Case presentation

In June 2024, a patient in her 40s from Ecuador with a history of two cesarean sections and no comorbidities was referred to her gynecologist due to persistent vaginal discharge and an abnormal Pap smear. Gross examination of the cervix was normal. A colposcopy was performed, and random cervical biopsies were negative for any disease. The endocervical curettage revealed papillary adenocarcinoma of the cervix. MRI performed in Ecuador showed a uterus measuring 9.6×5×5.4 cm, with a 6.5×5.2 cm lesion in the endometrium extending to the cervix, without bladder or rectum involvement and no lymphadenopathy.

At that point, the patient presented to Houston Methodist Hospital for a second opinion. Physical examination and vital signs were normal. Speculum examination showed mild bleeding with macroscopically normal but bulging cervix. The bimanual exam was consistent with thickened lower uterine segment but no pelvic mass and no evidence of parametrial involvement. A positron emission tomography/computed tomography (PET/CT) scan showed a cervical mass (~5 cm) with probable extension into the lower uterine segment and bilateral uterine horns with no evidence of metastatic disease. A repeat pelvic MRI showed a 6.4 cm polypoid cervical mass with intracavitary extension and without myometrial invasion. The pathology from the original curettage was reviewed at Houston Methodist Hospital.

Dr Deavers (pathology): what were the findings at the time of review?

The outside curettage contained small, detached fragments of adenocarcinoma without underlying stroma. Some artifactual changes were present in the tissue. The cells were columnar without marked atypia; focal luminal mitoses were also seen. Foamy histiocytes were noted in the core of one of the fragments. With immunohistochemical staining performed, there was diffuse staining for p16, scattered staining for estrogen receptor (ER), and focal staining for vimentin. Carcinoembryonic antigen (CEA) was negative. P53 was wild-type. There was a focal equivocal signal for high-risk human papillomavirus (HPV; RNA in situ hybridization (ISH). While the immunohistochemical panel was not definitive, an endocervical primary was favored.

At that time, the differential diagnosis included primary endometrial versus endocervical adenocarcinoma due to overlapping features. The radiation oncologist evaluated the patient and considered the tumor as an adenocarcinoma of the endometrium extending down to the upper cervix. This was primarily based on pathology review and the appearance of disease on MRI.

Dr Farach (radiation oncology): are there specific imaging techniques or protocols that you find particularly useful in determining the primary site of disease?

Distingushing between endocervical carcinoma and endometrial adenocarcinoma can be difficult as in this case. Imaging features can assist in determining anatomical origin. On MRI, cervical cancer typically presents in a round or oval shape perpendicular to the long axis of the uterus, while endometrial cancer typically presents parallel to the long axis of the uterus. Locally advanced cervical cancer often results in parametrial invasion while endometrial cancer results more commonly in myometrial invasion. Cervical cancer is also more commonly hypervascular on dynamic contrast enhanced MRI and with high-signal intensity on T2 weighted imaging, whereas endometrial cancer is hypovascular and with lower signal intensity. The presented case on MRI demonstrated a hypovascular, low-signal intensity intrauterine mass growing parallel to the uterus without parametrial or myometrial invasion (f igure 1). Given a normal exocervix on exam, this was felt to be more consistent with endometrial primary based on imaging.1

Figure 1

(A) Sagittal magnetic resonance imaging (MRI) T2 fast recovery fast spin echo (frFSE) sequence; (B) Axial T2 frFSE; (C) axial positron emission tomography/computed tomography. These sequences show the extension of disease to uterine cornu.

Dr Farach: what is the efficacy of disease control with radiation in the setting of a uterine tumor extending down to the cervix?

For patients with medically inoperable endometrial cancer treated with chemoradiation and definitive brachytherapy, pelvic control rates range from 70% to 100%.2

Dr Farach: If the original biopsy had shown an endocervical adenocarcinoma, what treatment would you have recommended for a patient with this tumor size and projecting into the uterine cavity?

According to EMBRACE data, the expected 5-year local control rate for stage IB3 cervical cancer would be 98% with appropriate chemoradiation and image-guided adaptive brachytherapy.3 According to National Comprehensive Cancer Network (NCCN) guidelines, both chemoradiation therapy and radical hysterectomy with pelvic and/or para-aortic lymphadenectomy are appropriate treatment options4. Given the uncertainty in diagnosis and the disease extension to the bilateral uterine cornue, adequate coverage of the entire uterine mass would be complex. Given this, surgery was recommended as the most appropriate initial treatment option with adjuvant therapy to be determined based on surgical pathology.

After further discussion, the patient was offered the recommendation to proceed with surgery. In June 2024, the patient underwent a total abdominal radical hysterectomy, bilateral salpingo-oophorectomy, and sentinel lymph node mapping. The intraoperative exam showed a 3×3 cm enlarged cervix and 8 cm anteverted uterus without adnexal masses. In the abdomen, the uterus and bilateral adnexa were normal in appearance. There was no evidence of enlarged lymph nodes. The operating room time was 240 min, with an estimated blood loss of 100 mL.

Dr Deavers: what were the findings on the final pathology?

The ectocervix was smooth and measured 4.6×4.6 cm, with a 0.5×0.3 cm os on gross examination. Bivalving the uterus revealed a 6.4×3.4×2.3 cm fungating mass that involved the endometrial cavity, the lower uterine segment, and extended to the cervical os, but did not appear to extend through the os. It was deeply invasive, up to 1.8 cm. The parametrial margins appeared negative, with the closest 0.7 cm from the tumor bilaterally. The vaginal cuff also appeared negative, 1.2 cm from the tumor at the nearest extent. The mass was friable, and large necrotic areas were identified. Microscopically, moderately differentiated adenocarcinoma involved the endocervix, lower uterine segment, and endometrium. Endocervical adenocarcinoma in situ (AIS) and cervical intraepithelial neoplasia three were identified in association with the tumor. High-risk HPV (RNA ISH) was positive. The stromal invasion was 1.8 cm and extended to 0.15 cm from the 12 o'clock margin, with no lymphvascular invasion. The background endometrium was secretory with no atypical hyperplasia. The background endometrium was secretory with no atypical hyperplasia. All margins and lymph nodes were negative. PD-L1 combined positive score (CPS): 2. The tumor was classified as International Federation of Gynecology and Obstetrics (FIGO) stage IB3 (Figure 2).

Figure 2

(A) Moderately differentiated endocervical adenocarcinoma with evident luminal mitoses (H&E, 100×); (B) a focus of adenocarcinoma in situ (AIS) associated with the tumor (H&E, 200×); (C) punctate nuclear signal for high-risk human papillomavirus in the cervical intraepithelial neoplasia 3 area (RNA ISH, 400×).

Dr Deavers: what are the primary histological challenges distinguishing between endocervical and endometrial adenocarcinoma, and how does HPV status and immunohistochemistry influence your diagnostic approach?

Endocervical adenocarcinoma and endometrial adenocarcinoma can have overlapping microscopic morphology, particularly when endocervical adenocarcinoma has endometrioid-like or serous-like features. In these cases, identifying a precursor lesion, such as endocervical AIS or endometrial atypical hyperplasia, can be helpful. Additionally, there are some microscopic clues, such as luminal mitoses, increased apoptotic bodies, and goblet cells, that are more frequently found in endocervical adenocarcinoma.

In difficult cases, immunohistochemical studies can be valuable. A typical panel for a well to moderately differentiated adenocarcinoma might include vimentin, CEA, ER, and p16, although vimentin and CEA are not used as frequently now as in the past. In general, most endocervical adenocarcinomas stain for CEA and p16 (diffuse pattern), and are negative for ER and vimentin, while the reverse is true for endometrioid endometrial adenocarcinoma.5

Of course, since most endocervical adenocarcinomas are HPV related, obtaining a direct HPV test (if available) can be very useful, particularly in higher grade carcinomas that may be positive for p16 whether HPV driven or not.6

Dr Ramirez (gynecologic oncology): what was your discussion with the patient regarding the decision to proceed with surgery versus radiotherapy?

Endometrioid carcinoma is the most common subtype of endometrial cancer. However, there are a number of patients who may present with disease in the lower uterine segment with extension to the upper endocervical canal, where determining the primary site of disease is challenging. Cervical involvement occurs in approximately 10–15% of all endometrial cancer cases7, and both cervical and endometrial cancer can present clinically as a cervical mass, making findings at the time of physical examination inconclusive. In addition, an element to consider is that, although rare, some primary endocervical cancers have endometrioid differentiation. Consequently, distinguishing between primary endometrial and endocervical adenocarcinoma based solely on a biopsy or curettage results may also be difficult. MRI may help differentiate between the cervical and endometrial primaries. In the setting of cervical cancer, MRI is used to assess parametrial extension and lymph node involvement, and in endometrial cancer, it is used to determine the depth of myometrial invasion and cervical involvement. However, in ambiguous situations, the results in the literature are conflicting. In a study by Ramirez et al8 the authors reported the sensitivity and specificity of MRI in detecting cervical involvement from endometrial cancer as 72% and 93%, respectively, with a positive and negative predictive value of 90% and 80%, respectively. Also, in that study, the investigators showed that MRI was inaccurate or unhelpful in detecting primary disease sites in 43% of cases.

In such cases, it is very important to communicate with the patient the uncertainty as to the primary site of disease and the rationale for proceeding with the given recommendation. The patient needs to be informed as to what will be potential treatment options based on the final pathology. She needs to know that regardless of what the final pathology shows, there will be a clear and definitive recommendation. This will assure the patient that despite the uncertainty on the primary site of disease, the multidisciplinary team will formulate a treatment plan based on established guidelines.

Dr Ramirez: what factors led to the decision to perform an abdominal radical hysterectomy instead of a simple hysterectomy?

When evaluating the differential diagnosis between primary endometrial and endocervical adenocarcinoma, several factors influenced the decision to perform an abdominal radical hysterectomy instead of a simple hysterectomy. This decision was driven by the need to adequately address both potential diagnoses, given the overlapping features of the tumor. Considerations included the tumor size, location, depth of invasion, and the necessity to achieve clear surgical margins to minimize the risk of recurrence. According to NCCN guidelines, radical hysterectomy may improve local control and survival compared with simple hysterectomy in cases where cervical cancer is suspected, even in the setting of large tumors as long as there is no local-regional involvement.4.

Radical surgery in the setting of stage II endometrial cancer can help obtain negative margins, comprehensively address potential diagnoses, and reduce he risk of residual disease. However, some retrospective studies have not supported the routine use of radical hysterectomy for stage II endometrial cancer.9 For cervical cancer FIGO stage IB3, the NCCN guidelines recommend definitive external beam radiotherapy with concurrent platinum-based chemotherapy and brachytherapy (category 1).10 However, radical hysterectomy with pelvic lymphadenectomy is also an option that may be considered (category 2B). In addition, as a third option, external beam radiotherapy with concurrent platinum-based chemotherapy and brachytherapy followed by completion hysterectomy is also offered (category 3).

Dr Farach: given the patient’s final pathology, is adjuvant treatment necessary? What specific chemoradiation regimen is recommended?

For early-stage cervical cancer, NCCN guidelines recommend adjuvant treatment after radical hysterectomy in the setting of high pathological risk.10–12 In patients with two or more of the following criteria, adjuvant pelvic radiation therapy is recommended with or without a brachytherapy boost: tumor size >4 cm, lymphovascular invasion, and/or deep cervical stromal invasion.12 In patients with the following higher-risk pathological features, platinum-based chemotherapy is added to adjuvant pelvic radiotherapy: positive margin, involved parametria, or nodal involvement.11 Given deep stromal invasion and tumor size, adjuvant radiation therapy alone was recommended.

Closing summary

This case illustrates the complex diagnostic and management challenges in differentiating between endocervical and endometrial adenocarcinoma, particularly when the tumor involves both the cervix and the uterine cavity. In cases where the anatomical pathology is inconclusive, the tools at our disposal, such as MRI and hysteroscopy, may not provide definitive answers. Recent advancements in molecular profiling, liquid biopsy and radiomics, and novel imaging techniques, such as diffusion-weighted MRI and PET/CT, offer the potential for improving diagnostic precision. However, their integration into routine clinical practice is still evolving.13–15 European Society of Gynaecological Oncology/ European Society for Medical Oncology (ESMO-ESGO) guidelines also suggest that expert vaginal ultrasound examinations, performed by an expert sonographer, can be used for the detection of myometrial invasion and cervical stromal invasion instead of pelvic MRI.16

The ability to distinguish between primary cervical and endometrial cancer is essential to formulate an effective treatment plan in the postoperative period. Although current guidelines provide strong evidence that the treatment for stage IB3 cervical cancer should primarily involve chemotherapy and radiation, the management of stage II endometrial cancer remains a topic of debate. Despite NCCN guidelines recommending radical hysterectomy, the literature presents conflicting opinions.4 17 ,17 Several retrospective studies have demonstrated that the type of surgical approach—whether simple or radical hysterectomy—does not significantly impact overall survival in these patients.18 19 In addition, the ESMO-ESGO guidelines recommend performing a simple hysterectomy in this patient population, and more extensive procedures should only be performed if required to achieve free surgical margins.16 Therefore, a multidisciplinary approach involving a tumor board with radiologists, pathologists, radiation oncologists, and surgeons is crucial to accurately determine the appropriate pathology diagnosis and guide subsequent management.

Ethics statementsPatient consent for publication

Not applicable.