Endometrial cancer is the most common malignant tumour of the uterine corpus and ranks among the leading gynaecological cancers worldwide [1] [2]. According to GLOBOCAN 2022 estimates, it was the sixth most common cancer in women, with an age-standardised incidence rate of 8.4 per 100,000 women [3]. Its incidence is rising in many countries, in relation to population ageing, obesity and diabetes [4]. In sub-Saharan Africa, it accounts for 6% to 9% of cancers in women [5]. In Gabon, epidemiological data on endometrial cancer remain scarce and outdated. They are based mainly on hospital-based series and on data from the Akanda Cancer Institute (ICA), which do not allow its incidence to be accurately estimated at national level. The study by Meye et al. in Gabon [6], in 2000, reported a frequency of 5.3%. A more recent study from the same country shows that endometrial cancer is the fourth most common cancer in women, after breast, cervical and ovarian cancer [7].

The diagnosis of this cancer is often made late, which accounts for the difficulties encountered in its management [8]. It is in this context that we undertook this work, in order to describe the epidemiological, diagnostic and therapeutic features of endometrial cancer at the Fondation Jeanne Ebori Mother and Child Teaching Hospital (CHUMEFJE).

Our study was conducted at the CHUMEFJE in Libreville, the national referral centre for the diagnosis and surgical management of gynaecological and breast cancers in Gabon. This was a retrospective descriptive study carried out from 1 January 2020 to 31 December 2025. The records of all patients with a histologically confirmed diagnosis of endometrial cancer during the study period were included. Records with incomplete or unusable data were excluded. Patients diagnosed with endometrial cancer in the department underwent clinical examination and a diagnostic work-up. This work-up comprised pelvic ultrasound, followed by endometrial sampling with a Novak curette or a Pipelle de Cornier in the obstetrics and gynaecology outpatient clinic; or hysteroscopy in the operating theatre under general anaesthesia when outpatient sampling was difficult or not feasible. Biopsy specimens were sent for histopathological examination, which determined the histological type and tumour grade. Staging work-up was based on pelvic MRI and CT of the chest, abdomen and pelvis, allowing staging according to the FIGO 2009 classification. In patients who did not undergo surgery, staging was based on clinical and radiological data. Cases were prepared and then discussed at a multidisciplinary team (MDT) meeting in order to define the treatment strategy. The treatment options agreed at the MDT meeting depended on the stage and on prognostic factors. A standardised data collection form was designed. Epidemiological, clinical, diagnostic, histopathological and therapeutic data were extracted from medical records and outpatient registers, and analysed using Excel.

During this period, 33 cases of endometrial cancer were identified out of a total of 463 gynaecological and breast cancers managed, i.e. a frequency of 7.1%.

The sociodemographic characteristics and past medical history of the patients are presented in Table 1. Mean age was 64 years (range 48 - 80). Patients not in paid employment accounted for 72.7% (n = 24). Thirty-one patients (93.9%) were postmenopausal and 24 (72.7%) were multiparous. Hypertension was found in 54.5% (n = 18) of cases, whereas a family history of breast cancer was reported in only one patient (3.0%).

Table 1. Sociodemographic characteristics and past medical history.

The clinical characteristics of the patients are presented in Table 2. Postmenopausal bleeding was found in 93.9% (n = 31) of cases. Pelvic pain was reported in 21.2% (n = 7) of cases, while 6.1% (n = 2) of premenopausal patients presented with abnormal uterine bleeding. An impaired general condition was noted in 42.4% (n = 14) of cases. On gynaecological examination, the cervix was grossly normal in 93.9% (n = 31) of cases.

Table 2. Clinical characteristics of the patients.

The time from symptom onset to first consultation was documented for all patients. In our series, the mean time to presentation was 7 months (range 1 - 24 months). The median duration was 5 months (interquartile range: 3 - 9 months).

The findings of the imaging investigations are presented in Table 3. On pelvic ultrasound, an intrauterine mass was found in 51.5% (n = 17) of cases. Pelvic MRI, performed in 30 patients (90.9%), showed extrauterine extension in 56.7% (n = 17) of cases. CT of the chest, abdomen and pelvis revealed distant metastases in 34.4% (n = 11) of cases.

Table 3. Findings on imaging investigations.

The histopathological characteristics are presented in Table 4. The diagnosis was established by biopsy curettage with a Novak curette in 84.8% (n = 28) of cases. Endometrioid adenocarcinoma accounted for 75.8% (n = 25) of the histological types identified. Tumour grade was assessed only in the 25 patients with endometrioid adenocarcinoma; of these, 16 (64%) had a grade 3 tumour.

Table 4. Histopathological characteristics of the patients.

Table 5 shows that advanced stages (FIGO III-IV) accounted for 54.5% (n = 18) of cases.

Table 5. Distribution by FIGO stage.

The treatment modalities are presented in Table 6. Surgery was indicated in 17 (51.5%) patients but performed in 12 (36.4%). Procedures comprised total hysterectomy with bilateral salpingo-oophorectomy in 6.1% (n = 2) of cases, total hysterectomy with bilateral salpingo-oophorectomy and pelvic lymphadenectomy in 8 patients (24.2%), and total hysterectomy with bilateral salpingo-oophorectomy, infracolic omentectomy, pelvic and para-aortic lymphadenectomy in 6.1% (n = 2) of cases. Chemotherapy was administered to 16 patients (48.5%), mainly with palliative intent (68.8%), while radiotherapy was delivered in 24.2% (n = 8) of cases.

Table 6. Treatment modalities of the patients.

TH: total hysterectomy BSO: bilateral salpingo-oophorectomy; PLND: pelvic lymphadenectomy; PALND: para-aortic lymphadenectomy.

This study has limitations, mainly inherent in its retrospective design. In particular, we faced inconsistent archiving of medical records, incomplete registers and missing data when completing the collection forms. These shortcomings made some information difficult to use and may have limited the completeness of the analysis.

In our series, endometrial cancer accounted for 7.1% of all gynaecological and breast cancers. This frequency is higher than that reported by Meye et al. [6] in Gabon (5.3%), but lower than those reported by Tonato et al. [9] in Benin (9.5%) and Engbang et al. [10] in Cameroon (11.6%). These differences should nevertheless be interpreted with caution, given the heterogeneity of the study populations, study periods, inclusion criteria and denominators used. Our frequency was calculated over all gynaecological and breast cancers, as in the Beninese study, whereas the Cameroonian study reports the proportion of endometrial cancer among female genital cancers, excluding breast cancer. Furthermore, the study periods cover different durations (6 to 10 years), which may influence the number of cases recorded. The frequency in our series is higher than that of Meye et al., which may be explained by differences in study period and in patient recruitment. It may also reflect the worldwide trend towards an increasing incidence of endometrial cancer described by several authors, attributed to improved diagnostic methods, population ageing and the rising prevalence of metabolic risk factors such as obesity, diabetes and hypertension [11]-[13]. Mean age at diagnosis was 64 years, which was higher than that reported by Anakwenze et al. in South Africa [8], Coulibaly et al. in Gabon [5], and Ndamba et al. in Cameroon [10], who reported 57.7, 63 and 53 years respectively. These differences may be explained by variations in the characteristics of the study populations, the study periods and the recruitment methods. Although mean age in our series was relatively low, endometrial cancer mainly affected postmenopausal women. With regard to parity, most of our patients were multiparous, as in several African series [5] [10].

The clinical presentation of endometrial cancer is dominated by postmenopausal bleeding, which is the main warning symptom of the disease. In our series, it was found in 93.9% of patients, close to the frequencies reported by Meye et al. (100%) [6] and Coulibaly et al. (95.8%) [5]. Pelvic pain was present in 21.2% of patients. The marked predominance of postmenopausal bleeding in our series, as in the studies cited above, underlines its value as a red-flag symptom and the importance of prompt diagnostic work-up in any woman presenting with this symptom.

Imaging plays a key role in the diagnosis and pre-treatment work-up of endometrial cancer. In our series, transvaginal ultrasound mainly showed heterogeneous endometrial thickening or an intracavitary mass. Pelvic MRI showed extrauterine extension in more than half of the patients, while CT of the chest, abdomen and pelvis revealed distant metastases in 34.4% of cases. These findings reflect advanced disease in our setting and underline the value of imaging in the assessment of tumour extension and in staging [1].

In our series, the Novak curette was the main method of endometrial sampling, used in 28 cases (84.8%). The other modalities were hysteroscopy in 1 case (3%), biopsy at laparotomy in 2 cases (6.1%), and diagnosis made on the hysterectomy specimen in 2 cases (6.1%). The predominance of Novak curette biopsy in our study is explained by its simplicity, its low cost, its availability and the ease with which it can be performed in an outpatient setting. This technique remains widely used as the first-line diagnostic test in many countries, particularly in resource-limited settings [14]. International guidelines nevertheless consider hysteroscopy with directed biopsy to be the most accurate technique where available, because of more representative sampling of the lesion and better concordance with the final histology [1] [14].

In our study, endometrioid adenocarcinoma was the predominant histological type, with 25 cases (75.8%), followed by serous papillary carcinoma (15.2%) and carcinosarcoma (9.1%). This distribution is broadly comparable to published data, in which endometrioid adenocarcinoma accounts for approximately 75% to 85% of endometrial cancers [1] [2] [15]. The frequency of serous papillary carcinoma (15.2%) appears slightly higher than the rates usually reported in Western series (5% to 10%). This non-endometrioid subtype is characterised by more aggressive behaviour, a higher risk of extrauterine spread and a poorer prognosis [2] [16]. Carcinosarcoma, a rare and aggressive tumour, accounted for 9.1% of cases in our series, a proportion higher than the frequencies usually described [16] [17]. This may be explained by referral bias towards advanced disease in a tertiary centre. The high proportion of grade 3 tumours further suggests a potentially more aggressive tumour profile. It should nevertheless be recalled that FIGO grading applies mainly to endometrioid carcinomas, whereas serous papillary carcinomas and carcinosarcomas are considered high-grade tumours by definition [1] [18].

In our study, staging was based on the FIGO 2009 classification (stages I to IV), without taking molecular features into account. Advanced-stage disease (stages III-IV) accounted for 54.5% of cases, versus 45.5% for stages I-II. This predominance of advanced-stage disease contrasts with data from Western series, in which endometrial cancer is mostly diagnosed at an early stage, particularly stage I [19] [20]. In our setting, this may be explained by diagnostic delay related to late presentation and to limited access to specialised centres, as well as by the predominance of certain aggressive histological subtypes. It is also consistent with the high proportion of high-grade tumours observed in our series. Comparisons with African series nevertheless remain difficult, because of the heterogeneity of the study populations and the variable availability of staging data, as illustrated by the African series previously reported [6] [9] [10].

Surgery is the cornerstone of the management of endometrial cancer. The standard procedure is total hysterectomy with bilateral salpingo-oophorectomy, with or without pelvic lymphadenectomy, when the tumour is confined to the uterine corpus [1]. In our series, 17 patients (51.5%) had an indication for surgery, but only 12 (36.4%) actually underwent an operation. Surgery consisted mainly of total hysterectomy with bilateral salpingo-oophorectomy, combined with pelvic lymphadenectomy in most cases and, for the most advanced disease, with para-aortic lymphadenectomy and omentectomy. In Western series [19] [20], surgery is the standard of care for resectable endometrial cancer, providing both tumour treatment and surgical staging. The relatively low proportion of patients who underwent surgery in our series may be explained by the high frequency of advanced-stage disease (54.5% stages III-IV), as well as by the predominance of aggressive histological types and high-grade tumours. With regard to additional treatments, 16 patients (48.5%) received chemotherapy: 3 (9.1%) in the neoadjuvant setting, 2 (6.1%) in the adjuvant setting and 11 (33.3%) with palliative intent. The predominance of palliative chemotherapy is consistent with the high frequency of advanced and metastatic disease in our series. In Western guidelines [1], chemotherapy is mainly indicated in high-risk, locally advanced or metastatic disease, whereas neoadjuvant chemotherapy may be considered in selected patients who are not candidates for upfront surgery. Radiotherapy was delivered to 8 patients (24.2%), a proportion that may be explained by indications selected according to stage, risk of recurrence and access to this treatment modality. In Western countries, management relies on more standardised and diversified approaches, including biomarker-driven targeted therapies, adjuvant pelvic radiotherapy, lymphadenectomy and different surgical approaches to hysterectomy (laparoscopic or open), evaluated in numerous studies [1] [8]. This contrast highlights the persistent inequalities in the management of endometrial cancer between resource-limited and high-income countries. It underlines the need not only to improve access to early diagnosis, but also to make adequate technical facilities and standardised treatment options available in low-resource countries. Improving health infrastructure, training health professionals and developing guidelines adapted to local realities are essential levers for reducing these disparities and optimising patient management [3]. Survival could not be assessed in our study, because patients were systematically referred to the ICA for continuation of treatment and no subsequent follow-up data were available in our department.

In our setting, endometrial cancer is characterised by a high frequency of advanced-stage disease (54.5% stages III-IV) and aggressive histological profiles, underlining the need not only for earlier diagnosis, but also for improved technical facilities and appropriate multidisciplinary management.

Eya’ama and Mba Edou wrote the article. Assoumou Obiang corrected it. They also covered the publication costs. The other authors work in the department where the patients were treated.