Endometrial carcinoma in a gravid uterus: a case report and literature review

Endometrial carcinoma (EC) is the fourth most common cancer in women in high-income countries; however, EC commonly occurs in peri- or postmenopausal women, and only 5% of women are diagnosed with adenocarcinoma before the age of 40 years [1, 2]. Therefore, the coexistence of EC and pregnancy is rare. Moreover, EC is rarely detected during pregnancy or within a year postpartum because the tumor can disrupt the pregnancy. Although a previous study had already reviewed the latest 35 reports on EC coexisting with pregnancy during the last 80 years [3], the outcome of EC associated with pregnancy and the effect of pregnancy on EC is not well known. Previous literature review also showed that there have been no reports of diagnosing EC during pregnancy in the surgical specimen of cesarean hysterectomy. Therefore, we report a case of EC diagnosed in a postoperative histopathological examination after total hysterectomy for placenta accreta spectrum (PAS), and we additionally present the results of a literature review on this matter.

A 35-year-old woman (gravida 2, para 1) was referred to our hospital due to placenta previa at gestational week 31. Her medical history was unremarkable, and her previous pregnancy was an uncomplicated, normal vaginal delivery at gestational week 38. Her current pregnancy was uncomplicated except for the placenta previa. She denied abnormal genital bleeding before the current pregnancy. Cervical cytology performed during early pregnancy was negative for intraepithelial lesions. Vaginal ultrasonography revealed total placenta previa and one lacuna (Fig. 1a). Magnetic resonance imaging (MRI) at gestational week 31 revealed total placenta previa and loss of the myometrium between the placenta and bladder wall (Fig. 1b). Other MRI findings of PAS such as uterine bulging, heterogenous placenta, and T2 dark band were not observed. Based on these findings, we suspected PAS, and an emergency cesarean delivery was performed owing to antepartum bleeding (approximately 100 mL) at gestational week 35. An abdominal midline incision was made, and a healthy male infant weighing 2274 g (− 0.42 SD) was delivered with Apgar scores of 8 and 9, at 1 and 5 min, respectively. The placenta was not delivered within 30 min after fetal delivery, thus requiring hysterectomy for PAS. Estimated blood loss was 1000 mL. The postoperative course was uneventful, and the patient and baby were discharged on the 8th postoperative day.

Part of the chorion and placenta were adhered to the uterus (Fig. 2a). The resected uterus was divided to 7 specimens in order to perform macroscopic and histopathological analyses. The surgical specimen showed a white polyp measuring 2 cm, which parted from the uterine fundus and the lower uterine segment (Fig. 2b). Histopathological examination of the tumor involving the lower uterine segment revealed endometrioid adenocarcinoma (Grade 1), with < 50% myometrial invasion and positive expression of estrogen and progesterone receptors, in addition to PAS (Fig. 3a and b). Notably, the tumor involving the uterine fundus did not show myometrial invasion. Histopathological findings were similar in both tumors located in the uterine lower segment and uterine fundus. A retrospective review of the MRI images obtained during pregnancy revealed the tumor involving the uterine fundus, although involvement of the lower uterine segment was difficult to detect (Fig. 3c). We performed a laparoscopic bilateral salpingo-oophorectomy and pelvic lymphadenectomy 102 days after cesarean hysterectomy and confirmed the absence of metastases. The tumor was a stage IA lesion based on the International Federation of Gynecology and Obstetrics system. Follow-up performed 4 years after cesarean hysterectomy revealed no recurrence.

Our case demonstrated the gross and histopathological findings, MRI findings, and clinical course of EC during pregnancy. To discuss this rare condition, we performed a literature review of cases of endometrioid carcinoma associated with pregnancy. We defined EC associated with pregnancy as diagnosed at delivery to within 15 months after pregnancy. We performed a search of PubMed, MEDLINE, and Scopus databases for the period between January 1995 and March 2019, using the following key words: “endometrial cancer”, “endometrial carcinoma”, “endometrioid cancer”, “endometrioid carcinoma”, “corpus cancer”, “pregnancy”, “abortion”, and “postpartum” in various combinations. We excluded non-English articles, discontinued journal and those published before 1995. We summarized the timing of diagnosis, outcome of EC, symptoms, diagnosis of histopathological examination, surgical stage (base on FIGO 2008) [4], and surgical treatment for EC. We also listed the authors’ opinions and discussions about the effect of pregnancy on the prognosis of EC.

A total of 18 studies with 25 cases of EC associated with pregnancy (including our case) have been reported [3, 5‐20]; 9 cases were identified postpartum up to the 14-month; 16 cases were diagnosed at the time of D&C for first-trimester spontaneous or elective abortion. These results suggest that clinicians should consider EC after pregnancy even though abnormal bleeding is often observed after pregnancy, and EC associated with pregnancy is rare. Our literature review revealed that there were no previous reports of a diagnosis of EC based on an examination of the resected uterus following cesarean hysterectomy for PAS [3, 7]. Although our case is extremely rare, clinicians should check the macroscopic finding of the resected uterus carefully regardless of the indication for hysterectomy.

Our literature review showed that the histopathological classification was endometrioid adenocarcinoma grade 1–2 in 23 of the 25 cases, unknown grade of endometrioid adenocarcinoma in 1 of the 25 cases, and poorly differentiated adenosquamous carcinoma in 1 of the 25 cases. Immunohistochemical (IHC) analysis was performed in 9 of the 24 cases and revealed a typical staining pattern as previously reported [21, 22]. Previous reports have shown that women younger than 45 years rarely developed EC, and the most common subtype of classification in younger women was endometrioid adenocarcinoma grade 1–2 [23, 24]. Although the number was limited, these results suggested that pregnancy did not affect the subtype and IHC staining pattern of EC. We considered that our literature review might be biased because we could include only published literature and cases that made it to the scientific publication stage and this condition might be under-reported; thus, this is the limitation of our study.

The case we presented is rare, and this report highlights several interesting points, as follows: 1. It describes the histopathological analysis of EC during pregnancy; 2. It describes a tumor involving the lower uterine segment and simultaneously the uterine fundus; and 3. It describes the MRI appearance of EC during pregnancy.

Histopathological examination of the specimen revealed EC that presented as a well-differentiated adenocarcinoma with a focal cribriform pattern, back-to-back structure, and a papillary area. Although IHC analysis showed positive expression of estrogen and progesterone receptors, our patient did not demonstrate any metastases, and no recurrence was observed 4 years after the cesarean hysterectomy. These features resemble those of typical grade 1 endometrioid adenocarcinoma [25‐27]. We concluded that the high-dose estrogen and progesterone condition during pregnancy did not promote progression of the EC. As shown in Table 1, most authors considered that pregnancy did not worsen the prognosis of EC. Further cases are expected to discuss how the pregnancy affects the prognosis of EC. Moreover, the histopathological and IHC findings in our case showed similar pattern to those of typical EC.

The reason for the presentation of separate tumors at the uterine fundus and lower uterine segment is unknown. Histopathological analysis of both tumors showed similar findings; thus, we concluded that the tumor presented as 2 separate growths at the aforementioned sites owing to enlargement of the uterus during pregnancy (although this remains speculative). Other possibilities considered were metastasis or multi-site involvement of EC. Myometrial invasion was insignificant; thus, we excluded metastasis as a possible etiology. The possibility of multi-site involvement of EC is difficult to exclude; however, the estimated frequency of this condition is low. Therefore, we concluded that the tumor separation could be attributed to the uterine enlargement during pregnancy.

MRI scans were retrospectively analyzed after the cesarean hysterectomy. We observed a lesion in the uterine fundus measuring approximately 3 cm in diameter with signal intensity similar to that of the placenta. Notably, this lesion was separate from the placenta. Clinicians must consider the possibility of EC in women with MRI scans showing such lesions during pregnancy.

In conclusion, our findings in this case suggest that careful analysis of MRI findings during pregnancy and gross examination of the resected uterus (in patients undergoing hysterectomy for obstetric complications) are essential, although EC during pregnancy is extremely rare. The literature review suggested that EC associated with pregnancy seemed to have a good prognosis.