An exceptional case of degenerative endometriosis of the uterine torus

The case involved a 44‐year‐old female patient (BMI: 32) with a long history of severe pelvic endometriosis, which had led to repeated surgeries over the previous 10 years, including a total right and partial left salpingectomy, a partial right oophorectomy, and a right colectomy. The MRI features were consistent with recurrence of previously reported endometriosis of the uterine torus, associated with external adenomyosis and a left hydrosalpinx (Figure  1A,C ). Following this, primary infertility was treated by IVF, followed by oocyte donation, which led to the birth of a child by cesarean section due to placenta previa 3 years prior to the current treatment. Two years after delivery, the patient was referred for recurrence of pelvic pain related to a left tubo‐ovarian complex measuring 52 × 33 × 30 mm, compressive on the left ureter and adherent to the sigmoid wall without any sign of invasion. Systematic screening for cervico‐uterine cytological abnormalities and human papillomavirus testing were negative. Treatment with GnRH agonist for 3 months was initiated. Clinical reevaluation (improvement) and an MRI performed remotely from the GnRH agonist treatment revealed the torus lesion with persistent non‐transmural recto‐sigmoidal retraction. Twenty months later, in the context of recurrent pelvic pain associated with a digestive compressive symptomatology, a new MRI identified an infiltration of the recto‐sigmoid junction (Figure  1B,D ) motivating additional investigations by rectal endoscopy. This revealed a heterogeneous tissular lesion, with transmural invasion of the digestive wall at the recto‐sigmoid junction, 20 cm from the anal margin, as well as suspicious mesorectal adenopathies. During colonoscopy, the hemorrhagic lesion measured 30 × 30 mm, and biopsies were performed. Histological analysis revealed a moderately differentiated adenocarcinoma, rather in favor of a gynecological origin (CK7+/CK20‐/P16+/PAX6+ profile), cervico‐uterine rather than endometrial, given the age of the woman and the negative hormone receptors. This profile is not in favor of a primary colorectal origin. There is a conservation of the four MMR proteins by the tumor cells (MMR proficient status). The PET‐CT scan of the tumor's extension reveals pelvic and lumbo‐aortic lymph node metastasis with no distant visceral or bone metastasis. The staging chosen according to the suspected etiologies is either FIGO stage IVA or IIIC2 if a cervical or uterine origin is considered, or IIIC if a primary ovarian epithelial etiology is considered. MRI images in T2 sequence. (A, C) Axial and sagittal sections 2 years before treatment showing the endometriosis lesion. (B, D) Axial and sagittal sections at the time of treatment with the malignant lesion infiltrating the digestive tract and uterus. After discussion with the national committee for rare gynecological tumors, it was proposed to start chemotherapy with carboplatin AUC5 + paclitaxel 175 mg/m 2 D1 = D21 for three to four cycles, then reassess. At the end of the four cycles, the MRI reassessment showed a 30% progression (recist 1.1 criteria compared to the previous MRI) of the lesion, measuring 82 mm in height, 72 mm in anteroposterior axis, and 78 mm in transverse axis, with the development of a central necrotic portion. This lesion, centered around the uterine torus, invaded the cervix, the left ovary, the right broad ligament, and the anterior wall of the rectosigmoid junction over a height of 37 mm. No endocervical or endometrial abnormalities were found. There was evidence of bilateral external iliac, right common iliac, and left subrenal adenopathy. Hemoglobin was 10.4 g/dL, renal function was normal, and markers showed CA 125 at 500 IU/L, CA 19‐9 at 108 IU/L and ACE at 0.8 IU/L. A repeated PET‐CT scan showed a partial response to therapy in the lymph nodes, stability of the pelvic mass, partial regression of the infiltration of the recto‐sigmoid junction, and no other suspicious hypermetabolic lesions. It was decided in a surgical committee to perform a median laparotomy. The per‐operative assessment revealed a distal right lateral parametrial invasion encompassing the right pelvic ureter, and a distal left caudal parametrial invasion in the prolongation of the recto‐sigmoid invasion (Figure  2 ) that had already been described. The surgical procedure consists of a monobloc resection of the uterus, the recto‐sigmoid hinge, the ureter, and the right distal parametrium, the left caudal parametrium type C2 (Querleu Morrow's classification 9 ), and the other parametrium type C1, pelvic and lumbo‐aortic curage, right ureteral reimplantation with a JJ probe, a middle colorectal anastomosis protected by an upstream colostomy, and an epiplasty on the left gastroepiploic artery. The postoperative follow‐up was marked by a complication of class IIIb of the Clavien Dindo classification 10 : left ureteral fistula treated by endoscopic placement of a JJ catheter under general anesthesia. The anatomopathological analysis of the surgical specimen showed a histological and immuno‐histochemical aspect of high‐grade clear cell carcinoma developed from endometriosis lesions of the uterine torus (Figure  3A,B ). The endometrial and endo‐cervical mucosa were normal. The lesion extends posteriorly to the bowel mucosa with focal ulceration of the latter (Figure  3C,D ), laterally to the parametrium on both sides (with a metastatic node on the left) and to the right ovary. All the excisional borders were healthy, there were numerous neoplastic vascular emboli as well as pelvic and lumbo‐aortic lymph node involvement (9 metastatic nodes out of 35 taken). The 2017 UICC pTNM staging is therefore pT4N2 R0, the 2014 FIGO stage IVA. The progressive iconographic features, the centro‐pelvic sus‐peritoneal location with posterior development on endometriosis lesions, the absence of other cancerous lesions as well as the histological type therefore suggest in priority a malignant degeneration of deep endometriosis. After discussion at the national committee for rare gynecological tumors, a reinstatement of chemotherapy with carboplatin AUC5 + paclitaxel 175 mg/m 2 D1 = D21 with bevacizumab 15 mg/kg D1 = D21 for 15 months was proposed. To date, the patient has not had any recurrence of her disease. Intraoperative macroscopic view of the posterior pelvic exenteration specimen (left, uterus and cervix; right, rectum with a clamp and sigmoid; middle, malignant endometriosic lesion). Pathology images. (A) Macroscopic photograph of the posterior pelvic exenteration specimen (median sagittal section). (B) Endometriosis of the torus (arrows) close to the tumor area (star); HES staining. (C) Extrinsic tumor infiltration of the digestive wall with tumor (arrow) ulcerating the digestive mucosa (star); HES staining. (D) Histological details of clear cell carcinoma; HES staining.

Philippe Merviel: Conceptualization; data curation; investigation; supervision; validation; writing – original draft; writing – review and editing. Christie Rebahi: Investigation; validation. Pandora James: Investigation; validation. Isabelle Thomas‐Kergastel: Investigation; validation. Fanny Bourhis‐Guizien: Investigation; validation. Virginie Conan‐Charlet: Investigation; validation. Fanny Derquin: Investigation; validation. Anne‐Claire Hardy‐Bessard: Investigation; validation. Pierre‐François Dupre: Investigation; validation. Karine Morcel: Data curation; investigation; supervision; validation; writing – original draft.

Written informed consent was obtained from the patient to publish this report in accordance with the journal's patient consent policy. French legislation on studies of routine medical care (Law no. 78‐17 of January 6, 1978 amended in 2004 relating to data processing, files, and freedoms): For all research involving human participants, informed consent to participate in the study should be obtained from participants (or their parent or legal guardian in the case of children under 16).

None.

External endometriosis is defined by the presence of endometrial tissue outside the uterus. 1 Malignant transformation of endometriosis occurs in 0.7%–1.6% of cases (about 1%), 2 , 3 and in 78.7% of cases it was an ovarian localization of endometriosis that evolved into ovarian cancer. 4 , 5 This association of endometriosis and ovarian cancer has existed since the first publication by Sampson in 1927. 11 More recently, the tubal origin of ovarian carcinogenesis has been suggested, 12 which could be responsible for the two types of cancer associated with endometriosis, endometrioid carcinomas and clear cell adenocarcinomas (grouped under the term EAOC: endometriosis associated ovarian cancer). 12 Saavalainen's study, 13 based on the Finnish national registry between 1987 and 2012 of 23,210 ovarian locations of endometriosis, 20,187 peritoneal, and 2372 cases of deep infiltration, shows that there is an increased risk of ovarian cancer (standardized incidence ratio—SIR: 1.76 [1.47–2.08]), associated with an ovarian location of endometriosis. However, in this study, the risk of ovarian cancer is not increased in case of peritoneal or DIE. Ferrari et al. 14 reported a case of clear cell carcinoma in a parietal endometriosis lesion (post‐cesarean section and transverse laparotomy), while Chen et al. 15 reported two cases of high‐grade serous cancer (HGSC) in an endometriosis of the utero‐sacral ligament, followed by Johnson et al. 16 and Song et al., 17 each of whom reported a case of HGSC from in a rectal endometriosis. If we add the cases of cervical and rectal degeneration associated with Lynch syndrome, 18 and vaginal degeneration associated with endometrial cancer, 19 , 20 there are less than 10 cases of degeneration described in the literature (Table  1 ). Cases of malignant degeneration on deep endometriosis lesions reported in the literature since 2011. 39 52 Uterosacral lig Uterosacral lig Pelvic pain Vaginal bleeding HGSC HGSC Abbreviations: CCC, clear cell carcinoma; EC, endometrioid adenocarcinoma; HGSC, high‐grade serous carcinoma. Today, Sampson's menstrual reflux theory published in 1925 21 can no longer explain the biological variability (aromatase activity, progesterone resistance) and the clonal aspect of endometriosis lesions. 22 , 23 The current hypothesis is that endometriosis develops from primitive endometrial cells present outside the uterus as early as the period of organogenesis. After puberty, these cells, due to immune alterations, endocrine disruptors, and a pro‐inflammatory peritoneal environment related to menstrual reflux (oxygen free radicals, microbiota), would evolve toward endometriotic implants. 24 Endometriotic cells acquire properties of adhesion, proliferation (by local synthesis of estrogens, increase in estrogen receptors and partial resistance to progesterone by mutation/methylation of its receptors), and invasion of surrounding tissues (deregulation of certain miRNAs). An inflammatory reaction, neovascularization, and neo‐neurogenesis develop around the endometriosis lesions, allowing them to become self‐sustaining and independent of the woman's hormonal regulations. The diagnosis of endometriosis degeneration requires the three criteria of Sampson, 11 i.e., (i) the presence of endometriosis and cancer in the same site, (ii) the negative search for another tumor location, and (iii) a histological type compatible with an endometriotic origin (glands and stroma). In 1953, Scott 25 added a fourth criterion: the morphological continuity between the benign and malignant epithelium within the endometriosis. 26 Twenty percent of degenerative cases of endometriosis are extra‐ovarian (recto‐vaginal septum, colon, and vagina in more than 50% of cases 27 ); however, few cases in the literature are reported (Table  1 ). In these cases, the women are 10–20 years younger than in ovarian localization 28 and the histological types are endometrioid carcinoma (69.1%), sarcoma (25%), and in 4.5% of cases clear cell carcinoma. 12 Conversely, for ovarian degeneration, the meta‐analysis by Pearce 29 shows an OR of 3.05 (95% CI: 2.43–3.84) for clear cell carcinoma. We studied the risk factors for endometriosis degeneration. Our patient was 44 years old, and women with endometriosis who are older than 50 years have a higher risk of EAOC than women of the same age without endometriosis (HR: 9.63; 95% CI: 3.27–28.37) and women with endometriosis under 30 years old (HR: 4.97; 95% CI: 1.03–24.09). 30 The duration of endometriosis development (average 8 years) 3 and the association with infertility also exposes the woman to an increased risk of endometriosis degeneration: 12 years and four IVF in our patient's case, 20 years for Chen. 15 In a registry of 20,686 women with endometriosis, we find a threshold of 10 years with an SIR: 4.2 (95% CI: 2–7.7). 31 This woman had a total right and partial left salpingectomy, which interrupted the menstrual reflux that may be responsible for the stimulation of endometriosis lesions. In the context of prevention of EAOC, the hypothesis of incessant menstruation has led some authors to propose tubal ligaturea as primary prevention of endometriosis degeneration. 32 , 33 This woman was treated with in vitro fertilization (six cycles) and oocyte donation. Hyperoestrogenism is associated with the degeneration of endometriosis, particularly ovarian endometriosis, and the environment of endometriosis favors excess estrogen. Perimenopause, obesity, and ovarian stimulation are factors involved in the degeneration of DIE. 34 It is likely that the difference in estrogen concentration between the ovary (high concentrations) and DIE explains the difference in the frequency of malignant degeneration of lesions. 35 , 36 Similarly ERβ (estrogen receptor β) and CYPIBI (cytochrome P450 family I subfamily B member I), an estrogen agonist, are also overexpressed in the ovary compared to deep infiltrative endometriosis. 37 Finally, cell division decreases with female age, while deep infiltrating lesions increase. 38 , 39 In contrast to the positive effects of macroprogestins in endometrial cancer, they seem to have no effect on the degeneration of deep infiltrating lesions of endometriosis. 4 Can we assess the risk of degeneration of endometriosis lesions and make an earlier diagnosis? There are no preoperative markers that can indicate degeneration of endometriosis. For extra‐ovarian endometriosis, the risk of degeneration is close to zero. 13 , 36 The discovery of somatic mutations 40 in DIE lesions (79% of cases), 26% of which concern genes involved in cancerous degeneration (KRAS—Kirsten rat sarcoma, PIK3CA—phosphatifylinositol‐4,5 biphosphate 3 kinase catalytic subunit alpha, PTEN, ARID 1A—adenine‐thymine‐rich interactive domain‐containing protein 1A). The ARID 1A gene mutation is thought to link endometriosis to EAOC. 41 , 42 It is a tumor suppressor and chromatin remodeler gene, which would lead to cell cycle perturbations. 43 These mutations could help in the future to discriminate lesions at risk or not (however, there was no mutation in both cases in Chen's study 15 ). Despite multiple interventions and complementary examinations (ultrasound, MRI) in our case, the degeneration of endometriosis lesions was not recognized. CA125 can be used but it is increased in benign endometriosis. 44 Only on biopsies of the intra‐luminal lesion of the recto‐sigmoid hinge could the gynecological origin be established, especially by CK7 and CK20 immuno‐histochemical staining, which can distinguish the rectal origin of the cancer (CK20+/CK7−) from the gynecological origin (CK7+/CK20−). 45 However our team alerts on the degenerative risk in case of recurrence, persistence, and growth of deep endometriosis lesions in a woman over 40 years, especially after an ovarian stimulation and in case of obesity.

We report an exceptional case of degeneration of a deep endometriosis lesion of the uterine torus into high‐grade clear cell carcinoma. Among the risk factors in this woman, we found obesity, ovarian stimulation, and perimenopause. Despite multiple procedures and additional examinations (ultrasound, MRI) this degeneration was not well known. This clinical case should alert us to the risk of endometriosis degeneration in case of recurrence, persistence, and growth of endometriosis lesions in a woman over 40 years of age.

We report the case of a 44‐year‐old female patient with a carcinoma developed at the site of a deep endometriosis lesion. This case should alert us to the risk of endometriosis degeneration in case of recurrence, persistence, and growth of endometriosis lesions in a woman over 40 years. Endometriosis affects 10%–15% of the female population of childbearing age and 20%–50% of infertile women. 1 Malignant transformation in endometriotic lesions is a rare event (1%), 2 , 3 most often described in cases of external endometriosis, in particular ovarian endometriosis 4 , 5 and more rarely in cases of deep infiltrating endometriosis (DIE). DIE is defined by infiltration of the peritoneum ≥5 mm or invasion of the bowel, ureter, or bladder by endometriotic lesions. 6 It includes endometriosis of the abdominal wall, rectovaginal septum, utero‐sacral, and utero‐ovarian ligaments. 7 Degeneration of deep infiltrating lesions is exceptional and less than 10 cases have been published to date. 8 Our case report was therefore uniqueness. We report the case of a 44‐year‐old female patient with a high‐grade clear cell carcinoma of the uterine torus developed at the site of a deep endometriosis lesion.

The authors report no conflicts of interest in relation to the present study.