Large cervical endometrioid adenocarcinoma mimicking endometrial endometrioid adenocarcinoma: A case report.

A 58-year-old Japanese woman (gravida 1, para 1) presented to her family doctor as an outpatient with complaints of gradually increasing lower abdominal pain and irregular vaginal bleeding, which had persisted for 1 year. She was petite (146 cm, 37 kg) and lean (body mass index: 17.9 kg/m 2 ) with a pale face. Her medical, surgical, allergic, family, and social histories were insignificant. She was married and had children (twins). She had no history of HPV vaccination, experienced menopause at the age of 56 years, and did not take any medications, smoke, drink alcohol, or use illicit drugs. Abdominopelvic computed tomography (CT) without contrast revealed an enlarged uterus. Laboratory tests showed elevated white blood cell count (32,200 per mm 3 ; reference range 3,300-8,600/mm 3 ), elevated C-reactive protein level (17 mg/dL; reference range 0-0.14 mg/dL), and severe anemia (hemoglobin level: 5.5 g/dL; reference range 11.6-14.8 g/dL). Uterine cancer was suspected, and she was referred the same day to a gynecologist at another institution for further evaluation. She was immediately hospitalized in the gynecology department. Speculum examination showed a large tumor that easily bled in the vaginal canal, which did not allow for further evaluation of the extension onto the vaginal or pelvic wall. Bimanual examination enabled the palpation of the enlarged uterus with restricted mobility. The uterine fundus was palpable midway between the symphysis pubis and the umbilicus. She was started on transfusion therapy with 6 units of packed red blood cells for anemia. Tumor markers, SCC and CA125, were elevated to 5.3 ng/mL (reference range 0.6-2.5 mg/mL) and 112.7 U/mL (reference range ≤35.0 U/mL), respectively. Cervical cancer was suspected, warranting a total body CT for staging. Contrast-enhanced CT (delayed phase) revealed no distant metastases but demonstrated multiloculated, thick-walled cystic lesions in both uterine spaces and fluid retention extending from the pelvic cavity to the left lateral paracolic gutter, suggestive of a pelvic abscess. She was concurrently started on antimicrobial therapy with ceftriaxone (4 g/day) and metronidazole (1.5 g/day). Contrast-enhanced magnetic resonance imaging (MRI) revealed an 11 × 9-cm tumor with central necrosis in the uterine cervix; however, the endometrial cavity was spared and distended with secretions ( Fig. 1 A, B, and C). Contrast-enhanced MRI also indicated no direct extension into the rectum, parametrium ( Fig. 1 D), or vagina and regional lymphadenopathy; nevertheless, bladder mucosal invasion could not be ruled out. She underwent a colposcopy-directed biopsy. Histological examination of the biopsy specimen, stained with hematoxylin and eosin, indicated endometrioid adenocarcinoma. Subsequently, the patient was transferred to our department for further evaluation to distinguish endometrioid ECA from endometrioid carcinoma of the LUS. Fig. 1 Pretreatment MRI of the pelvis and pathological profiles of cervical biopsy. (A) Sagittal T2-weighted image with turbo spin echo sequence shows a large cervical cancer with distension of the endometrial cavity. (B) Sagittal diffusion-weighted image shows the tumor with hyperintensity. (C) Sagittal apparent diffusion coefficient map shows the tumor with low signal intensity. (D) Axial T2-weighted image with turbo spin echo sequence shows the absence of parametrial invasion. (E) Hematoxylin and eosin staining demonstrates columnar epithelium with densely stained and enlarged nuclei that form fused glandular ducts with a 6-50% solid nest without severe atypia. (F) Immunoreactivity shows diffuse-positive staining for vimentin. (G) Immunohistochemistry shows diffuse-positive staining for estrogen receptors. (H) Immunohistochemistry shows almost negative staining for progesterone receptors. (I) Immunohistochemistry shows patchy-positive staining for p16. Fig 1 Pretreatment MRI of the pelvis and pathological profiles of cervical biopsy. (A) Sagittal T2-weighted image with turbo spin echo sequence shows a large cervical cancer with distension of the endometrial cavity. (B) Sagittal diffusion-weighted image shows the tumor with hyperintensity. (C) Sagittal apparent diffusion coefficient map shows the tumor with low signal intensity. (D) Axial T2-weighted image with turbo spin echo sequence shows the absence of parametrial invasion. (E) Hematoxylin and eosin staining demonstrates columnar epithelium with densely stained and enlarged nuclei that form fused glandular ducts with a 6-50% solid nest without severe atypia. (F) Immunoreactivity shows diffuse-positive staining for vimentin. (G) Immunohistochemistry shows diffuse-positive staining for estrogen receptors. (H) Immunohistochemistry shows almost negative staining for progesterone receptors. (I) Immunohistochemistry shows patchy-positive staining for p16. Speculum examination was limited because of the large tumor occupying the vaginal canal; however, rectal examination confirmed a barrel-shaped cervix extending from the LUS to the hymenal ring. The bimanual evaluation indicated that the tumor involved the lower third of the vagina, with no extension to the pelvic wall. A cervical biopsy revealed a columnar epithelium with densely stained and enlarged nuclei that formed fused glandular ducts ( Fig. 1 E). The tumor exhibited a 6-50% solid nest without severe atypical atypia. Immunoreactivity was diffuse-positive for vimentin ( Fig. 1 F) and estrogen receptors ( Fig. 1 G), almost negative for progesterone receptors ( Fig. 1 H), and patchy-positive for p16 ( Fig. 1 I), signifying a grade 2 endometrioid adenocarcinoma. Both the HPV test (Eurofins Genetic Lab, Hokkaido, Japan) for 16 high-risk HPV subtypes (HPV types 6, 11, 16, 18, 30, 31, 33, 35, 39, 45, 51, 52, 56, 58, 59, and 66) and the microsatellite instability test (SRL, Tokyo, Japan) yielded negative results. Cystoscopy revealed an intact bladder mucosa. Overall, we tentatively diagnosed the patient with the International Federation of Gynecology and Obstetrics (FIGO) stage IIIA endometrioid ECA (cT3a, cN0, M0). The patient received concurrent chemoradiotherapy (CCRT) with a total of 50 Gy in 25 fractions for whole-pelvis irradiation using the central shielding technique after a total dose of 40 Gy and weekly cisplatin (40 mg/m 2 ). Surgery was not indicated as the cervical cancer had progressed beyond stage IB2. The tumor shrank toward the posterior lip of the external os after 40 Gy of irradiation ( Fig. 2 A), and the tumor origin was then determined again. Diffusion-weighted MRI revealed residual diffusion restriction ( Fig. 2 B). Intracavitary brachytherapy with a high-dose-rate remote after loading was also administered to maximize local control (18 Gy in 3 fractions). The patient developed a urinary tract infection during treatment; however, she completed 6 cycles of weekly cisplatin. At the end of CCRT, cervical cytology still showed viable tumor cells Subsequently, systemic chemotherapy was administered to induce remission. After 3 cycles of tri-weekly chemotherapy with paclitaxel (180 mg/m 2 ; 3-h intravenous infusion) plus carboplatin (area under the curve: 6 mg/mL/min), the tumor shrank further ( Fig. 2 C), and its diffusion restriction disappeared ( Fig. 2 D). After 6 cycles of systemic chemotherapy ( Fig. 2 E), the cervical biopsy revealed no viable cells, and contrast-enhanced CT (portal venous phase) of the chest, abdomen, and pelvis showed no distant metastasis. Thus, the patient was shifted to post-treatment surveillance. After treatment, follow-up MRIs at 6 months ( Fig. 2 F) and ten months ( Fig. 2 G) confirmed tumor shrinkage and the return to normal anatomical structure. Fig. 2 Changes in the tumor during and after treatment. (A) Sagittal T2-weighted image with turbo spin echo sequence after 40 Gy of irradiation shows tumor shrinkage. (B) Axial diffusion-weighted image after 40 Gy of irradiation still shows the tumor with hyperintensity. (C) Sagittal T2-weighted image with turbo spin echo sequence after 3 cycles of chemotherapy shows further tumor shrinkage. (D) Axial diffusion-weighted image after 3 cycles of chemotherapy shows the tumor without diffusion restriction. (E) Sagittal T2-weighted image with turbo spin echo sequence after 6 cycles of chemotherapy shows residual tumor originating from the posterior wall of the cervix. (F) Sagittal T2-weighted image with turbo spin echo sequence shows the radiographic residual tumor, with the contemporary colposcopy [inset] shows a small amount of oozing from the external os (6 months after treatment). (G) Sagittal T2-weighted image with turbo spin echo sequence shows an almost normal anatomical structure, with the contemporary colposcopy [inset] shows normal findings (10 months after treatment). Fig 2 Changes in the tumor during and after treatment. (A) Sagittal T2-weighted image with turbo spin echo sequence after 40 Gy of irradiation shows tumor shrinkage. (B) Axial diffusion-weighted image after 40 Gy of irradiation still shows the tumor with hyperintensity. (C) Sagittal T2-weighted image with turbo spin echo sequence after 3 cycles of chemotherapy shows further tumor shrinkage. (D) Axial diffusion-weighted image after 3 cycles of chemotherapy shows the tumor without diffusion restriction. (E) Sagittal T2-weighted image with turbo spin echo sequence after 6 cycles of chemotherapy shows residual tumor originating from the posterior wall of the cervix. (F) Sagittal T2-weighted image with turbo spin echo sequence shows the radiographic residual tumor, with the contemporary colposcopy [inset] shows a small amount of oozing from the external os (6 months after treatment). (G) Sagittal T2-weighted image with turbo spin echo sequence shows an almost normal anatomical structure, with the contemporary colposcopy [inset] shows normal findings (10 months after treatment). At 18 months after the initial treatment, liver metastasis was suspected on routine follow-up CT. Further dynamic contrast-enhanced MRI revealed a 1.5 × 1.4-cm well-defined mass located just below the umbilical portion of the portal vein in segment 3 ( Fig. 3 A) with diffusion restriction ( Fig. 3 B). The tumor was adjacent to the portal vein without stenosis or distortion, and R0 resection was considered. The patient underwent left hepatectomy with laparotomy through a J-shaped incision without a preoperative needle biopsy. The estimated blood loss was 430 mL, and the postoperative course was uneventful. The final pathological examination showed a metastatic carcinoma originating from the uterine cervix ( Fig. 3 C). The patient again received paclitaxel plus carboplatin chemotherapy for 6 cycles and achieved complete remission. She was last examined at 23 months after hepatectomy and remained well, with no evidence of disease on whole-body CT scanning. Fig. 3 Appearance of liver metastasis. (A) Axial, 3-dimensional T1-weighted image with gradient echo sequence shows a solitary tumor located just below the umbilical portion of the portal vein in segment 3 (arrow). (B) Diffusion-weighted image shows the tumor with hyperintensity. (C) Hematoxylin and eosin staining of the resected hepatic tumor shows endometrioid adenocarcinoma (metastatic tumor). Fig 3 Appearance of liver metastasis. (A) Axial, 3-dimensional T1-weighted image with gradient echo sequence shows a solitary tumor located just below the umbilical portion of the portal vein in segment 3 (arrow). (B) Diffusion-weighted image shows the tumor with hyperintensity. (C) Hematoxylin and eosin staining of the resected hepatic tumor shows endometrioid adenocarcinoma (metastatic tumor).

Due to the retrospective nature of this case report, the need for institutional review board approval was waived.

Written informed consent was obtained from the patient and her husband for the publication of anonymized information in this report.

We report the case of a woman with a large and bulky endometrioid ECA that was diagnosed based on the 2020 WHO/2018 IECC guidelines and successfully treated with CCRT. Even with an ECA classification based on HPV status, diagnosing an endometrioid carcinoma arising from the upper endocervix or LUS of equivocal origin presents a challenge. Although pathological assessment should be considered, the final diagnostic decision should be made clinically. MRI plays a crucial role in accurately assessing prognostic indicators of cervical and corpus cancers at initial diagnosis. Additionally, it is important during post-treatment follow-ups to distinguish between mimics that can simulate both cancers. Further prospective studies focusing on the clinical, pathological, radiological, and molecular profiles of ECA based on the 2020 WHO/2018 IECC guidelines are needed.

This case report presented 2 important clinical issues. First, we encountered a rare case of a large endometrioid ECA newly diagnosed after the release of the 2020 WHO/2018 IECC guidelines. In cases with bulky or large uterine tumors of uncertain origin, the pretreatment diagnosis can be challenging and may change during the course of treatment. In this case, the patient was primarily diagnosed with endometrioid ECA via clinical and radiographic assessments and confirmed with certainty by sequential intra- and post-treatment MRI. Second, the tumor was not amenable to curative surgery owing to its size and progression but was successfully treated with CCRT followed by systemic platinum-based chemotherapy. Although the tumor recurred in the liver, it was successfully treated with multidisciplinary therapy. The 2020 WHO/2018 IECC guidelines define endometrioid ECA as having endometrioid morphology with “confirmatory endometrioid features.” These features include at least focal identification of low-grade endometrioid glands lined by columnar cells with pseudostratified nuclei showing no more than moderate atypia, with or without squamous differentiation and/or associated endometriosis, and lacking HPVA features [5] . Table 1 summarizes endometrioid ECA cases reported after 2018, when the new IECC classification system was published [9] , [10] , [11] , [12] , [13] . We conducted a literature search in PubMed/MEDLINE for articles published in English with available abstracts and with publication dates ranging from January 1, 2018, to January 1, 2024, using the terms “cervical adenocarcinoma” and “endometrioid.” As shown in this table, all reported cases of endometrioid ECA in the literature underwent hysterectomy, except for our case. Therefore, we have discussed the diagnostic rationale for endometrioid ECA when hysterectomy is not feasible. Table 1 Reported cases of endometrioid endocervical adenocarcinoma after 2018. Table 1 References Year Age (years) Tumor diameter (cm) FIGO stage HPV status p16 expression MMR status Coexistent precursor/cancer First-line treatment Outcome Seay et al. [9] 2020 76 2.4 IB1 Neg * Focal pMMR (IHC) None RoH * + PLNB + OMB, followed by CCRT NED (15 M) Akizawa et al. [10] 2020 68 3.0 IB1 n/a Neg n/a Endometrial SCC TAH + BSO, followed by PTX + CBDCA NED (12 M) Uehara et al. [11] 2020 49 4.0 IIB Neg Focal n/a Atypical endometriosis RaH + BSO + PLND, followed by DOX + CDDP NED (12 M) Li et al. [12] 2023 75 3.4 IB1 n/a Patchy n/a Endometrial EC TAH + BSO + PLND n/a Monist et al. [13] 2023 60 9.0 IB1 n/a Patchy n/a Endometrial EC TAH + BSO + OM + PALND n/a Present case 2024 58 11 IIIA Neg Focal pMMR (microsatellite instability) n/a CCRT, followed by PTX + CBDCA Liver metastasis (18 M) ⁎ Status after BSO. Abbreviations: FIGO, International Federation of Gynecology and Obstetrics; HPV, human papillomavirus; MMR, mismatch repair; IHC, immunohistochemistry, RoH, robotic-assisted simple hysterectomy; PLNB, pelvic lymph node biopsy; OMB, omental biopsy; CCRT, concurrent chemoradiotherapy; n/a, not available; SCC, squamous cell carcinoma; TAH, total abdominal hysterectomy; BSO, bilateral salpingo-oophorectomy; PTX, paclitaxel; CBDCA, carboplatin; RaH, radical hysterectomy; PLND, pelvic lymph node dissection; DOX, docetaxel; CDDP, cisplatin; EC, endometrioid adenocarcinoma, PALND, para-aortic lymph node dissection. Reported cases of endometrioid endocervical adenocarcinoma after 2018. Status after BSO. Abbreviations: FIGO, International Federation of Gynecology and Obstetrics; HPV, human papillomavirus; MMR, mismatch repair; IHC, immunohistochemistry, RoH, robotic-assisted simple hysterectomy; PLNB, pelvic lymph node biopsy; OMB, omental biopsy; CCRT, concurrent chemoradiotherapy; n/a, not available; SCC, squamous cell carcinoma; TAH, total abdominal hysterectomy; BSO, bilateral salpingo-oophorectomy; PTX, paclitaxel; CBDCA, carboplatin; RaH, radical hysterectomy; PLND, pelvic lymph node dissection; DOX, docetaxel; CDDP, cisplatin; EC, endometrioid adenocarcinoma, PALND, para-aortic lymph node dissection. Three possible differential diagnoses can be made for endometrioid adenocarcinoma involving the cervix—(i) a primary ECA, (ii) a primary endometrial adenocarcinoma arising in the fundus and extending to involve the cervix (hereinafter called “uterine fundus cancer”), and (iii) a primary endometrial adenocarcinoma arising in the LUS and secondarily involving the cervix (hereinafter called “uterine isthmus cancer”) [14] . Imaging studies such as MRI are considerably useful in ruling out uterine fundus cancer from ECA by judging whether the endometrial cavity is spared and/or distended with secretions, irrespective of the histological type [15] . Distinguishing endometrioid ECA from endometrioid uterine isthmus cancer is difficult even in hysterectomy specimens using immunohistochemistry, which can be attributed to the following 2 reasons. First, bulky tumors often destroy not only the normal anatomical structure of the upper endocervix and uterine isthmus but also the precursor lesions, and tumors with extensive involvement of the uterine corpus and endocervix deviate from the definition of uterine isthmus cancer [8] . Second, their immunohistostaining patterns overlap: positive for estrogen and progesterone receptors, patchy (almost synonymous with focal) for p16, and negative for HPV [16] . The panel of markers should be valued only for distinguishing usual ECA (categorized as HPVA) from endometrioid uterine isthmus cancer [17] . Therefore, for endometrioid carcinoma of equivocal origin, especially with bulky uterine masses, the tumor origin should be determined based on the clinical findings and course. Table 2 presents the clinicopathological features of the 3 carcinomas [6 , 15 , 18] . Some cases were primarily diagnosed and treated as ECA but were consequently re-diagnosed as uterine isthmus adenocarcinoma after hysterectomy [8] . Lynch syndrome-related endometrial cancer with high microsatellite instability during CCRT was also reported [19] . Table 2 Comparison of clinicopathological features of cervical endometrioid adenocarcinoma arising in the endocervix, uterine fundus, and uterine isthmus. Table 2 Endometrioid adenocarcinoma in the uterine fundus extending to involve the cervix Endocervical endometrioid adenocarcinoma Endometrioid adenocarcinoma in the uterine isthmus involving the cervix Immunohistochemistry ER and PR positivity; patchy or focal p16 staining [18] HPV test Negative [18] Microsatellite instability test n/a n/a ∼30% [6] MRI findings The uterine cavity was spared and distended with secretions [15] Indistinguishable in large and bulky cases Rectal examination Presence or absence of continuity between the cervix and tumor Abbreviations: HPV, human papillomavirus; MRI, magnetic resonance imaging; n/a, not acceptable. Comparison of clinicopathological features of cervical endometrioid adenocarcinoma arising in the endocervix, uterine fundus, and uterine isthmus. Abbreviations: HPV, human papillomavirus; MRI, magnetic resonance imaging; n/a, not acceptable. The recent paradigm shift in diagnosing ECA, as well as its originally low morbidity, could limit the data on the prognosis of patients with endometrioid ECA. Stolnicu et al. conducted a central pathological review of 371 cases of ECA, including 41 (11%) cases of the endometrioid type, and reported that only 3 (0.8%) of the 41 endometrioid cases met the new 2018 IECC criteria [5] . They also found that NHPVA cervical cancers as an entire cohort were larger (median, 38 vs. 21 mm) and present at higher stages (48% vs. 8.6% in stage II/III) in older patients (median, 55 vs. 42 years) than HPVA cervical cancers [5] . Another study identified age, stage, and tumor size as independent prognostic factors for overall survival [20] . Remarkably, in these 2 studies, endometrioid ECA accounted for only approximately 1% of all ECA tumor types according to the 2018 IECC criteria, whereas usual ECA accounted for approximately 70% [5 , 20] . A recent study comparing usual ECA with endometrioid ECA found no statistically significant differences in overall survival [21] . In that study, using propensity score matching to adjust for background, the researchers designed a matched cohort of usual ECA (n = 280) and endometrioid ECA (n = 280) from the Surveillance, Epidemiology, and End Results (SEER) database [21] . Of note, patients with ECA were diagnosed based on the 2014 WHO criteria, resulting in a high percentage of endometrioid ECA (10%) compared with usual ECA (90%).

Cervical cancer ranks as the fourth most common cancer in women globally, accounting for an estimated 604,000 new cancer cases and approximately 34,2000 deaths by 2020 [1] . Endocervical squamous cell carcinoma and endocervical adenocarcinoma (ECA) are the 2 major histological subtypes of cervical cancer, with human papillomavirus (HPV) infection being the cause for almost all cases of the former [2] and 85% of the latter [3] . The incidence of squamous cell carcinoma has decreased owing to the spread of HPV vaccinations and screening programs; in contrast, the incidence of ECA has increased in recent years [4] . Recent findings suggest that current screening methods for preinvasive disease and invasive cervical cancer may be more effective for detecting squamous histology than adenocarcinoma histology [4] . The International Endocervical Adenocarcinoma Criteria and Classification (IECC) proposed a new pathological classification for ECA in 2018, in which ECA was divided into 2 categories according to the morphological features of HPV infection—namely, HPV-associated (HPVA) and non-HPV-associated (NHPVA) [5] . The 2020 World Health Organization Classification (WHO 2020) incorporated this concept and further divided NHPVA into 4 subcategories (gastric, clear cell, mesonephric, and endometrioid adenocarcinomas). The rare incidence and difficulty in distinguishing endometrioid ECA from endometrioid endometrial carcinoma make clinical management difficult. Uterine endometrioid adenocarcinoma can arise from the endocervix, uterine corpus proper (uterine body and fundus), and lower uterine segment (LUS, a synonym for the uterine isthmus) [6] . The LUS, which is located in the lower uterine body, is an area showing histological transition from the uterine lining (endometrium) to the endocervical glands [7] . In rare cases, a carcinoma localized in the LUS expands macroscopically or microscopically from the lower uterine body through the upper cervix and is subclassified as an LUS carcinoma or uterine isthmus cancer among uterine corpus cancers [8] . LUS carcinoma is reportedly associated with Lynch syndrome [6] . The identification of the tumor origin is crucial for first-line organ-specific therapy and appropriate prognostic prediction in patients. Herein, we report a rare case of a large endometrioid ECA in which the tumor origin was difficult to identify. In addition, we review and discuss the key points of differentiation from its mimics.