Prognosis of uterine and extrauterine low-grade endometrial stromal sarcoma: an observational cohort study

Objective: Little is known about the survival differences between uterine and extrauterine low-grade endometrial stromal sarcoma (LGESS). Survival outcomes, consisting of disease-free survivals and overall survivals (OS), were compared in these two entities. Methods: From February 2012 to June 2019, all primary LGESS cases and LGESS cases with first recurrence in the study center were reviewed. The clinicopathological characteristics and survival outcomes of extrauterine and uterine LGESS patients were compared for both primary and recurrent diseases. Results: During the study period, 143 patients with primary LGESS and 56 patients with recurrent LGESS were included and followed up to 1 June 2020, among whom 8 (5.6%) and 10 (17.8%) patients were identified as having extrauterine LGESS. Patients with primary and recurrent extrauterine LGESS had similar clinicopathological characteristics to those of patients with uterine LGESS. In primary or in recurrent LGESS cases, in univariate analysis, patients with uterine and extrauterine LGESS had similar disease-free intervals after the last treatment, and they also had similar OSs after the diagnosis. Ovarian preservation led to significantly increased recurrence for primary LGESS [hazard ratio (HR) 4.9, 95% CI: 2.3–10.1, P<0.001) and repeated recurrence for recurrent LGESS (HR 3.1, 95% CI: 1.3–7.3, P=0.009). Surgical treatment for recurrent LGESS decreased repeated recurrence after the first recurrence (HR 0.2, 95% CI: 0.1–0.7, P=0.006). No factors were found to be associated with the OS of primary or recurrent LGESS. Conclusion: The clinical characteristics and survival outcomes of extrauterine LGESS are similar to those of uterine LGESS. Surgery is the treatment of choice for recurrent LGESS. Ovarian preservation is detrimental to disease-free survival but not to OS in both uterine and extrauterine LGESS.


Introduction
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, accounting for only 0.2% of female genital tract malignancies, but is the second most common uterine malignant mesenchymal tumor [1] .
LGESS had a favorable prognosis [2] , with a 5-year disease-specific survival rate of 90% for stage I or II disease and of 50% for stage III or IV disease [3,4] .Although less common, primary extrauterine LGESS, or low-grade endometrioid stroma sarcoma, namely, LGESS without uterus involved, has been studied in many case reports.However, the prognostic significance of extrauterine LGESS is not as well-known as that of uterine LGESS due to the scarcity of cases for study.Some authors have suggested that extrauterine LGESS may have a similar survival prognosis to its uterine counterpart [5] , but others suggest that it tends to have a worse prognosis [6] .Ovarian LGESS seems to have a much better prognosis than other primary ovarian sarcomas [7] .There is no

HIGHLIGHTS
• Extrauterine low-grade endometrial stromal sarcoma (LGESS) had similar clinical characteristics and survival outcomes compared with uterine LGESS in primary or in recurrent disease status.• Surgical treatment is the choice for all sorts of LGESS to achieve advanced disease-free survivals.• Ovarian preservation would cause detriment to diseasefree survival but not to overall survival.direct evidence demonstrating the prognostic differences between uterine and extrauterine LGESS.Recent studies either had very limited case samples [8,9] or simply described the extrauterine LGESS cases themselves [5,7] .These factors have limited an in-depth understanding of the oncological behavior of extrauterine LGESS.
In this observational cohort study, all patients diagnosed with LGESS from February 2012 to June 2019 in our study center and were reviewed and classified into the primary and recurrent LGESS cohorts.The clinical characteristics, treatment characteristics, and prognoses of extrauterine LGESS were compared to those of uterine LGESS as to characterize the specific survival outcomes of extrauterine LGESS.

Ethical approval
The Institutional Review Board from the study center has approved this observational 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.This work has been reported in accordance with STROCSS standards [10] .

Study design and participants
This is an observational cohort study.Patients diagnosed and treated for LGESS in one study center, from February 2012 to June 2019 were retrospectively included.LGESS is defined as a malignant stromal tumor with cells resembling proliferativephase endometrial stroma and displaying infiltrative (permeative) growth with or without lymphovascular invasion [1] .Primary extrauterine LGESS is defined as a low-grade mesenchymal neoplasm with a morphology resembling that of proliferativetype endometrial stroma without uterus involved [1] .
The main study outcomes were survival outcomes, and the secondary outcomes were fertility outcomes and factors relevant to survival and fertility outcomes.Follow-up of fertility and oncologic outcomes were carried out up to 1 June 2020.The prognosis of extrauterine LGESS was compared with that of primary uterine LGESS during the same periods; the prognosis of recurrent extrauterine LGESS was also compared with that of recurrent uterine LGESS in terms of survival outcomes after first recurrence.For primary LGESS cohort, only patients undergoing surgical treatment during the study period were included.For recurrent LGESS, all patients were included no matter whether they underwent surgical treatment or not.Patients with pathological findings rather than LGESS, even patients with mixed high-grade and low-grade endometrial stromal sarcoma, were excluded.As some patients developed recurrence during the study period, their survival outcomes were also included in the analysis for patients after first recurrence.
All clinicopathological data were collected from the case reports.Data on personal medical history and survival outcomes were collected by outpatient interviews.All histological findings of LGESS were strictly reviewed and confirmed by two pathologists (Y.Y. and H.W.).For extrauterine LGESS, close attention was paid to the examination of uterine histology.Efforts were made to ascertain the status of the uterus if this information was not obvious at the time the extrauterine foci were removed.If the patient had her uterus removed before diagnosis of extrauterine LGESS, the previous uterine pathology, if applicable, was comprehensively reviewed.If the uterus remained intact at first treatment, the uterine pathology after surgery for recurrence was checked.If uterine histology was not available, indirect evidences of uterine lesions, that is, histology from endometrial sampling and imaging evaluation, were reviewed together.

Interventions and measures
Data on surgical treatment, including surgical routes, retroperitoneal lymphadenectomy, and pathological results (FIGO staging, lymphovascular space invasion, metastasis to lymph nodes) were collected from the case and pathology reports.For recurrent LGESS disease, surgical routes and residual lesions reported by surgical records were also collected from the case reports.Attention was given to the chemotherapy regimens and courses, hormone therapy regimens and courses, and radiotherapy.Oncologic outcomes consisted of disease-free survival (DFS) since last treatment and overall survival (OS) since the diagnosis, which were followed up to 1 June 2020.Fertility outcomes consisted of conception rates and birth rates during the same follow-up periods.
In this study, ovarian preservation denotes preservation of at least one ovary with or without an intact uterus.All recurrences were confirmed by histological diagnosis by biopsy and/or repeated surgeries.All deaths were confirmed by certification of death.

Statistics
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 of specific clinicopathological characteristics by using nonparametric tests.Survival curves were generated with the Kaplan-Meier method, and proportional hazards models were used to estimate the hazard ratios (HRs), 95% CI and P values for the effects of treatment modalities on DFS, PFI, and OS.Unless otherwise stated, all analyses were performed with a two-sided significance level (P values) of 0.05 and were conducted with the use of the software Statistical Product and Service Solutions (SPSS) Statistics 20.0 (IBM Corporation).

Patient characteristics
The flow diagram is presented in Figure 1.During the study periods, among 143 patients with primary disease and 56 patients with recurrent disease were enrolled, 8 (5.6%) and 10 patients (17.8%) were identified as having extrauterine LGESS, respectively.The total number of patients with extrauterine LGESS was 15, since three patients with primary diseases experienced recurrences during the study period.The baseline characteristics of the patients with primary and recurrent disease are summarized in Table 1 and Table 2, respectively.The patients with extrauterine and uterine LGESS were well balanced in terms of almost all these characteristics for both primary disease and recurrent disease.However, due to the location of the disease, 10 and 4 patients with uterine and extrauterine LGESS underwent hysteroscopy and vaginal surgeries, respectively, which was unique in this population.All patients' data were supplied in Supplementary Table 1 (Supplemental Digital Content 1, http://links.lww.com/JS9/B820 and Supplementary Table 2, Supplemental Digital Content 2, http://links.lww.com/JS9/B821).
The clinicopathological characteristics of all 15 patients with extrauterine LGESS are listed in Table 3.No patient underwent myomectomy previously, by either laparoscopic or open surgery.In 10 cases, the uterus removed before the diagnosis of LGESS due to adenomyosis (two cases) or leiomyoma (eight cases).The median interval from hysterectomy to the diagnosis of LGESS was 93.5 (range 39-148) months.Two examples of vaginal LGESS were illustrated in Supplementary Figure 1 (Supplemental Digital Content 3, http://links.lww.com/JS9/B822).Among 10 patients with a hysterectomy history, 6 had a comprehensive review of the uterine histology, which had ruled out occult LGESS.Among five patients with an intact uterus at first treatment, one was proved of normal histology confirmed by pathological examination in specimens of recurrence, one patient underwent endometrial sampling at first treatment, and the pathological examination was confirmed normal; the other three had no available histology for review, and their extrauterine LGESS were diagnosed by surgical records, pathological reports, and imaging reviews.In total, 8 of 15 patients had direct evidence of extrauterine LGESS.Among seven patients without direct evidences of extrauterine LGESS were all in the recurrent cohort (LGE0008, LGE0010, LGE0013, LGE0014, LGE0016, LGE0018, and LGE0048).

Survival difference between uterine and extrauterine LGESS in primary cases
As shown in Figure 2A and B, in univariate analysis, extrauterine LGESS was not associated with increased risks of recurrence (HR 1.5, 95% CI: 0.4-4.8,P = 0.528) or mortality (HR 0.04, 0.0-not available, P = 0.769).As shown in Figure 3A and B, in this population, ovarian preservation was associated with an increased risk of recurrence (HR 4.9, 95% CI: 2.3-10.1,P < 0.001) but not with mortality (HR 1.1, 95% CI: 0.1-12.6,P = 0.916).The difference in recurrence remained after adjustment for the presence of extrauterine LGESS.No risk factor was found to be associated with OS.No patient with extrauterine LGESS achieved successful conception in this population.Among the four patients with vaginal LGESS, three underwent vaginal tumor resection, and two had disease relapse later.One patient with extrauterine LGESS (LGE0011) had fertility spared and attempted conception, but unfortunately, she experienced recurrence before successful conception.

Discussion
In this cohort study, we compared the clinical characteristics and survival outcomes of patients with uterine and extrauterine LGESS who had primary or recurrent disease.The comparisons revealed that the prognoses of extrauterine LGESS were similar to those of their uterine counterparts.To our knowledge, although this is not the largest cohort of extrauterine LGESS in the literature, it is the first to compare the survival outcomes of extrauterine LGESS to those of uterine LGESS with long-term follow-up.We also provide information for patients with recurrent disease for the first time.Since the tumor behavioral characteristics of extrauterine LGESS are similar to those of uterine LGESS, ovarian preservation is not recommended for older patients not desiring fertility preservation, especially in recurrent situations, as it would significantly increase the risk of recurrence [11,12] .As with uterine LGESS, surgical treatment is of great importance in the management of extrauterine LGESS, for both the primary and recurrent disease states.This study brings to light previously unreported prognostic features that add to the body of knowledge with respect to extrauterine LGESS and may aid in the clinical management of this rare disease.Primary extrauterine ESS is a rare entity and consists of lowgrade and high-grade types [5,6] .In our study, extrauterine disease made up 5.6% of cases in the primary LGESS cohort.Many case reports have discussed LGESS originating from the vagina [8,13,14] , cervix [15,16] , ovaries [5,7,17] , placenta [18] , stomach [19] , ileum [20] , colorectum [21][22][23] , omentum [24] , sciatic nerves [25] , retroperitoneal cavity [26] , and nonspecific sites within the pelvic cavity [27] .The most common sites involved are the ovaries [5] .Many cases may have nonspecific origin sites within the abdominal and/or pelvic cavity [5,6] .These findings suggest that the identification and differentiation of extrauterine LGESS are of clinical importance, as LGESS often has an indolent course with a better prognosis than other ovarian sarcomas, indicating the importance of a correct diagnosis [7] .
The differential diagnosis of extrauterine LGESS involves primarily metastatic uterine sarcoma of the same cell type.One fundamental element of good pathology practice, namely, awareness of the clinical history and knowledge of the uterine status, is crucial in this regard [7] and paramount in this distinction.
The characteristic gross finding of nodular infiltrative growth with vascular involvement as seen in uterine LGESS is rarely seen within the ovarian parenchyma [6,7] .A number of extrauterine LGESSs have the same genetic alterations as their uterine counterparts, suggesting that extrauterine LGESS has a histogenesis that is similar to that of uterine tumors [28] .Comprehensive pathological review could only be available when hysterectomy is performed, which is critical for young patients desiring fertility.When hysterectomy precedes removal of the extrauterine disease or if the uterus is left in situ, a definitive diagnosis may rely on review of prior pathology material, on review of imaging evaluations, on endometrial sampling if applicable, or on thorough sampling of the extrauterine tumor to assess histological features that could aid in the exclusion of a uterine primary [5] .However, these methods have a high risk of missed diagnosis of primary uterine disease.To rule out the possibility of metastasis from the uterus, we made efforts to review uterine histology, and all extrauterine LGESS patients had direct or indirect evidence supporting an uninvolved uterus.In our study, ovarian preservation led to an increased risk of recurrence, and no extrauterine LGESS patient with an intact uterus achieved successful conception.However, for primary uterine LGESS, fertility-sparing treatment has been reported in many case reports [29][30][31][32][33][34][35][36][37] .Therefore, uterinesparing treatment seemed inappropriate for patients suspected of extrauterine LGESS.
An unexpected location and unusual presentation of extrauterine LGESS may make the diagnosis challenging.The differential diagnosis of extrauterine LGESS with other neoplasms in the same sites is fraught with difficulties, particularly given the potential for less common histological patterns that mimic other entities.Additionally, one-quarter of cases could be misdiagnosed at initial treatment [5] .When a primary ovarian LGESS is determined, the differential diagnosis is most often with a sex cordstromal tumor [5,7] .The immunohistochemical [5] and genetic characteristics [11,12,38,39] of the tumor are very important for the differential diagnosis of this rare neoplasm.
Conditions, such as adenosarcoma and endometriosis, should be considered in the differential diagnosis when true glandular/epithelial differentiation is present.In our study, endometriosis and adenomyosis were noted in six (40.0%) and two (13.3%)patients with extrauterine LGESS, respectively.Many reports have suggested endometriosis as the origin of LGESS [9,16,20,22,25,[40][41][42][43] .Endometriosis was noted in almost half of extrauterine LGESS cases [5] , and endometriotic foci were identified adjacent to the tumor in almost half of ovarian LGESS cases [7] .A history of endometriosis might provide a clue for the differential diagnosis of extrauterine LGESS from other types of sarcomas, which lack such an association [44,45] .When there is an apparent lack of associated endometriosis, it is possible that the extrauterine LGESS originated from gland-poor foci of endometriosis (stromal endometriosis) or that overgrowth of the tumor has obscured the underlying endometriosis [5] .
The observational design and limited sample sizes of our study are its main limitations.Additionally, seven of 15 patients lacked direct evidence of a noninvolved uterus, which would weaken the study power.Due to the limited sample size, we could not sufficiently evaluate other important clinicopathological characteristics, such as tumor size, mitotic index, location, vascular invasion, and stages, in the context of survival outcomes.In a well-designed multicenter study, a stringent sample size estimation should be performed as to guarantee enough statistic power.We did not perform genetic testing although two thirds of LGESS harbor genetic fusions involving polycomb family genes [1] .We did not perform FIGO staging in the extrauterine LGESS, as Masand et al. [5] observed, the conditions of patients with multifocal disease were not worse than those of patients with single-site disease, suggesting that multifocal disease does not represent advanced-stage disease in extrauterine LGESS.The optimal adjuvant therapy after surgery remains unclear.Besides, histological characteristics could provide prognostic information for LGESS [46] .

Conclusions
Extrauterine LGESS had similar clinical characteristics and survival outcomes to those of uterine LGESS in both the primary or recurrent disease states, and surgical treatment is the choice for all types of LGESS to achieve improved survival.Ovarian preservation would be detrimental to DFS but not to OS.
Data Statement: The data that support the findings of this study are available from the corresponding author upon reasonable request.The data are not publicly available due to privacy or ethical restrictions.

Provenance and peer review
This paper was not invited.

Statement of submission
The paper is not under consideration by another journal, and the results presented in this work have not been previously presented or published.

Figure 1 .
Figure 1.Flow diagram of the study.LGESS, low-grade endometrial stromal sarcoma.* In our study, ovarian low-grade endometrioid stroma sarcoma was regarded as LGESS.

Figure 2 .
Figure 2. Survival outcomes of patients with low-grade endometrial stromal sarcoma by Kaplan-Meier analysis.(A) Disease-free survival of patients with primary extrauterine or uterine disease.(B) Overall survival (OS) of patients with primary extrauterine or uterine disease.(C) Progression-free interval of patients with recurrent extrauterine or uterine disease.(D) OS of patients with recurrent extrauterine or uterine disease.

Figure 3 .
Figure 3. Survival outcomes of patients with low-grade endometrial stromal sarcoma (LGESS) by Kaplan-Meier analysis.(A) Disease-free survival of primary LGESS patients with or without ovarian preservation.(B) Overall survival (OS) of primary LGESS patients with or without ovarian preservation.(C) Progression-free interval (PFI) of recurrent patients with or without ovarian preservation.(D) OS of recurrent patients with or without ovarian preservation.(E) PFI of recurrent patients with or without surgical treatment.(F) OS of recurrent patients with or without surgical treatment.

•
No patient with extrauterine LGESS achieved successful conception.State Key Laboratory for Complex, Severe and Rare Diseases, c Department of Pathology, Peking Union Medical College Hospital, d National Clinical Research Center for Obstetric and Gynecologic Diseases, Beijing and e Department of Obstetrics and Gynecology, the Second People's Hospital of Jingdezhen, Jingdezhen, People's Republic of China Sponsorships or competing interests that may be relevant to content are disclosed at the end of this article.
a Department of Obstetrics and Gynecology, b

Table 1
Epidemiological, surgical, and pathological characteristics between patients with primary LGESS.

Table 2
Epidemiological, surgical, and pathological characteristics between patients with recurrent LGESS.
a PFI denotes progression-free interval since last treatment for first recurrence.b OS denotes overall survival since the diagnosis of first recurrence.DFS, disease-free survival; LGESS, low-grade endometrial stromal sarcoma; OS, overall survival; PFI, progression-free interval.

Table 3
Detailed epidemiological characteristics, treatment and prognosis of patients with extrauterine LGESS.
a The normal histology was confirmed by pathological examination in specimen of recurrence.bThis patient underwent endometrial sampling at first treatment, and the pathological examination was confirmed normal.cThese histological results had not been reviewed and confirmed due to lack of available materials.d