Fertility-sparing treatment for low-grade endometrial stromal sarcoma.

The Institutional Review Board of the study center approved this retrospective study (No. SK-1289). All procedures performed in the study involving human participants were in accordance with the ethical standards of the institutional and National Research Committee and with the 1964 Declaration of Helsinki and its later amendments or comparable ethical standards. Specifically, before surgeries, all patients signed their informed consents, and realized the potential risk of sparing fertility. This was a retrospective cohort study. Patients who were diagnosed with and treated for primary LGESS at the study center from February 2012 to June 2019 were reviewed. Follow-up of fertility and oncologic outcomes was carried out up to June 1, 2020. The pathological diagnoses were reviewed and confirmed by pathologists of this study. The margin status of resected tumors was given special attention. Patients were excluded if they had repeated LGESS, had an ambiguous diagnosis or misdiagnosis of LGESS, or if they did not accept surgical treatment. Detailed epidemiological information, including age at diagnosis, menopausal status, gravidity, parity, and personal and family history of cancer, was collected by reviewing case reports and by interviewing patients via email and/or telephone. All patients underwent surgical treatment, with or without adjuvant chemotherapy, radiotherapy, and/or hormone therapy. Patients were divided into groups according to surgical entity: those with or without fertility-sparing surgery and those with hysterectomy with or without ovarian preservation. In this study, fertility-sparing surgery involves preservation of the uterus and at least one ovary. Patients who underwent fertility-sparing surgery also had records of regular menstruation in this study after all relevant treatments. Ovarian preservation refers to the preservation of at least one ovary with or without an intact uterus. However, not all patients undergo evaluations of the ovarian reserve after ovarian preservation. Data on surgical treatment, including surgical routes, retroperitoneal lymphadenectomy, and pathological results (FIGO staging, lymphovascular space invasion [LVSI], estrogen and progesterone receptor status, mitotic activity, and metastasis to lymph nodes), were also collected from the case and pathology reports. Ambiguous or missing pathological information was supplemented by pathologists of this study. The same attention was given to the chemotherapy regimens and courses, hormone therapy regimens and courses, and radiotherapy. For patients who undergo pelvic radiotherapy after the diagnosis of LGESS, treatment is not regarded as fertility sparing even if their ovaries are not resected by surgery. Oncologic outcomes consisted of DFS and OS, which were followed up during the study period. All recurrences were confirmed via histological diagnosis via biopsy and/or repeated surgeries. Biopsy was performed for intractable lesions by paracentesis or partial resection under direct vision. All deaths were confirmed by certification of death. The fertility outcomes consisted of conception rates and birth rates in the same follow-up periods. For patients who underwent fertility-sparing surgery, the trend in pregnancy attempts and reasons for not attempting pregnancy were confirmed by telephone and/or outpatient clinical interviews. Data on the interval from last treatment of hormone treatment to conception, conception methods, pregnancy weeks and complications were recorded in detail. Comparisons of continuous variables were conducted with parametric methods if assumptions of normal distribution were confirmed. Non-normally distributed variables and categorical data were compared between various groups with and without fertility-sparing surgery or ovarian preservation via nonparametric tests. Survival curves were generated via the Kaplan‒Meier method, and proportional hazards models were used to estimate the hazard ratios (HRs) and 95% confidence intervals (95% CIs) for the effects of fertility-sparing surgery and ovarian preservation on DFS and OS. Unless otherwise stated, all analyses were performed with a two-sided significance level of 0.05 and were conducted with the use of the software Statistical Product and Service Solutions (SPSS) Statistics 20.0 (IBM Corporation, Armonk, NY, USA).

The flow diagram is presented in Fig.  1 . In total, 143 eligible patients with a definite diagnosis of LGESS were reviewed, and 135 patients with primary uterine LGESS were included in the study. The raw data of this study are shown in Supplementary Table S1. One hundred and two (75.6%) patients had stage I disease. Hormone treatment was utilized by 61 (45.2%) patients. Twenty-one (15.6%) and 114 (84.4%) patients accepted fertility-sparing and nonfertility-sparing surgery, respectively, and 42 (31.1%) and 93 (68.9%) patients accepted ovarian preservation and no ovarian preservation, respectively. The baseline characteristics of the patients in the various groups are summarized in Table  1 . Patients who underwent fertility-sparing surgery or ovarian preservation were significantly younger (both p  < 0.001) and had lower gravidity or parity (all p values < 0.005). Eighteen patients underwent retroperitoneal lymphadenectomy, and only one (5.6%) had lymph node metastasis. Ten patients in the fertility-sparing groups underwent only hysteroscopy to remove the lesions. Hormone treatment was utilized by 12 of 21 (57.1%) patients who underwent fertility-sparing surgery and 23 of 42 (54.8%) patients who underwent ovarian preservation. Three patients selected the LNG–IUS for long-term treatment. Fig. 1 Flow diagram of the study. LGESS, low-grade endometrial stromal sarcoma Table 1 Epidemiological and surgical–pathological characteristics of patients with or without fertility-sparing surgery and ovarian preservation Fertility sparing p Ovarian preservation p No (n = 114) Yes (n = 21) No (n = 93) Yes (n = 42) Age at diagnosis (years), mean ± SD 43.1 ± 8.7 29.1 ± 7.5  < 0.001 44.6 ± 8.2 32.7 ± 8.3  < 0.001 Menopause, n (%) 16 (14.0) 0 (0.0) 0.056 15 (16.1) 1 (2.4) 0.016 Gravidity, mean ± SD 2.4 ± 1.9 1.0 ± 1.3 0.002 2.5 ± 1.9 1.4 ± 1.8 0.002 Parity, mean ± SD 1.1 ± 0.9 0.5 ± 0.7 0.004 1.2 ± 0.8 0.7 ± 0.9 0.005 Personal cancer history, n (%) 3 (2.6) 0 (0.0) 0.600 3 (3.2) 0 (0.0) 0.324 Family cancer history, n (%) 15 (13.2) 1 (4.8) 0.246 12 (12.9) 4 (9.5) 0.403 FIGO stages, n (%) 0.279 0.275 I/II 105 (92.1) 18 (85.7) 84 (90.3) 40 (95.2) III/IV 9 (7.9) 3 (14.3) 9 (9.7) 2 (4.8) Surgical routes, n (%)  < 0.001  < 0.001 Hysteroscopy 0 (0.0) 10 (47.6) 0 (0.0) 10 (23.8) Laparoscopy 58 (50.9) 5 (23.8) 43 (46.2) 20 (47.6) Laparotomy 56 (49.1) 6 (28.6) 50 (53.8) 12 (28.6) Retroperitoneal lymphadenectomy, n (%) 17 (14.9) 1 (4.8) 0.185 15 (16.1) 3 (7.1) 0.155 Max diameter of the tumor (mm), mean ± SD 66.3 ± 28.0 (n = 86) 66.0 ± 34.8 (n = 16) 0.971 66.6 ± 28.3 65.4 ± 31.1 0.845 LVSI, n/n (%) 24/99 (24.2) 1/21 (4.8) 0.035 20/80 (25.0) 5/40 (12.5) 0.112 Positive estrogen receptor, n (%) 103 (90.4) 20 (95.2) 0.470 85 (91.4) 38 (90.5) 0.546 Positive progesterone receptor, n (%) 98 (86.0) 18 (85.7) 0.600 79 (84.9) 37 (88.1) 0.626 Mitotic activity > 5 per 10 high power field, n (%) 24 (21.1) 6 (28.6) 0.307 22 (23.7) 8 (19.0) 0.551 Metastasis to retroperitoneal lymph nodes, n/n (%) 1/17 (5.9) 0/1 (0.0) 0.944 0/15 (0.0) 1/3 (33.3) 0.167 Hormone treatment, n (%) 49 (43.0) 12 (57.1) 0.231 38 (40.9) 23 (54.8) 0.133 Hormone treatment period (months), mean ± SD 10.2 ± 6.1 (n = 49) 19.8 ± 26.3 (n = 12) 0.022 11.1 ± 6.4 (n = 38) 13.7 ± 19.8 (n = 23) 0.453 Chemotherapy, n (%) 9 (7.9) 1 (4.8) 0.518 8 (8.6) 2 (4.8) 0.346 Radiotherapy, n (%) 15 (13.2) 0 (0.0) 0.068 15 (16.1) 0 (0.0) 0.003 Recurrence, n (%) 18 (15.8) 10 (47.6) 0.002 10 (10.8) 18 (42.9)  < 0.001 Recurrence beyond pelvic cavity, n/n (%) 4/18 (22.2) 4/10 (40.0) 0.284 3/10 (30.0) 5/18 (27.8) 0.615 Disease free survival (months), median (range) 41 (5–98) 26 (6–90) 0.161 45 (5–98) 26 (5–90) 0.005 Mortality, n (%) 3 (2.6) 0 (0.0) 0.600 2 (2.2) 1 (2.4) 0.676 Overall survival (months), median (range) 48 (10–98) 55 (21–94) 0.520 48 (10–98) 54 (12–94) 0.955 LVSI, lymphovascular space invasion. SD, standard deviation Flow diagram of the study. LGESS, low-grade endometrial stromal sarcoma Epidemiological and surgical–pathological characteristics of patients with or without fertility-sparing surgery and ovarian preservation LVSI, lymphovascular space invasion. SD, standard deviation A review of histological pathology revealed similar proportions of patients with advanced FIGO stages (III/IV), estrogen or progesterone receptor status, and mitotic activity and similar tumor diameters between patients with and without fertility-sparing treatment or ovarian preservation (Table  1 ). However, nonfertility-sparing patients had a greater proportion of positive LVSI than fertility-sparing patients did ( p  = 0.035). Among the 21 fertility-sparing patients, marginal status was available for evaluation in 16 patients and was negative. Five patients had no available margin status for evaluation, because fractured tissue was resected by hysteroscopy (Table  2 ). Table 2 Fertility and oncologic outcomes of patients with fertility-sparing surgery No Recurrence Recurrent sites DFS (months) OS (months) Age at diagnosis Stages Surgical routes Margin status Adjuvant therapy Attempting pregnancy Interval to conception* LGE0001 Yes Uterus, lung 39 88 26 years I Laparotomy Negative PEI Yes, and succeeded once 12 months LGE0002 Yes Uterus, lung 72 94 28 years IA Hysteroscopy N/A Progestin Yes, and succeeded once 14 months LGE0004 Yes Retrospective lymph nodes 19 69 25 years I Laparotomy Negative None Yes, and succeeded twice 10 months LGE0006 Yes Pelvic and peritoneal cavity 23 43 29 years IIIB Laparotomy Negative None Yes, but failed - LGE0007 Yes Pelvic cavity 9 66 37 years IIB Laparotomy Negative None No, during hormone treatment - LGE0019 Yes Uterus 19 43 35 years I Laparoscopy Negative GnRHa Unmarried - LGE0020 Yes Pelvic and peritoneal cavity 17 53 39 years IIIB Laparotomy Negative Progestin No - LGE0021 Yes Pelvic cavity 6 58 16 years IC Laparotomy Negative Progestin Unmarried - LGE0022 Yes Uterus 26 77 35 years I Hysteroscopy N/A None No - LGE0023 Yes Uterus 58 58 20 years I Hysteroscopy N/A None Unmarried - LGE0063 No - 57 57 33 years I Hysteroscopy Negative GnRHa Yes, and succeeded once 11 months LGE0064 No - 25 25 33 years I Laparoscopy Negative GnRHa/AI Yes, and succeeded once 10 months LGE0065 No - 55 55 26 years I Hysteroscopy Negative Progestin Yes, and succeeded once 20 months LGE0066 No - 21 21 20 years I Hysteroscopy Negative Progestin Yes, and succeeded once 10 months LGE0067 No - 66 66 37 years I Hysteroscopy N/A None Yes, but failed - LGE0073 No - 55 55 33 years I Laparoscopy Negative GnRHa/LNG–IUS No, during hormone treatment - LGE0074 No - 90 90 15 years I Hysteroscopy Negative GnRHa/LNG–IUS Unmarried - LGE0075 No - 25 25 37 years I Hysteroscopy N/A LNG–IUS No - LGE0076 No - 37 37 25 years I Hysteroscopy Negative Progestin Unmarried - LGE0077 No - 29 29 23 years I Laparoscopy Negative None Unmarried - LGE0078 No - 25 25 40 years I Laparoscopy Negative None No - AI, aromatase inhibitor. DFS, disease-free survival. GnRHa, gonadotropin-releasing hormone agonist. LNG–IUS, levonorgestrel intrauterine system. OS, overall survival. PEI, chemotherapy with cisplatin, etoposide and ifosfamide Fertility and oncologic outcomes of patients with fertility-sparing surgery AI, aromatase inhibitor. DFS, disease-free survival. GnRHa, gonadotropin-releasing hormone agonist. LNG–IUS, levonorgestrel intrauterine system. OS, overall survival. PEI, chemotherapy with cisplatin, etoposide and ifosfamide Comparisons of survival outcomes across various groups are presented in Fig.  2 and Supplementary Figure S1. The HRs calculated via proportional hazards models are listed in Supplementary Table S2. After a median follow-up time of 38.5 (range 5–98) months, no patient was lost to follow-up during the study period; 10 (47.6%) and 18 (15.8%) recurrences occurred in the fertility-sparing and nonfertility-sparing groups ( p  = 0.002), and 18 (42.9%) and 10 (10.8%) recurrences occurred in the ovarian preservation and nonovarian preservation groups ( p  < 0.001), respectively. Specifically, according to the Kaplan‒Meier analysis, between patients who preserved the ovaries but underwent hysterectomy (21 patients) and those who had both the uterus and ovaries removed (93 patients), there were significant difference in DFS (HR 0.2 with the former as reference, 95% CI 0.1–0.6, p  = 0.002) but no significant difference in OS (HR 0.4, 95% CI 0.04–4.5, p  = 0.464). Fig. 2 Survival outcomes of patients with and without fertility-sparing surgery or ovarian preservation according to Kaplan‒Meier analysis . A Disease-free survival (DFS) of patients with and without fertility-sparing surgery ( p  = 0.001). B DFS of stage I patients with and without fertility-sparing surgery ( p  = 0.034). C Overall survival (OS) of patients with and without fertility-sparing surgery ( p  = 0.440). D DFS of patients with and without ovarian preservation ( p  < 0.001). E DFS of stage I patients with and without ovarian preservation ( p  = 0.007). F OS of patients with and without ovarian preservation ( p  = 0.922) Survival outcomes of patients with and without fertility-sparing surgery or ovarian preservation according to Kaplan‒Meier analysis . A Disease-free survival (DFS) of patients with and without fertility-sparing surgery ( p  = 0.001). B DFS of stage I patients with and without fertility-sparing surgery ( p  = 0.034). C Overall survival (OS) of patients with and without fertility-sparing surgery ( p  = 0.440). D DFS of patients with and without ovarian preservation ( p  < 0.001). E DFS of stage I patients with and without ovarian preservation ( p  = 0.007). F OS of patients with and without ovarian preservation ( p  = 0.922) According to the Kaplan‒Meier analysis, fertility-sparing surgery (Fig.  2 A, B) or ovarian preservation (Fig.  2 D, E) resulted in a significantly increased recurrence risk in the whole cohort (HR 3.5 and 5.2, 95% CI 1.6–7.6 and 2.4–11.3, p  = 0.002 and < 0.001, respectively) and in stage I patients (HR 2.6 and 3.1, 95% CI 1.0–6.4 and 1.3–7.4, p  = 0.041 and 0.011, respectively). No other factors, including age group, menopausal status, FIGO stage, pathological characteristics, retroperitoneal lymphadenectomy, or adjuvant therapy, were significantly associated with recurrence risk (Figure S1). As a continuous variable, age at diagnosis was not significantly associated with recurrence (HR 0.8, 95% CI 0.4–1.6; p  = 0.576). Menopausal status seemed to have a protective effect against recurrence, but the effect was marginal ( p  = 0.100). Notably, three patients utilized the LNG–IUS and remained free of disease during the follow-up periods. Three of the 5 patients without available margins experienced recurrences. No factor was found to be associated with increased mortality risk. In particular, neither fertility-sparing surgery (Fig.  2 C, HR 0.04 [95% CI 0.0–not available], p  = 0.626) nor ovarian preservation (Fig.  2 F, HR 1.1 [95% CI 0.1–12.4], p  = 0.922) increased mortality. Among patients who underwent ovarian preservation, fertility-sparing surgery was not associated with increased recurrence or mortality ( p  = 0.810 and 0.285, respectively; Supplementary Figured S2 and Supplementary S3). Multivariate analysis was not performed because of the limited number of endpoint events and meaningful statistical power. The specific fertility and oncologic outcomes of the patients who underwent fertility-sparing surgery are listed in Table  2 . The mean age of the patients who underwent fertility-sparing surgery was 29.1 (± 7.5) years. Twelve out of 21 (57.1%) patients did not attempt pregnancy after their surgeries, since six patients had no fertility requirement due to being unmarried, 2 were still undergoing treatment, and 4 wanted only fertility spared. Among the 9 patients who attempted pregnancy, 7 achieved 8 live births, and 2 did not. No miscarriage occurred. Seven births were achieved by natural conception, and one live birth was achieved by in vitro fertilization and embryo transfer. The median interval from the last treatment to conception was 11 months (range 10–20 months). Severe complications during pregnancy included preterm birth (36 weeks) and hypertensive disease in two patients. All live deliveries were performed by cesarean sections. No death occurred and no other severe adverse events occurred during the pregnancy puerperium in mothers and infants. Previously receiving hormone treatment or not had no impact on the success of pregnancy (5/12 [41.7%] vs 2/9 [22.2%], p = 0.324). In general, fertility-sparing treatment was associated with increased recurrence. Ten patients experienced recurrence after fertility-sparing treatment, 4 (40.0%) had distant metastasis beyond the pelvic cavity, and only 3 (30.0%) cases of recurrence were limited to the uterus. All three patients with Stage III disease experienced recurrence without successful conception. Postrecurrence treatments for patients who underwent fertility-sparing surgery are listed in Supplementary Table S3. After recurrence, seven patients underwent nonfertility-sparing surgeries and remained free of disease, and 3 patients who underwent repeated fertility-sparing surgeries experienced disease relapse and underwent nonfertility-sparing surgeries again, after which they all remained disease free during the follow-up period. No patients who underwent fertility-sparing surgery died during the follow-up period. Unfortunately, among the 6 single (unmarried) patients who underwent fertility-sparing surgery, 3 experienced local recurrences within the uterus and underwent nonfertility-sparing surgeries. Specifically, one patient (Supplementary Table S3) was diagnosed with first recurrence during her cesarean section for the second live birth, when metastasis to retroperitoneal lymph nodes was noticed. She subsequently underwent repeated fertility-sparing surgeries (resection of metastatic lymph nodes) without adjuvant therapy. However, 5 months after the surgery, recurrence in the uterus was noticed, and the patient underwent nonfertility-sparing surgeries and remained disease free during the follow-up period.

In this study, we report the fertility and oncological outcomes of LGESS patients with and without fertility-sparing surgery or ovarian preservation. To our knowledge, this is the largest cohort to address fertility-sparing surgery in LGESS patients. The evidence in this study suggests that either fertility-sparing surgery or ovarian preservation significantly increases the recurrence risk of LGESS. Among the 9 patients who attempted pregnancy, 7 achieved 8 live births. The results are promising for patients desiring fertility. However, patients should be informed of the 3.5-fold increased risk of recurrence and that recurrence would occur in nearly half of patients with or without successful live births, although fertility-sparing treatment was not associated with an increased mortality risk. Besides, although ovarian preservation increased the recurrent risk, but had no significant association with the mortality risk, just as we had discovered in other LGESS cohort [ 8 ]. Little is known about the prognostic factors in LGESS [ 9 ]. Owing to the infrequency of LGESS, very few studies have explored the risk factors for recurrence in this population [ 10 , 11 ], and none have specified the prognostic role of fertility-sparing treatment. For the first time, we found that fertility-sparing treatment significantly increased the risk of recurrence. The potential reasons may be related to ovarian preservation in this population. Despite these controversies [ 2 , 12 , 13 ], ovarian preservation has been proven to increase the risk of LGESS recurrence [ 14 ], which is further supported by the findings of a systematic review [ 15 ]. Both estrogen and tamoxifen treatment are contraindicated in patients with LGESS, since they also increase the risk of recurrence [ 16 , 17 ]. Menopause is a protective factor associated with improved progression-free survival among patients with LGESS [ 18 ]. In addition, numerous reports have revealed the role of endometriosis in the pathogenesis of extrauterine LGESS [ 19 ]. These findings suggest that LGESS is a highly estrogen-dependent disease, and this knowledge is essential for decision-making before surgical therapies. For patients not desiring fertility, ovarian preservation leads to a greater risk of recurrence and, therefore, is not practical in terms of effectiveness, especially for older patients or patients not desiring fertility. Owing to the limited sample size, currently, no evidence exists regarding the quality of life after oophorectomy in premenopausal patients. However, it is prudent not to offer estrogen therapy to survivors of LGESS and women with granulosa cell tumors of the ovaries [ 20 ]. In this study, neither ovarian preservation nor fertility-sparing treatment had an adverse effect on OS. This finding has been confirmed by other studies [ 15 ], including a National Cancer Database analysis [ 21 ]. The lack of survival difference could be explained by the good prognosis of patients with isolated tumor relapse and the strict follow-up in such a cohort. A currently published report suggested that age and marital status are important parameters in the nomogram for predicting the OS of patients with LGESS. In their study, ovarian preservation was associated with an even lower risk of mortality than no ovarian preservation [ 22 ]. The discrepancy in the recurrence and mortality risks led to different recommendations in the relevant guidelines [ 23 – 25 ]. In our view, ovarian preservation is safe in terms of long-term survival and is appropriate for patients requiring fertility preservation. However, since multivariate analysis was not applied in this study, it should be acknowledged that the impact of FSS on DFS or OS cannot be further evaluated when considering multiple factors simultaneously. During the surgical counseling, cautions should be provided to patients, that due to the limited sample size, the oncological risk of conservative treatment consisting of ovarian preservation and/or fertility sparing should be carefully considered. We suggest that once a patient achieves a live birth, salpingo-oophorectomy could be a practical recommendation to decrease further potential recurrence. Little is known how to improve the fertility-sparing outcomes. In this study, previously receiving hormone treatment or not had no impact on the success of pregnancy (5/12 [41.7%] vs 2/9 [22.2%], p = 0.324). This finding risked more bias than definiteness. The impact of hormone treatment and its protocols, courses should be explored in studies of larger samples. This study supports hysteroscopy as an effective fertility-sparing treatment modality for LGESS. In the fertility-sparing treatment cohort in this study, ten patients had their lesions removed only by hysteroscopy. Among these 10 patients, 7 (70%) remained disease free after surgery, and 4 of 5 (80%) successfully achieved live births. Previous case reports have also supported the diagnostic and treatment role of hysteroscopy for LGESS [ 6 , 26 – 28 ]. The selection of eligible patients with comprehensive imaging and pathological evaluation is critical for the successful management of fertility-sparing treatment by hysteroscopy. However, hysteroscopy has the limitation of being unable to evaluate margin status because of the presence of fractured tissues, while a positive resection margin is an independent adverse factor for relapse [ 4 ]. Half of the patients who underwent hysteroscopy had no available margin status for evaluation, and 3 of the 5 patients experienced recurrence, which was likely associated with positive margins. Unfortunately, three of the six single (unmarried) patients (50.0%) in this study experienced recurrence after fertility-sparing surgeries. Effective sustained treatment is urgently needed for such situations. Adjuvant hormonal treatment has been considered a feasible adjuvant therapy for reducing the recurrence risk of patients with LGESS while bearing little benefit to OS [ 29 ]. In this study, three patients utilized the LNG–IUS and remained free of disease. The LNG–IUS design can achieve long-term local control by slowly releasing levonorgestrel, which is effective in the conservative treatment of adenomyosis [ 30 ], leiomyoma [ 31 ], and endometrial precancerous or cancerous diseases [ 32 ]. It may have protective effects on ovarian cancer [ 33 ]. Most adverse events associated with the LNG–IUS are mild and acceptable [ 30 ]. However, its use in the treatment of LGESS has not been reported in the English-language literature. This study is the first to report the possible effectiveness of sustained therapy in this population, which has provided a promising method for LGESS patients desiring fertility sparing but unable to be conceptive. The retrospective design and limited sample size of this study are its main limitations. The prognostic roles of other important clinicopathological parameters, including age, menopausal status, stages, and adjuvant therapies, are undetermined. We could not perform multivariate analysis to investigate the prognostic role of fertility-sparing treatment because of the limited number of endpoint events and meaningful statistical power. As described previously [ 13 ], long-term follow-up is essential for LGESS, a rare and indolent disease. Most importantly, how to select eligible patients for fertility-sparing treatment remains unresolved. As shown in this study, advanced-stage patients all experienced relapse without successful conception. However, other conditions had little supportive evidence due to the limited sample sizes. The conclusions regarding overall survival/mortality are based on only 3 deaths that occurred in the entire cohort during the follow-up period, which would limit the ability to make statistically robust conclusions about overall survival or mortality, or at least that any conclusions here should be interpreted with caution given the small number of deaths that occurred.

In this retrospective cohort study, fertility-sparing surgery or ovarian preservation in patients with LGESS resulted in a significantly increased risk of recurrence but had no impact on mortality. Fertility-sparing treatment had promising results in terms of live births. Hysteroscopy and the LNG–IUS may provide less invasive and sustained treatment for LGESS.

Low-grade endometrial stromal sarcoma (LGESS) is a malignant tumor composed of cells resembling stromal cells of the proliferative-phase endometrium, displaying permeative, infiltrative growth into the myometrium and/or Iymphovascular spaces [ 1 ]. LGESS represents < 1% of all uterine malignancies but is the second most common uterine malignant mesenchymal tumor and has a favorable prognosis [ 2 ]. The 5-year disease-specific survival rates are 90% for stage I or stage II disease and 50% for stage III or IV disease [ 3 ]. Hysterectomy with salpingo-oophorectomy is the standard treatment for LGESS. Ovary-sparing procedures and myomectomy are adverse factors for relapse [ 4 ]. However, the age range at diagnosis of LGESS is broad, and many patients are of reproductive age and desire pregnancy. For primary uterine LGESS, the fertility and oncological outcomes of fertility-sparing procedures remain ambiguous, since fertility-sparing treatment has been reported only in case reports [ 5 , 6 ]. In addition, these reports did not compare patients who did not undergo fertility-sparing procedures and had no specific information on subsequent obstetric outcomes. Limited evidences form a systemic review suggested that sparing fertility in women with LGESS would cause a relatively high risk of tumor relapse, though it does not increase the risk of death[ 7 ]. These limitations restrict effective decision-making for young patients who desire fertility preservation after the diagnosis of LGESS, especially early stage patients. In this retrospective cohort study, we reviewed all LGESS cases diagnosed and treated in the last 7 years at the study center to explore the prognostic role of fertility preservation in these patients. The primary objectives were to explore the disease-free survival (DFS) and overall survival (OS) rates after surgery for LGESS, and the secondary objective was to explore the fertility outcomes of these patients.

Additional file1 Additional file2 Additional file3 Additional file4 Additional file5 Additional file6 Additional file1 Additional file2 Additional file3 Additional file4 Additional file5 Additional file6