMT), Mullerian adenosarcoma,\nendometrial undifferentiated carcinoma, endometrial stro-\nmal sarcoma, lymphoma, Ewing’s sarcoma/peripheral\nprimitive neuroectodermal tumor (ES/PNET), and com-\nposite tumors are distinct entities that may mimic each\nother in terms of clinical presentation and gross and\nmicroscopic appearances. Diverse endometrial tumors,\nparticularly MMMT and endometrial stromal sarcoma,\ncommonly present with bulky, friable polypoid masses that\ninvolve the uterine wall deeply, fill the uterine cavity, and\nprotrude through the external os, as a common finding\nshared with SCC. MMMT may occur in elderly postmen-\nopausal women with extrauterine disease at initial presen-\ntation, like SCC.\nAs an initial approach, the possibility of metastatic\ncarcinoma of the endometrium should be excluded by\nusing detailed clinical information and radiology findings.\nSCCs exhibit the same histology independently from the\nprimary site. The lung is the most common site and it is not\nFig. 3 PD component exhibited a discriminating immunoreactivity;\nPANCK highlighted both the adenocarcinoma component and non-\nneoplastic endometrial glands remaining in the adenomyosis foci\ninfiltrated diffusely by PD component ( a). CD10 showed up the\nendometrial stromal cells surrounding the non-neoplastic endometrial\nglands as evidence of adenomyosis ( b). Synaptophysin ( c), CD56\n(d) (diffusely, with strong intensity), and chromogranin A (e) (focally,\nwith moderate intensity) were expressed in the PD component,\nwhereas the adenocarcinoma components were entirely negative for\nthese markers. Immuno peroxidase; 940, 9100, 940, 9100, 9200,\nrespectively\nTable 1 Immunohistochemical characteristics of tumor components\nPrimary antibody Clone Dilution Source Cellular localization Tumor components\nPD Adenocarcinoma\nPANCK SD3 ? LP34 1/400 Neomarkers Cytoplasmic Negative Diffuse strong staining\nVimentin V9 1/200 Neomarkers Cytoplasmic Diffuse strong staining Diffuse strong staining\nSynaptophysin Polyclonal 1/50 Invitrogen Cytoplasmic, membranous Diffuse strong staining Negative\nChromogranin A Cocktail 1/1000 Neomarkers Cytoplasmic, membranous Focal moderate staining Negative\nCD56 S6CO4 1/100 Neomarkers Cytoplasmic, membranous Diffuse strong staining Negative\nTTF-1 8G763/1 1/250 Cell Marque Nuclear Negative Negative\nDesmin DE-R-11 1/150 Novocastra Cytoplasmic Negative Negative\nCaldesmon H-Cald 1/400 Neomarkers Cytoplasmic Negative Negative\nSMA 1A4 1/500 Dako Cytoplasmic Negative Negative\n30 Int Canc Conf J (2012) 1:27–32\n123\n\nsurprising for a primary lung SCC to accompany an ade-\nnocarcinoma. However, IHC evidence of Ttf1 expression\nwould reliably indicate pulmonary origin [15]. Endometrial\nSCCs also differ from pulmonary SCCs by diffuse\nexpression of vimentin [ 8, 16], with the exception of a\nunique report exhibiting vimentin negativity in a primary\nendometrial SCC [ 12]. Cytokeratin expression is not dif-\nferentiating, and is variable in SCCs. Endometrial carci-\nnomas should also be investigated for an ovarian primary\ndisease.\nIn the present case, the initial pathologic diagnosis in\nendometrial curettage was interpreted as MMMT, at an\noutside center. The SCC component of a combined tumor\nmay be misdiagnosed as undifferentiated sarcoma, leading\na diagnostic dilemma with MMMT. MMMTs have a\nbiphasic appearance both with carcinomatous and sarcoma-\nlike elements. The former may exhibit any type of endo-\nmetrial carcinoma commonly including endometrioid and\nserous carcinoma, whereas the latter may reveal homolo-\ngous (fibrosarcoma or leiomyosarcoma) or heterologous\n(chondrosarcoma, osteosarcoma, or rhabdomyosarcoma)\nelements. Detailed IHC examination with desmin, SMA,\nmyoglobin, and S100 antibodies would differentiate the\nsarcomatous component of MMMTs, in general. Differ-\nentiation of SCC from sarcoma definitely requires IHC\nevidence of NE markers. NE differentiation in MMMTs is\nparticularly seen as focal expression which is distinct from\nthe diffuse expression observed in SCCs [ 17].\nMullerian adenosarcoma, which defines the benign\nglandular component admixed with malignant mesenchy-\nmal component, should also be kept in mind in the dif-\nferential diagnosis. In the present case, the SCC component\ninvaded the endometrium leaving residue non-neoplastic\nand atrophic glands, and even in adenomyosis areas,\nleading to a misdiagnosis of an adenosarcoma. A malignant\nepithelial component of a combined tumor may be over-\nlooked especially in curettage materials and also inade-\nquate sampling of the tumor may lead to a misdiagnosis of\nMMMT or a combined tumor as adenosarcoma.\nEndometrial undifferentiated carcinoma, an aggressive\ntumor commonly presenting with an advanced stage, is a\ncontroversial entity that fails to show evidence of glandular\nor squamous differentiation according to the World Health\nOrganization definition [ 18]. It is commonly associated\nwith an endometrioid adenocarcinoma and, thus, may be\nmisdiagnosed as a high-grade endometrioid adenocarci-\nnoma component. However, diagnosis of high-grade\nendometrioid adenocarcinoma requires the definite pres-\nence of glands which is not an expected finding for\nendometrial undifferentiated carcinoma [ 19]. The NE\ncomponent of a combined tumor, particularly LCNEC,\nmay be misinterpreted as an endometrial undifferentiated\ncarcinoma because of its similar cytological features.\nNevertheless, IHC evidence of diffuse NE differentiation\nwould distinguish SCC from the latter which may also\nexhibit NE differentiation, but in a limited number of\ntumor cells, approximately 10% [ 20].\nEndometrial stromal sarcomas are indolent tumors\ncomposed of uniform, small, oval cells with scanty cyto-\nplasm and may metastasize even many years after initial\ndiagnosis. Their differentiation from endometrial combined\ntumors is relatively straightforward on the basis of absence\nof significant cytological atypia and resemblance to normal\nproliferative phase endometrium with the characteristic\nfeature of regularly distributed spiral arterioles. Further-\nmore, endometrial stromal sarcomas commonly occur in\npremenopausal women in contrast to the aforementioned\ntumors. CD10 has been considered as a specific marker for\nboth neoplastic and non-neoplastic endometrial stromal\ncells [21], and probably distinguishes endometrial stromal\nneoplasms from others. However, there has been a report\nstressing CD10 positivity both in MMMT and Mullerian\nadenosarcoma, and indicating that CD10 is a characteristic\nof Mullerian system-derived neoplastic mesenchymal cells\n[22].\nLymphoma and ES/PNET are practically distinguished\nby IHC analysis of LCA antibody for the former, and CD99\nfor the latter.\nIn conclusion, it has long been known that diverse\nendometrial tumors with different prognosis and thera-\npeutic modalities may reveal similar histopathologic find-\nings. Although characteristic histological features may\nusually lead to a correct diagnosis, the NE component may\nbe misdiagnosed as undifferentiated carcinoma or sarcoma,\nparticularly in the absence of IHC analysis. Therefore, it is\nadvisable to carry out IHC in the histopathology exami-\nnation of the combined tumors to avoid misinterpretations.\nConflict of interest The authors declare that they have no conflict\nof interest.\nReferences\n1. Barakat MT, Meeran K, Bloom SR (2004) Neuroendocrine\ntumors. Endocr Relat Cancer 11:1–18\n2. Eichhorn JH, Young RH (2001) Neuroendocrine tumors of the\ngenital tract. Am J Clin Pathol 115(Suppl 1):94–112\n3. Manivel C, Wick MR, Sibley RK (1985) Neuroendocrine dif-\nferentiation in mullerian neoplasms. An immunohistochemical\nstudy of a ‘ ‘pure’’ endometrial small-cell carcinoma and a mixed\nmullerian tumor containing small-cell carcinoma. Am J Clin\nPathol 85:438–443\n4. Huntsman DG, Clement PB, Gilks CB et al (1994) Small cell\ncarcinoma of the endometrium. A clinicopathological study of\nsixteen cases. Am J Surg Pathol 18:364–375\n5. Van Hoeven KH, Hudock JA, Woodruff JM et al (1995) Small\ncell neuroendocrine carcinoma of the endometrium. Int J Gynecol\nPathol 14:21–29\nInt Canc Conf J (2012) 1:27–32 31\n123\n\n6. Katahira A, Akahira J, Nikura H et al (2004) Small cell carci-\nnoma of the endometrium: report of three cases and literature\nreview. Int J Gynecol Cancer 14:1018–1023\n7. Rajab KE, Sandhu AK, Malik A et al (2005) Small cell neuro-\nendocrine carcinoma of the endometrium, a rare aggressive\ntumor. Saudi Med J 26:1130–1132\n8. Melgoza F, Brewster WR, Wilczynski S et al (2006) p16-Positive\nsmall cell neuroendocrine carcinoma of the endometrium. Int J\nGynecol Pathol 25:252–256\n9. Mulvany NJ, Allen DG (2007) Combined large cell neuroendo-\ncrine and endometrioid carcinoma of the endometrium. Int J\nGynecol Pathol 27:49–57\n10. Bige O, Saatli B, Secil M et al (2008) Small cell neuroendocrine\ncarcinoma of the endometrium and laparoscopic staging: a clin-\nicopathologic study of a case and a brief review of the literature.\nInt J Gynecol Pathol 18:838–843\n11. Albores-Saavedra J, Martinez-Benitez B, Luevano E (2008)\nSmall cell carcinomas and large cell neuroendocrine carcinomas\nof the endometrium and cervix: polypoid tumors and those arising\nin polyps may have a favorable prognosis. Int J Gynecol Pathol\n27:333–339\n12. Hwang JH, Lee JK, Lee NW et al (2010) Primary small cell\ncarcinoma of the endometrium: report of a case with immuno-\nhistochemical studies. J Reprod Med 55:81–86\n13. Shaco-Levy R, Manor E, Piura B et al (2004) An unusual com-\nposite endometrial tumor combining papillary serous carcinoma\nand small cell carcinoma. Am J Surg Pathol 28:1103–1106\n14. Posligua L, Malpica A, Liu J et al (2008) Combined large cell\nneuroendocrine carcinoma and papillary serous carcinoma of the\nendometrium with pagetoid spread. Arch Pathol Lab Med\n132:1821–1824\n15. Travis WB, Nicholson S, Hirsch FR et al (2004) Small cell car-\ncinoma. In: Travis W, Brambilla E, Muller-Hermelink HK et al\n(eds) Pathology and genetics. Tumors of the lung, pleura, thymus\nand heart. World Health Organization classification of tumors.\nIARC, Lyon, pp 31–34\n16. Upton MP, Hirohashi S, Tome Y et al (1986) Expression of\nvimentin in surgically resected adenocarcinomas and large cell\ncarcinomas of lung. Am J Surg Pathol 10:560–567\n17. George E, Manivel JC, Dehner LP et al (1991) Malignant mixed\nmullerian tumors: an immunohistochemical study of 47 cases,\nwith histogenetic considerations and clinical correlation. Hum\nPathol 22:215–223\n18. Silverberg SG, Kurman RJ, Nogales F et al (2003) Epithelial\ntumors and related lesions. In: Tavassoli FA, Devilee P (eds)\nTumors of the breast and female genital organs: World Health\nOrganization classification of tumors. IARC, Lyon, p 227\n19. Altrabulsi B, Malpica A, Deavers MT et al (2005) Undiffer-\nentiated carcinoma of the endometrium. Am J Surg Pathol\n2:1316–1321\n20. Silva EG, Deavers MT, Malpica A (2007) Undifferentiated car-\ncinoma of the endometrium: a review. Pathology 39:134–138\n21. Toki T, Shimizu M, Takagi Y et al (2002) CD10 is a marker for\nnormal and neoplastic endometrial stromal cells. Int J Gynecol\nPathol 21:41–47\n22. Mikami Y, Hata S, Kiyokawa T et al (2002) Expression of CD10\nin malignant mullerian mixed tumors and adenosarcomas: an\nimmunohistochemical study. Mod Pathol 15:923–930\n32 Int Canc Conf J (2012) 1:27–32\n123","source_license":"CC0","license_restricted":false}