{"paper_id":"5f24d34f-5309-405f-9df1-c41ff5da4dd5","body_text":"Endometriosis is\na frequent benign disorder. Several observations of the coexistence of endometriosis\nand cancer have been published [ 1 ,  2 ]. Malignancy arising in extraovarian endometriosis\nis a rare event [ 1 ]. Here, we report a case of clear cell adenocarcinoma \nderived from\npathologically confirmed endometriosis in the abdominal wall. We discuss the\nepidemiological and clinicopathological features of malignancy arising in\nabdominal wall endometriosis.\n\nA 49 year-old woman,\ngravida 3, para 0, underwent a myomectomy, 20 years before, for uterus\nleiomyoma through a midline incision. She is presented with a painful large abdominal wall\nmass. On physical examination, a firm indurated mass was palpated in the lower\nabdominal wall. Abdominal and pelvic ultrasounds followed by computed tomography showed a\nheterogeneous intramuscular mass of 8.5 cm diameter without local extension ( Figure \n1 ). The clinical impression was of desmoid tumor and the patient taken to\nsurgery and the tumor was resected without rupture. The surgeon did not perform\nany peritoneal washing or biopsies because of the absence of widespread tumor\nin the peritoneal cavity. The surgical specimen consisted of 11 cm cutaneous and\nmuscular tissues, occupied by ill-defined white tumor, which contained cystic\ncavities and abundant foci of necrosis. Surgical margins were positive. \nMicroscopically, the tumor showed a predominant papillary and tubulocystic\ngrowth pattern ( Figure 2 ). The tumor cells were round or polygonal most with\nhobnail configuration ( Figure 3 ). The cytoplasm was clear, and the nuclei were\nround with prominent nucleoli. Cellular atypia was moderate, and mitosis was rare. Benign\nendometriotic foci were observed in the proximity of the tumor ( Figure 4 ). Immunohistochemically,\ntumor cells showed diffuse and strong cytoplasmic positivity with vimentin,\nepithelial membrane antigen, and cytokeratin 7, but no staining for cytokeratin\n20 progesterone and estrogen receptor. Calretinin and mesothelin were negative. \nPelvic ultrasound and computed tomography identified normal-sized ovaries and\nuterus. Endometrial curettage was negative for malignancy.\nThe diagnosis of\nclear cell adenocarcinoma arising from abdominal endometriosis foci was\nretained. Adjuvant chemotherapy was indicated, but the patient was lost to\nfollow up. She returned six months after. At that time, ultrasound and computed\ntomography showed a recurrent mass at the abdominal wall with extension to the\nbladder. Three cycles of combination chemotherapy with cyclophosphamide and cisplatin\nwere given, but the tumor did not regress. She underwent surgery again with a\nresection of 5 cm encapsulated nodule. At that time, uterus, ovaries, and tubes\ndid not show any abnormalities. Histology demonstrated the same type of tumor. \nMargins were free of tumor. Three cycles of chemotherapy were also given but\nfailed to control the disease; the chemotherapy she got is not precised. The\ncomputed tomography showed again a recurrent mass with extension to the bladder\nand pelvic bone, and adjuvant radiotherapy was indicated.\n\nEndometriosis,\ndefined as the presence of endometrial-like tissue outside the uterine cavity,\nis usually located in the ovaries and pelvic peritoneum [ 2 ,  3 ]. Parietal\nendometriosis is very rare and constitutes 1 to 2% of endometriosis cases [ 1 ]. \nIt arises usually in a surgical scar of cesarean section or hysterectomy, and\nless frequently in a surgical scar of hernia or of appendicectomy [ 1 ,  4 ]. Cases\nof endometriosis without scar have been described [ 1 ]. The incidence of \nabdominal wall endometriomas is of 0.04% among\nparturients undergoing cesarean section and it is more frequent than\nendometriosis following conventional gynaecologic surgery [ 4 ,  5 ]. In our case,\nabdominal wall endometriosis occurred in a surgical midline scar of myomectomy. \nThe etiopathogenetic mechanism is more likely related to iatrogenic\ntransplantation of endometrium during gynecological surgery rather than hematogenous\ndissemination or metaplasia [ 4 ,  6 ]. Clinical diagnosis remains difficult, and many\npatients are asymptomatic [ 4 ]. Symptoms related to pelvic endometriosis are\nnoted in 26% of cases. Ultrasound can show a cystic lesion in many cases [ 5 ]. \nWomen with pelvic endometriosis have a higher frequency of malignancy, but\nmalignant change in extrapelvic endometriosis is a rare event [ 1 ,  2 ,  4 ,  7 ]. \nTwenty percent of malignancy in endometriosis occurs in extragonadal site [ 1 ,  8 ,  9 ]. There is extensive clinicopathological, molecular, and genetic evidence\nsupporting the hypothesis that endometriosis is a neoplastic process with a\npotential for malignant transformation [ 3 ]. The natural course of malignant\ntransformation of endometriosis is long and can be explained by estrogenic\nstimulation [ 8 ,  10 – 12 ]. Malignant\ntransformation\nin endometriosis was first described by Sympson in 1925 in\n[ 9 ], who\nproposed three criteria for diagnosis: demonstration of a clear example of the\nendometriosis in proximity to the tumor, no other primary site for the tumor,\nand histologic appearance consistent with an origin from endometriosis. Scott\nin [ 1 ,  8 ] recommended the presence of transitional area between endometriosis and\ncancer. Atypical endometriosis, a term first coined by LaGrenade and Silvergerg\nin 1988 in [ 8 ,  13 ], is rare and is characterized by endometriotic glands with cytological\nand/or architectural atypia (hyperchromatic or pale nuclei with moderate to marked\npleomorphism increased nuclear to cytoplasmic ratio, cellular crowding, stratification,\nor tufting). The rate of atypical endometriosis ranges from 1.7 to 3.6% in ovarian endometriosis\n[ 13 ]. Fukunaga et al. [ 13 ] demonstrated that atypical endometriosis in ovary is\noften associated with epithelial neoplasm and showed direct transition from\natypical epithelium to malignant tumor. In our case, the criteria of Sympson\nwere fully satisfied. The demonstration of endometriosis might require the\nexamination of multiple levels and sections, that is why preoperative biopsy cannot\nmake the diagnosis of malignancy arising in endometriosis [ 6 ]. Tumors that can\narise in endometriosis include in decreasing order: endometrio id carcinoma\n(75.9–69.1%), sarcoma\n(25–11.6%), clear\ncell carcinoma (13.5–4.5%), and\nmucinous or serous carcinoma (4.6%–1%) [ 4 ]. In\nextrapelvic localization, clear cell carcinoma is the most common histological\nsubtype, followed by endometrioid carcinoma [ 1 ]. Due to the rarity of malignant\ntransformation of endometriosis at extragonadal sites, it is difficult to\nestablish a treatment protocol. First-line treatment is surgery, removing as\nmuch endometriosis as possible, staging at this point is also necessary. \nSecond-line treatment, with chemotherapy, radiotherapy, and even hormonotherapy\nmay be needed. Prognosis is variable from 10 to 100% five-year survival,\ndepending on histological type and localization of the disease [ 7 ].\n\nCutaneous\nlocalization of endometriosis is unusual and appears most frequently in\nsurgical scars from obstetric or gynecological interventions. It is important\nto recognize the possibility of tumors arising from endometriosis when the\npathologist is confronted to an extraovarian tumor with endometrial appearance. \nExamination of multiple sections is required to demonstrate endometriosis foci.","source_license":"CC0","license_restricted":false}