Female reproductive tract parenchyma develops from epithelia of the Mullerian Duct and urogenital sinus [5]. The Mullerian duct is derived from intermediate mesoderm and composed of epithelial and mesenchymal cells [1]. The development of the Mullerian ducts includes specification, invagination and elongation. Around gestational week 8, the Mullerian ducts’ terminal ends join together. A midline epithelial septum separates the lumen of two adjacent Mullerian ducts temporarily. During the 9th week, the midline septum disappears largely resulting in the formation of the midline uterovaginal canal. At about 12 weeks gestation, the basement membranes disappearing makes the uterus created. At about 14–15 weeks, gland formation in the uterine corpus and uterine cervix initially appears. The primary glands are found distant from the uterine fundus which proves that the first forming glands are possible to be cervical glands. The uterine and cervical glands which are initially lined by a simple columnar epithelium are elongated and branched within the stroma by 20 weeks [6, 7]. Based on the embryological observations, it can be speculated that the two uterine cavities of double uterus have the same risk of pathological changes.
We reviewed published articles about uterus didelphys combining endometrial cancer in PubMed from 1990 to 2020 [8,9,10,11,12,13,14,15,16,17,18,19,20,21,22]. There are 15 cases included, the age of onset ranged from 33 to 75 years old. Among them, 4 cases [11, 12, 15, 16] were less than 45 years old. Only 1 case had finished child-bearing before operation [12], 3 cases had staged operation, and only 1 case had bilateral adnexa reserved [11]. Many guidelines and experts achieved consensus on fertility preservation therapy for endometrial cancer at home and abroad. There are no discussion about the preservation of fertility function in young patients with uterus didelphys and endometrial carcinoma due to the rare occurrence.
In this case, an infertile woman with uterus didelphys was diagnosed as endometrial cancer in the left uterine cavity at the age of 28 years. The patient's endometrium did not achieve reversal after the classic high-efficiency progesterone treatment which led to the surgery to remove the left uterus. According to the embryological observations, uterus didelphys is due to the failure of fusion of the Mullerian ducts to form the uterus. Whether to keep the contralateral uterus is worthy of more discussion and attention.
By reviewing the previous 15 cases of double uterine malformation with endometrial cancer, we concluded that 7 cases (7/15) occurred in the right uterus with cancer staging from Ia–II and differentiation types from G1–G3. Among the 7 cases, only 1 case combined with atypical hyperplasia in the left endometrium. 4 cases (4/15) occurred in the left uterus with cancer staging from Ia-IIIc and differentiation type from G1 to G3, 1 of 4 cases was associated with atypical hyperplasia in the right endometrium. 4 cases (4/15) occurred in bilateral uterus with 1 case in stage IA (differentiation type G1), 1 case in stage II (differentiation type G2), 1 case in stage IIIA (differentiation type G1), and 1 case in stage IV (differentiation type G3).
From the limited data point of view, most of the patients (11/15) with uterus didelphys had endometrial cancer on one side of the uterus, two cases had atypical hyperplasia of endometrium on the contralateral uterus. Atypical hyperplasia of endometrium is the precancerous lesion of endometrial adenocarcinoma. 29% of complex atypical hyperplasia without treatment will develop into cancer [23]. Advanced endometrial carcinoma was more common in bilateral uterus, only one case of advanced endometrial cancer occurred in the left uterine cavity with the differentiation type as G3 [14].
Back to our case, the patients was diagnosed as stage IA of moderately differentiated endometrial carcinoma in the left uterus. No cancer or precancerous lesions were detected in the contralateral uterus given the previous hysteroscopy results. Only the affected side of the uterus and appendix were removed in the surgery. It is certain that the patient needs reviewing transvaginal ultrasound, MRI, and hysteroscopy of the reserved uterus in the follow-up treatment.
In order to ascertain the safety of preserving the fertility of this patient, we would like to search for prognostic biomarkers to predict the progression amongst the normal endometrium, endometrial atypical hyperplasia and endometrial cancer. PTEN is the most commonly investigated biomarker in endometrial cancer. However, PTEN is expressed normally in proliferative endometrial glands and stroma but expressed decreasingly in secretory cycle which may results in variable staining [24], Recently, scientists put more interests on PAX2 which is less frequently lost in both normal proliferative and secretory endometrium than PTEN [25].
PAX2 can be expressed by immunohistochemistry inside normal endometrial glandular cells’ nucleus. In 2018, Emma found a progressive decrease in PAX2 expression from proliferative endometrium to endometrial intraepithelial neoplasia (EIN) to endometrial endometrioid carcinoma (EEC) which is also observed in previous studies. It is acknowledged that it can take 30–40 years from normal endometrium to endometrial cancer. Despite of the small sample size, this study is quite popular since its long follow-up of these patients. In Emma’s study, they presented a method and threshold with PAX2 expression H-score as 75. EIN lesions with a PAX2 expression H-score below or equal to 75 (high risk) had significantly lower progression-free survival than those with an H-score above 75 (low risk). Furthermore, more EIN patients with high-risk PAX2 showed progression (73%; 8/11) compared to low-risk patients (8%; 2/25) [26].
In this study, our patient has done hysteroscopic surgeries 4 times in Peking University First Hospital among which the right uterine endometrium was examined 3 times. To better understand the risk of whether the reserved side of uterus will have pathological progress, our pathologist examined PTEN and PAX2 on three right endometrial pathological sections of the patient. The results showed that the right endometrium obtained from three hysteroscopic operations showed strong positive expression of PTEN and PAX2. This may tell clinicians that the possibility of endometrial lesions on the right side is not significant from the onset to the last operation. In the follow-up process, the PTEN and PAX2 immunohistochemical analysis of endometrium can be carried out. If possible, the H score of PAX2 can be used to better complete the follow-up of endometrial lesions, so as to adjust the treatment plan in time.
Endometrial cancer occurring in patients with uterus didelphys is quite rare in published articles, especially in young patients with needs for childbearing. In this case report, the patient has the unilateral uterus with endometrial cancer removed due to the unresponsiveness to drugs. According to the IHC analysis of the reserved side uterine endometrium, there is no pathological progress until the last hysteroscopic surgery. However, with less evidence, it requires more follow-up in later period to trace the patient’s tumor and fertility outcome. In order to shorten the time required for fertility, the patient chose assisted reproductive technology in our hospital. During the process of ovarian hyperstimulation and After the achievement of fertility outcome, we will keep monitoring the patient’s transvaginal ultrasound, MRI, and hysteroscopy. The application of H score of PAX2 will be considered to evaluate the risk of reserving the contralateral uterus.