Oncological and fertility outcomes following fertility-sparing strategies for endometrial hyperplasia and cancer in Singapore

INTRODUCTION Endometrial cancer (EC) is the most common gynaecological cancer in Singapore. Rising incidence of EC in younger women,[1] compounded by delayed childbearing,[2] has increased interest in fertility-sparing management for EC and its precursor, endometrial hyperplasia (EH). While hysterectomy with bilateral salpingectomy with or without bilateral oophorectomy, along with assessment of draining lymph nodes, is the standard initial treatment for EC and EH with atypia, this is not feasible for women who have not completed their family. Thus, international societal guidelines now include recommendations on fertility-preserving management of EC.[3,4] In our institution, patients with EC are considered for fertility-sparing treatment if they fulfil the following criteria: (a) presence of well-differentiated (International Federation of Gynecology and Obstetrics [FIGO] grade 1) endometrioid adenocarcinoma; (b) minimal or no myometrial invasion on magnetic resonance imaging of the uterus; (c) no extrauterine spread on imaging; (d) no suspicion of Lynch syndrome; (e) patient committed to long-term surveillance till definitive treatment; and (f) no contraindication to long-term use of progestins. Treatment of EC and EH with or without atypia consists of progestin in the form of levonorgestrel-releasing intrauterine system (LNG-IUS) and/or oral megestrol, with or without metformin. Transcervical resection of the endometrium may be considered in selected EC cases. Overweight patients are strongly encouraged to attend weight management clinics. Follow-up consists of endometrial biopsy at 3–6 months. If there is disease persistence or progression, definitive surgical management is advised. If regression to normal endometrium is noted, referral is made to the subfertility clinic for assisted reproduction if the patient is ready for pregnancy; otherwise, treatment is continued for at least one more year. Definitive surgery is strongly encouraged upon completion of childbearing. In this retrospective cohort study, we evaluated the oncological and fertility outcomes of patients with EC and EH who received fertility-sparing treatment. METHODS Patients undergoing fertility-sparing treatment for EC and EH with or without atypia (diagnosed histologically according to World Health Organization criteria)[5] were identified from the oncology tumour group database of the Department of Obstetrics and Gynaecology, National University Hospital, Singapore, between January 2015 and June 2023. Patients with at least 6 months follow-up and one repeat biopsy were included. De-identified information, including age, body mass index (BMI), histopathological diagnosis, presumed stage of disease based on scan results, treatment received, surveillance biopsy results, fertility and pregnancy outcome, was obtained from patients’ electronic medical records and analysed using Microsoft Excel (Microsoft Corp, Redmond, WA, USA). Ethical approval was obtained from the NHG Health Domain Specific Review Board (Ref: 2022/00203). Based on histopathological examination, response to treatment is classified as complete regression (CR) or partial regression (PR), the latter referring to cases with histological improvement in the specimen but where the endometrium has not returned to a normal state. Pregnancy rate is defined as the number of patients who became pregnant among those who actively tried to conceive after achieving CR. Live birth rate refers to the number of live births that occurred among those who actively tried to conceive after achieving CR. RESULTS A total of 37 women with EC and EH desiring fertility preservation were identified between January 2015 and June 2023. Median duration of follow-up was 30.3 months. Of these women, 20 had EC and 17 had EH. Sixteen of the patients with EC had stage IA grade 1 cancer and four had stage IA grade 2 cancer. Patients with grade 2 cancer were counselled on the deviation from standard guidelines, but they declined definitive treatment. Of the 17 patients with EH, ten had EH with atypia and seven had EH without atypia. The average age of the women with EC and EH was 33.0 (range 25–42) years and 33.5 (range 24–46) years, respectively. The cohort comprised 20 (54%) Chinese, ten (27%) Malays, three (8.1%) Indians and four (10.8%) patients of other ethnicities. Of the 37 patients, four had normal BMI (18.5–22.9 kg/m2), nine were overweight (BMI 23–27.4 kg/m2) and 24 were obese (BMI ≥27.5 kg/m2). Patient characteristics are summarised in Table 1, and treatment details are summarised in Table 2.Table 1: . Patient characteristics (N=37).Table 2: Treatment details.Of the seven EH patients without atypia, CR was noted in five patients. Median time taken to achieve regression was 13 months. Progression to cancer was noted in one patient who achieved an initial CR but was lost to follow-up without receiving maintenance therapy. Of the ten EH patients with atypia, CR was noted in nine patients. Median time to achieve CR was 9 months. Progression to cancer was noted in two of the nine patients after the initial CR, and time from CR to EC was 20 months and 39 months, respectively. Both patients lacked compliance to maintenance therapy. Among the 20 patients with EC, CR was observed in 17 patients, PR to EH was observed in two patients and EC persisted in one patient. Median time to achieve CR was 10 months. One patient with persistent EC was to undergo definitive surgical treatment as per the protocol. Of the two PR cases, one regressed to EH without atypia and continued to be on medical management. The other PR case regressed to EH with atypia in 16 months while on megestrol and levonorgestrel-releasing intrauterine system (LNG-IUS). Three months after PR, she progressed to metastatic poorly differentiated carcinoma with loss of MLH1 and PMS2 on immunohistochemical staining. At the time of this writing, she was undergoing systemic treatment for metastatic EC with plans for interval surgery. Recurrence to cancer was noted in four out of 17 patients after CR was achieved. The average time taken for recurrence was 8 months (range 3–20). Notably, one patient was diagnosed with stage IIIC serous uterine cancer with null p53 identified on immunohistochemistry studies. Table 3 shows a summary of the patient outcomes.Table 3: Summary of outcomes.Overall, 31 patients achieved CR in all groups, of whom eight had actively tried to conceive. Of the 17 patients with EC, five actively tried to conceive, while three underwent in-vitro fertilisation (IVF) but failed to conceive. Two patients conceived spontaneously, resulting in one live birth and one miscarriage. In the EH group (with and without atypia), three women underwent IVF; one was unsuccessful, one conceived but suffered a miscarriage, and one was awaiting embryo transfer at the point of data analysis. The overall pregnancy rate was 37.5%, and live birth rate was 12.5%. DISCUSSION This study demonstrates that conservative treatment of EC is feasible, with CR achieved in 85% of patients with EC and 82.4% of patients with EH. Recurrence after CR was observed in 23.5% of patients with EC, despite continued treatment. Our data are comparable to existing literature , which reports complete response rates of 75%–80%, disease recurrence rates of up to 40%, pregnancy rates of up to 50%, and live birth rates of 20%–30%.[6,7,8] In patients with EH with atypia, CR was achieved in 90% of patients, compared to 66.6% reported in another retrospective study performed in Singapore.[9] We opine that this could be related to the different strategies used in the medical management of atypical EH and EC. Lee et al.’s[9] study consisted of more patients managed with oral progestin alone, while most of our patients received Mirena as treatment [Table 2]. There is increasing evidence that Mirena has improved efficacy compared with oral progestin in the management of EH with atypia.[10] In our cohort, three out of eight (37.5%) women who tried to conceive achieved pregnancy, resulting in one live birth (12.5%). Although fertility preservation is a priority, only 25% of women (n = 8/31) in our cohort were ready for pregnancy at the time of CR. Reasons may include being medically unfit for pregnancy due to concomitant medical problems or personal reasons.[7] Majority (89%) of our patients were overweight. In obese women, a metabolic state that promotes oncogenesis exists; this is related to hyperoestrogenism, inflammation and insulin resistance, leading to multiple changes in oncogenic signalling pathways.[11] Obesity increases long-term mortality independent of EC, decreases fertility and IVF success[12] and is associated with poorer pregnancy outcomes for both the mother and fetus.[13] Weight loss plays a key role in sustaining treatment effect. The higher proportion of Malay patients (27%) observed in our cohort compared with the Singapore population census[14] may be partly due to a higher prevalence of obesity in the Malay population compared to the Chinese population.[15] Aside from the standard progestin treatment, there is emerging evidence regarding the use of metformin[16] and gonadotrophin-releasing hormone agonists (GnRHa).[17] Metformin demonstrates anticancer properties, inhibits PI3K/AKT/mTOR (phosphatidylinositol 3-kinase/protein kinase B/mammalian target of rapamycin) pathway, induces tumour suppressor genes, decreases estrogen receptor expression and increases progesterone receptor expression.[16,18] Meta-analyses suggest that hysteroscopic resection in combination with progestin results in higher rates of complete remission, lower recurrence rates and comparable pregnancy rates when compared to LNG-IUS with GnRHa/progestin and oral progestin alone.[19] Molecular classification should be considered in the selection of patients who are suitable for fertility-sparing management. The Cancer Genome Atlas classification divides EC into four subgroups with distinct genetic profiles and clinical outcome.[20] Molecular testing was not performed at the initial diagnosis of EC for two patients who developed advanced EC. They had null p53 phenotype serous carcinoma and deficient mismatch repair, respectively, associated with poorer prognosis. Progesterone and oestrogen receptor status is predictive of response to hormonal treatment[21] and should be incorporated during decision-making. This retrospective study is notable as the first of its kind in Southeast Asia. Its limitations include the single-institution design and the relatively small number of patients. In conclusion, fertility-sparing options in the management of EH and cancer are safe strategies if patients can strictly comply with clearly defined care pathways. Financial support and sponsorship Nil. Conflicts of interest There are no conflicts of interest.