Article Text
Abstract
Objective To evaluate the effect of ovarian preservation on oncologic outcomes for women with low-grade endometrial stromal sarcoma of the uterus.
Methods A systematic search of the Medline, Embase, Cohrane, and Web of Science databases was performed based on the Preferred Reporting Items for Systematic Reviews and Meta-analyses guidelines. Studies including patients with low-grade endometrial stromal sarcoma who had hysterectomy were identified. Data on tumor recurrence and death rate were pooled using a random effects model.
Results A total of 17 studies met the inclusion criteria and reported on 786 patients. Based on available information, ovarian preservation was noted in 190 patients while 501 had bilateral salpingo-oophorectomy. A significantly increased tumor recurrence rate was observed in the ovarian preservation group (89/190, 46.8%) compared with the bilateral salpingo-oophorectomy group (121/501, 24.2%) (OR 2.70, 95% CI 1.39 to 5.28). Based on data from 162 patients, no difference in death rate was noted between the ovarian preservation (2/34, 5.9%) and bilateral salpingo-oophorectomy (9/128, 7%) groups (OR 0.80, 95% CI 0.18 to 3.47).
Conclusions Approximately one-quarter of patients with low-grade endometrial stromal sarcoma were managed with ovarian preservation. These women experienced a higher recurrence rate. Hormone exposure may be responsible for this elevated risk. Given the apparent high salvage rate, however, ovarian preservation may be an option only in a well-informed patient population.
- sarcoma
- uterus
- endometrial stromal sarcoma
- low grade
- ovarian preservation
Statistics from Altmetric.com
Highlights
Approximately one in four patients with low-grade endometrial stromal sarcoma are managed with ovarian preservation.
Ovarian preservation is associated with an increased risk of tumor relapse.
Based on limited evidence, it may not result in an increased risk of death.
Introduction
Uterine sarcoma is a rare group of tumors arising from mesenchymal cells and represents approximately 3% of all uterine cancers.1 Low-grade endometrial stromal sarcoma is the second most prevalent uterine sarcoma. Microscopically it consists of well-differentiated endometrial stromal cells with mild nuclear atypia.1 On presentation, low-grade endometrial stromal sarcoma is typically confined to the uterus and exhibits an indolent behavior.2 Five- and 10-year survival rates above 90% have been reported for women with stage I disease.3 4
Surgical management of low-grade endometrial stromal sarcoma includes hysterectomy with or without bilateral salpingo-oophorectomy. The value of lymphadenectomy and adjuvant treatment has yet to be established.2 Low-grade endometrial stromal sarcoma tumors consistently express estrogen and progesterone receptors. Therefore, bilateral salpingo-oophorectomy serves to remove occult ovarian micrometastases and also eliminates the endogenous production of hormones that may stimulate tumor growth.5 As more than half of women diagnosed with low-grade endometrial stromal sarcoma are pre-menopausal, however, surgical menopause may have long-term implications.1 To date, only small retrospective studies have evaluated the safety of ovarian preservation in women with low-grade endometrial stromal sarcoma, often reaching opposite results, making their use for the basis of management recommendations challenging.3–21 In contrast, the safety of ovarian preservation in pre-menopausal women with International Federation of Gynecology and Obstetrics grade I endometrioid adenocarcinoma of the endometrium, another hormonally responsive tumor, has been extensively evaluated and is currently offered to carefully selected patients.22 23
The aim of this comprehensive systematic review and meta-analysis is to determine the effect of ovarian preservation compared with bilateral salpingo-oophorectomy in women with low-grade endometrial stromal sarcoma on oncologic outcome.
Methods
Search strategy
This systematic review was conducted in accordance with the Preferred Reporting Items for Systematic Review and Meta-Analyses guidelines24; a protocol was determined beforehand by participating authors. The present study was registered with the PROSPERO Registry with the unique identifying number CRD42018082378.
A comprehensive electronic search from January 1, 1970 to May 28, 2018 of Pubmed/Medline, EMBASE, Cochrane, and Web of Science databases was performed with the use of the following keywords: “(Uterus or Uterine) AND ((endometrial stromal sarcoma) OR (stromal sarcoma) OR ESS)”. In addition, all references of included studies were systematically hand-searched.
Eligibility of studies
Our investigation was limited to complete articles available in the English language. Inclusion criteria were: (1) studies on women with low-grade endometrial stromal sarcoma who had hysterectomy; (2) studies with at least five cases; (3) studies providing data on the performance of oophorectomy; and (4) studies with data on recurrence rate and/or overall survival. Exclusion criteria were: (1) studies on high-grade or undifferentiated endometrial stromal sarcoma; (2) studies reporting on all histologic subtypes of uterine sarcoma; (3) studies on women who had morcellation of incidental sarcomas; (4) database-based, overlapping, or republished studies; (5) studies on fertility-preserving surgery or patients who did not undergo hysterectomy; and (6) studies in which the number of events (deaths or recurrences) could not be calculated from the reported data.
Data extraction
The methodological quality of the studies was assessed using the Newcastle–Ottawa Quality Assessment Scale for non-randomized studies.25 From each eligible study the following parameters were extracted when applicable: general study characteristics (author, country of origin, date of publication, years of recruitment), patient demographics (age, percentage of pre-menopausal women, race), performance of lymphadenectomy, adjuvant treatment (radiation therapy, chemotherapy, hormonal therapy), recurrence, and death rate. If the required data for the pooled analysis (ie, number of events) were not readily available in the published articles, the corresponding authors were contacted. In our study we defined hormone replacement as the use of estrogen with or without progesterone to replace hormones lost by bilateral salpingo-oophorectomy and hormone therapy as the use of selective estrogen receptor modulators, aromatase inhibitors, and/or progestins to treat the neoplastic process.
Synthesis of data
The level of statistical heterogeneity was evaluated with the I 2 statistic. A random effects model was used to compare outcomes between the two groups and OR and 95% CI were calculated with the DerSimonian and Laird approach. Forest plots were created for each comparison while graphical funnel plots were generated to determine the presence of publication bias by visual inspection. P values <0.05 were considered statistically significant. If significant heterogeneity was noted, a sensitivity analysis was performed by the sequential omission of each study. Statistical analysis was performed with the Cochrane Review software (Review Manager version 5.2).26 The present study was exempt from Institutional Board Review.
Results
A total of 17 studies reporting on 786 patients met the inclusion criteria.3 5–20 Supplemental figure 1 depicts the PRISMA flow diagram of the search strategy. All studies were retrospective in nature and consisted of 16 cohort studies and one case-control study. The majority originated from North America (eight studies) followed by Asia (five studies) and Europe (four studies). No publication bias was present as assessed by funnel plots.
Median age at diagnosis ranged from 41 to 51 years. Approximately one-third (33.2%, 240/722) underwent lymphadenectomy during their staging procedure. Most patients had stage I disease, (74.9%, 472/630). Adjuvant radiation therapy and chemotherapy were administered to 16.6% (113/679) and 21.4% (145/679) of women respectively. Adjuvant hormonal treatment was employed in the management of 20.9% (142/679) of patients. Table 1 summarizes the demographic and clinicopathological characteristics of the included studies.
Seventeen studies provided information on the tumor recurrence rate and included 190 women in the ovarian preservation group and 501 women in the bilateral salpingo-oophorectomy group. There was moderate heterogeneity among the included studies (p=0.01, I2 =49%). A total of 89/190 (46.8%) women in the ovarian preservation group experienced a tumor recurrence compared with 121/501 (24.2%) in the bilateral salpingo-oophorectomy group. Based on the random effects model, there was an increased risk of tumor recurrence in the ovarian preservation group (OR 2.70, 95% CI 1.39 to 5.28) (Figure 1). A sensitivity analysis was performed by excluding one study at each time with no change in the final results.
Three studies provided information on the death rate among women with low-grade endometrial stromal sarcoma, which included 34 women in the ovarian preservation group and 128 women in the bilateral salpingo-oophorectomy group. There was no heterogeneity among the included studies (p=0.69, I2 =0%). A total of two (5.9%) and nine (7.0%) deaths were observed in the ovarian preservation and bilateral salpingo-oophorectomy groups, respectively. Based on the random effects model, there was no statistically significant difference in the risk of death between women in the ovarian preservation and bilateral salpingo-oophorectomy groups (OR 0.80, 95% CI 0.18 to 3.47) (Figure 2).
Seven studies provided data on tumor recurrence exclusively among pre-menopausal women and included 91 in the ovarian preservation group and 87 women in the bilateral salpingo-oophorectomy group. There was low heterogeneity among the included studies (p=0.23, I 2 =28%). A total of 29/91 (31.9%) women in the ovarian preservation group experienced a tumor recurrence compared with 20/87 (23%) in the bilateral salpingo-oophorectomy group. Based on the random effects model, there was no statistically significant difference in recurrence risk between the the ovarian preservation group and the bilateral salpingo-oophorectomy group (OR 1.38, 95% CI 0.55 to 3.50) (Figure 3).
Eight studies provided data on tumor recurrence among patients with early stage disease (stage I–II) and included 45 women in the ovarian preservation group and 151 in the bilateral salpingo-oophorectomy group. There was moderate heterogeneity among the included studies (p=0.01, I 2 =62%). A total of 15/45 (33.3%) women in the ovarian preservation group experienced a tumor recurrence compared with 25/151 (16.6%) in the bilateral salpingo-oophorectomy group. Based on the random effects model, there was a no statistically significant difference in tumor relapse risk between the two groups (OR 2.69, 95% CI 0.55 to 13.50) (Figure 4).
Seven studies provided data on tumor recurrence stratified by the administration of adjuvant hormonal therapy (92 and 225 patients in the hormonal therapy and non-hormonal therapy groups, respectively). There was no heterogeneity among the included studies (p=0.76, I 2 =0%). A total of 23/92 (25%) women in the hormonal therapy group had a tumor relapse compared with 79/225 (35.1%) in the non-hormonal therapy group. Based on the random effects model, there was a decreased risk of tumor recurrence in the hormonal therapy group (OR 0.41, 95% CI 0.21 to 0.78) (Figure 5).
Discussion
The removal of ovaries achieves resection of possible subclinical ovarian metastases and induces surgical menopause in patients with a hormonally responsive tumor. Data supporting bilateral salpingo-oophorectomy for women with low-grade endometrial stromal sarcoma, however, are limited.21 Although several investigations suggest that the prevalence of occult ovarian metastases for low-grade endometrioid adenocarcinoma of the endometrium appears to be small,22 23 only limited data addressing this question for low-grade endometrial stromal sarcoma are available. Dos Santos et al reported no occult ovarian metastases among 20 patients with disease grossly confined to the uterus and grossly normal ovaries.27 Further studies are warranted since the majority of women present with apparent early stage disease.
Low-grade endometrial stromal sarcoma consistently expresses steroid hormone receptors and is considered a hormone sensitive tumor. Online supplementary table 1 summarizes published studies examining the expression of estrogen and progesterone receptor in low-grade endometrial stromal sarcoma; cumulative rates of estrogen receptor and progesterone receptor positivity were 76% and 81.2%, respectively.28–45 Removal of the ovaries results in loss of endogenous hormone production that could theoretically stimulate growth of residual tumor cells.
Supplemental material
Existing retrospective studies investigating the safety of ovarian preservation individually present conflicting results.3–21 The lack of conclusive evidence is reflected in the discrepancies among major cooperative group guidelines. The National Comprehensive Cancer Network clinical practice guidelines suggest that providers should consider re-resection of the ovaries for patients with low-grade endometrial stromal sarcoma.46 However, the Gynecologic Cancer InterGroup recommendations state that ovarian preservation could be offered to young women with small tumors. In addition, the recently updated European Society of Medical Oncology guidelines conclude that the value of bilateral salpingo-oophorectomy is not established, particularly for pre-menopausal women.47
In our meta-analysis, women with preserved ovaries had a significantly increased risk of tumor relapse. After excluding studies without menopausal status, a trend towards an increased risk of relapse was also noted when the analysis was limited to the 178 known pre-menopausal women (from an initial cohort of 673 patients). The lack of statistical significance in this difference in the subset analysis is likely due to the limited number of subjects.
Only three studies reported death rates for low-grade endometrial stromal sarcoma. While ovarian preservation was associated with an increased risk of recurrence in our meta-analysis, no difference in death rate was observed between women who did and did not undergo bilateral salpingo-oophorectomy. Median follow-up ranging from 40 to 115 months does not appear to be an issue.7 16 17 Our recent analysis of the Surveillance, Epidemiology, and End Results database found similar results, with no difference in overall survival between women aged <50 who did (n=369) and did not (n=151) undergo bilateral salpingo-oophorectomy.21 After a median follow-up of 116 months, 36 (6.7%) deaths were observed from which only 13 (36%; 2.4% of the total population) were attributed to the tumor.21 However, lack of central pathology review is a limitation of that study. Previous analyses of the Surveillance, Epidemiology, and End Results dataset also did not detect any survival difference between patients managed with bilateral salpingo-oophorectomy or with ovarian preservation.48 49 We hypothesize that a high salvage rate with hormonal therapy and/or bilateral salpingo-oophorectomy (secondary cytoreduction) in patients with recurrence may be responsible for the lack of survival difference.
Recurrence in patients with low-grade endometrial stromal sarcoma may be treated with secondary cytoreduction, oophorectomy, and administration of hormonal therapy. Complete secondary cytoreduction of recurrent disease correlates with survival. Bai et al reported 5-year survival rates following relapse for women with (n=15) and without (n=34) macroscopic residual tumor after salvage surgery of 65% and 100%, respectively (p<0.001).4 None of the patients with locally recurrent disease died of their disease.4 Similarly, Yoon et al reported that performance of secondary cytoreductive resection was independently associated with better survival in a group of 33 patients with recurrent low-grade endometrial stromal sarcoma.3 Hormonal management also appears to be successful in the management of recurrence. Cheng et al reported that, among 24 patients with recurrent low-grade endometrial stromal sarcoma who received hormonal therapy, 17% and 10% had complete and partial response, respectively, while 53% had stable disease.10 In 11 patients with recurrent or residual disease, an objective response to hormonal treatment was observed in nine (82%) patients.50 A response rate of 88% to progestin therapy was reported by Chu et al in eight women with recurrent low-grade endometrial stromal sarcoma.11 Aromatase inhibitors also represent an attractive option with a response rate of 77.4% in a review of data from 30 patients.51 Optimal agents and dosage are yet to be established. Despite the limitations, these findings seem to indicate that low-grade endometrial stromal sarcoma recurrences can potentially be salvaged, especially if localized. Online supplementary table 2 summarizes the location of tumor relapse in the included studies. Unfortunately only one study provided information on the pattern of recurrence based on the preservation or otherwise of ovaries. Li et al5 provided information on the location of tumor relapse among patients who underwent ovarian preservation or bilateral salpingo-oophorectomy. They did not indentify any differences in the pattern of recurrence between the two groups.5
Supplemental material
The increased risk of recurrence for patients with low-grade endometrial stromal sarcoma managed with ovarian preservation suggests a hormonal mechanism. Ovarian preservation could allow stimulation of otherwise undetectable hormonally responsive tumor cells. However, as studies included in this analysis were retrospective, non-hormonal confounding factors could be responsible for the findings. At best, the efficacy of adjuvant hormonal treatment is mixed. Three investigations with small patient populations suggest an advantage to adjuvant hormonal treatment. Malouf et al did not observe any recurrence in a group of 10 women with early stage low-grade endometrial stromal sarcoma who received adjuvant hormonal treatment.13 In addition, Beck et al observed a trend towards a decreased recurrence rate among women (n=42) with early stage low-grade endometrial stromal sarcoma who received adjuvant hormonal therapy (14.3% vs 38.5%, p=0.26).8 Similarly, Chu et al reported a lower recurrence rate among women (n=22) who received adjuvant progestin therapy (31% vs 67%).11 In contrast, larger cohort studies (n>100) have failed to confirm a benefit for adjuvant hormonal treatment.3 17 These discrepancies may be due to the heterogeneity of the regimens used; medroxyprogesterone acetate, megestrol acetate, aromatase inhibitors, mifepristone, and gonadotropin-releasing hormone analogs have all been employed in different dosages. Alternatively, these findings may suggest a lack of impact of estrogen and progesterone levels on recurrence in low-grade endometrial stromal sarcoma. On the other hand, based on limited evidence, hormonal replacement therapy with or withour progesterone may be associated with a higher relapse rate and is not recommended.52
A major strength of our study is the fact that we were able to pool data from a large number of patients with low-grade endometrial stromal sarcoma providing enough statistical power to draw conclusions. However, certain limitations should be noted. First, we could not fully control for clinicopathological characteristics such as stage and tumor size and for the administration of any adjuvant therapy. Also, due to the lack of central pathology review, we could not exclude the inadvertent inclusion of non-low-grade endometrial stromal sarcoma cases, especially given the recent reclassification of endometrial sarcomas. However, the majority of the alterations in the 2014 World Health Organization classification system are most relevant to high-grade endometrial stromal sarcoma and undifferentiated endometrial stromal sarcoma. Our investigation is limited to low-grade endometrial stromal sarcoma (<10 mitoses). It should be noted that mitotic figures are no longer required to discriminate low-grade from high-grade endometrial stromal sarcoma since the mutation (YWHAE-FAM22) is more useful. Moreover, while all included studies provided data on tumor relapse, subsequent subset analyses included a fraction of the total patient population. Lastly, due to the extremely low incidence of low-grade endometrial stromal sarcoma, the performance of prospective randomized trials is not feasible. All studies included in the present meta-analysis were retrospective so of low quality.
In conclusion, ovarian preservation for women with low-grade endometrial stromal sarcoma is associated with an increased risk of tumor relapse but, based on limited evidence, not with an increased risk of death. Based on the results of our study, ovarian preservation cannot be routinely recommended. However, it could potentially be considered in carefully selected young pre-menopausal women with disease confined to the uterus following extensive counseling. Unfortunately, there are no prospective randomized investigations to guide recommendations. Given the limited data, providers should review available studies, national and international guidelines with pre-menopausal patients with low-grade endometrial stromal sarcoma. After an informed discussion, an individualized plan should be made with each patient. Due to the rarity of this tumor, the creation of an international uterine sarcoma registry could aid in elucidating the optimal management of these patients.
References
Footnotes
Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.
Competing interests None declared.
Provenance and peer review Not commissioned, externally peer reviewed.