Comparative analysis of hormonal therapy in low-grade endometrial stromal sarcoma: A retrospective study.

This is a retrospective cohort of patients diagnosed with LG-ESS who are enrolled in the Rare Gynecologic Malignancy Registry at the University of Texas MD Anderson Cancer Center and Lyndon B. Johnson Hospital, an affiliated county hospital. Patients with uterine or extrauterine LG- ESS, undergoing active treatment between January 2000 and July 2023, of all disease stages and at any phase of treatment were included. Patients with unavailable pathology reports, disease reported as ESS features or differentiation, or empty medical charts were excluded. Diagnoses were reclassified using the WHO terminology and classification system [ 2 ] and staged using the International Federation of Gynecology and Obstetrics (FIGO) for leiomyosarcoma and ESS. Study data were collected and managed using REDCap electronic data capture tools hosted at MD Anderson [ 8 ]. This analysis was approved by the institutional review board of MD Anderson Cancer Center (PA18-0064). Summary statistics were used to summarize demographic and clinical characteristics. Overall survival (OS) and recurrence-free survival (RFS) were estimated using the methods of Kaplan and Meier and modeled via Cox proportional hazards regression. OS was calculated from date of diagnosis to earliest date of death (due to any cause) or last follow up. RFS was calculated from date of diagnosis to earliest date of recurrence or death (due to any cause). Patients alive and without evidence of disease were censored at last gynecological visit. RFS2 was calculated from date of diagnosis to earliest date of second recurrence or death (due to any cause). Statistical analysis was performed using Stata/MP v17.0 (College Station, TX).

A total of 221 patients with confirmed diagnosis of LG-ESS were included in the analysis. Most patients were white (82%, 167/203) and non-Hispanic or Latinx (86%, 179/208). The mean age at diagnosis was 46 years (range, 16 to 96). Common routes to diagnosis were incidental finding at the time of hysterectomy (37%, 81/221) and symptoms leading to exam (81%, 180/221), most common of which were abnormal uterine bleeding (71%, 128/180) and abdominal/ pelvic pain (41%, 74/180). In terms of medical history and risk factors, among the 221 patients, 51% (112/221) had a prior history of uterine fibroids, and 31% (64/207) had a history of obesity before their diagnosis. The mean body mass index (BMI) at the time of diagnosis was 28 kg/m 2 (range, 17 to 49). 83% (122/147) have used hormonal contraceptives and 21% (37/ 173) have used hormonal replacement therapy (HRT) for a mean duration of 11.06 (SD= 29.18) and 11.89 (SD= 30.75) years, respectively. 185 patients (84%) had LG-ESS of uterine origin and 36 (16%) extrauterine origin. Stage distribution at diagnosis was as follows: 58% (91/157) stage I, 12% (19/157) stage II, 13% (20/157) stage III and 17% (27/157) stage IV. The demographic and clinical characteristics of patients are summarized in Table 1 . The median follow-up time for patients was 6.70 years (range: 0.05-41.85). Median RFS was 6.67 years (95% CI: 5.17 – 8.32) and median OS was 18.16 years (95% CI: 16.02 – 26.32). 63% of patients (69/109) had lympho-vascular invasion on pathology. Of patients who underwent immunohistochemistry (IHC) staining, estrogen receptor (ER) and progesterone receptor (PR) were positive among 97% of patients (89/92 and 88/91, respectively). The rest of IHC staining is summarized in Table 2 . Only 17% of patients (38/221) underwent some form of special testing (tumor molecular testing, FISH, microsatellite instability, tumor mutation burden, PD-L1 receptor testing…), and among those, JAZF1 gene rearrangement was the most identified mutation (7/38, 18%). Lympho-vascular invasion was associated with higher recurrence rate (p= 0.033) but did not affect OS (p= 0.227) ( Table 3 ). Negative status of ER, SMA, and desmin were associated with both higher recurrence rate (p= 0.039, p= 0.002, and p= 0.015, respectively) and poorer OS (p= 0.008, p= 0.012, and p= 0.013, respectively) ( Table 3 ). Negative status of PR on pathology did not affect RFS (p=0.197) but was associated with poorer OS (p= 0.003) ( Table 3 ). Negative status of CD10 and cytokeratin did not impact RFS (p= 0.399 and 0.092, respectively) or OS (p=0.368 and 0.447, respectively) ( Table 3 ). Surgery was the primary treatment for most patients (98%, 213/218). Surgery as primary treatment was associated with better RFS (HR= 0.22, 95% CI 0.08 to 0.61, p=0.004) and better OS (HR= 0.17, 95% 0.04 to 0.72, p=0.016). Among the patients who received surgery alone (n=113), 79% (68/86) were classified as stage I. These patients demonstrated an RFS of 6.25 years (95% CI: 4.43-8.32) and an OS of 20.29 years (95% CI: 15.59-28.72). 59 patients of the surgery alone group experienced recurrence; 34 underwent second surgery, 38 received hormonal therapy, 9 chemotherapy, and 5 radiotherapy as part of their first recurrence treatment (not mutually exclusive). Out of 100 patients who received adjuvant therapy, 79% (79/100) were on hormonal treatment for a mean duration of 34.04 (SD= 43.03) months ( Figure 1 ). Within this group, 58% (46/79) received Megace, 24% (19/79) received Letrozole, and 18 % (14/79) received other hormonal therapy (Leuprolide Acetate, Anastrozole, Medroxyprogesterone Acetate, Tamoxifen) ( Figure 1 ). Mean duration of Megace therapy was 37.75 months (95% CI: 20.95 – 54.55) versus Letrozole, 30.26 months (95% CI: 14.35 – 46.16, p= 0.883). For adjuvant therapy, receiving selective estrogen receptor modulators (SERM) such as Tamoxifen was associated with lower RFS (HR 5.03, 95% CI 1.16 – 21.86, p=0.031). Patients who received Megace as adjuvant therapy had a median RFS of 13.48 years and OS of 15.95 years, while patients who received Letrozole had a RFS of 6.06 years and OS was not evaluable (NE). There was no significant difference in RFS (HR=2.14, 95% CI 0.74-6.18, p=0.150) and OS (HR=4.16, 95% CI 0.55-31.41, p=0.134) between patients receiving Megace versus Letrozole as first line adjuvant therapy ( Figure 2 ). Out of 221 patients, 114 (52%) had at least one recurrence. Treatment modalities administered at first recurrence varied: surgery was performed in 53% (60/114), hormonal therapy in 67% (76/114), chemotherapy in 18% (21/114), and radiotherapy in 11% (12/114). It is important to note some patients received multiple modalities of treatment and these options were not mutually exclusive. Of those who received hormonal therapy, 37 patients received Megace, 15 received Letrozole, and 24 received other types of hormonal therapy. Chemotherapy at first recurrence was associated with shorter RFS (HR=2.76, 95% CI 1.58 - 4.83, p<0.001) and shorter OS (HR=2.78, 95% CI 1.51-5.10, p= 0.001). In contrast, hormonal therapy was associated with better RFS (HR=0.55, 95% CI 0.35-0.89, p=0.014) and OS at first recurrence (HR=0.51, 95% CI 0.3-0.87, p=0.014). Patients who received Megace at first recurrence had a median RFS of 16.28 years and OS of 27.46 years, while patients who received Letrozole had a RFS of 4.18 years and OS of 7 years. Patients given Megace at first recurrence had a similar RFS to those who received Letrozole (HR=0.64, 95% CI 0.27-1.49, p= 0.298), but a better OS (HR= 0.3, 95% CI 0.11-0.81, p=0.012) ( Figure 3 ).

Our study reflects the indolent disease course of LG-ESS, its favorable outcomes, and the current landscape of LG-ESS treatment. There were no significant differences in oncologic outcomes between Megace and Letrozole as adjuvant therapies for LG-ESS. While both treatments demonstrated comparable efficacy in the primary setting, Megace exhibited a potential survival advantage in recurrent disease. These findings advocate for further investigations to determine the most effective agents and their optimal sequence in the treatment of this rare gynecologic malignancy.

This retrospective study investigated the disease course and treatment outcomes in LG-ESS, with a particular focus on hormonal therapies. There were no significant differences in oncologic outcomes between Megace and Letrozole as adjuvant therapies in primary disease, but Megace exhibited a potential survival advantage in recurrent disease. The median OS in our study was 18.16 years (95% CI: 16.02 – 26.32) aligning with the favorable prognosis previously reported in LG-ESS, where the 5-year survival rate exceeds 90% [ 4 ]. Median RFS was 6.67 years (95% CI: 5.17 – 8.32) indicating a tendency for late recurrences in this disease [ 5 ]. These results highlight the positive outcomes and indolent disease course associated with LG-ESS. Among 38 patients who underwent tumor genetic testing, 7 individuals (18%) had a JAZF1 gene rearrangement mutation. The chromosomal rearrangement T(7;17) (p15;q21) which results in the JAZF1/SUZ12 gene fusion protein is commonly observed in LG-ESS [ 9 - 11 ]. While the pathogenesis of stromal tumors remains poorly understood, this chromosomal rearrangement may play a significant role [ 12 ]. We identified an association between LVSI and poorer RFS, although it did not impact OS. The poor prognostic role of LVSI is well-established in endometrial cancer [ 13 , 14 ] and uterine sarcomas [ 15 , 16 ]. In LG-ESS, LVSI was also associated with higher recurrence rate and lower disease-free survival (DFS) [ 17 ]. In our study, 97% of patients who underwent IHC staining were ER and PR positive. A study on uterine sarcomas revealed 63% ER positivity among 54 patients [ 18 ]. Additionally, in LG-ESS, positive IHC staining for ER and PR were found in 79-94% and 93-94% of patients, respectively [ 9 , 19 ]. Survival analysis indicated that ER-positive sarcomas were associated with improved OS compared to ER-negative cases (median OS 36 vs. 16 months, p = 0.004) and ER positivity was an independent predictor of survival (HR = 0.32, CI 0.12–0.89, p = 0.03) [ 18 ]. Notably, our findings align with these results, as the negative status of ER on pathology in our study was associated with poorer RFS and OS. CD10, WT-1, vimentin, and actins are typically positive on IHC in LG-ESS [ 20 ], consistent with our own findings as demonstrated in Table 2 . The majority of patients in our study underwent surgery as the primary treatment, and this was correlated with improved RFS and OS. These findings support the formal recommendation of THBSO as the primary treatment for resectable LG-ESS [ 20 ]. Despite the lack of consensus on adjuvant treatment, the expression of ER and PR in LG-ESS has prompted the use of hormonal therapies, particularly progestins like Megace, and AIs such as Letrozole for recurrence prevention [ 20 ]. Our study revealed that 58% received Megace, 24% received Letrozole, and 18% received other hormonal therapies as first-line adjuvant therapy, reflecting the current variability in treatment choices ( Figure 1 ). Notably, a study by Reich et al. [ 21 ] suggested a preference for AIs due to their better tolerability compared to progestins. Although no comparative studies exist on the outcomes of different hormonal adjuvant therapy options, we found no significant difference in outcomes between patients receiving Megace versus Letrozole, indicating comparable efficacy. Additionally, Tamoxifen was administered as adjuvant therapy for two patients in our study, and its use was associated with lower RFS. This observation aligns with the contraindication of Tamoxifen in patients post-ESS treatment [ 12 ], emphasizing the need for caution in employing such treatments as adjuvant therapy for LG-ESS. Out of 221 patients, 114 (52%) experienced at least one recurrence, aligning with the reported 30-50% recurrence rate documented in the literature [ 12 ]. There is currently no standard therapy for patients with recurrent LG-ESS, and various modalities, including hormone therapy, radiation, surgical re-excision, or a combination of these approaches, have been utilized [ 12 , 22 ]. The use of chemotherapy has also been reported in some cases of recurrent disease [ 23 ]. This highlights the variability in treatment modalities of recurrent LG-ESS, mirroring the variability observed in our patient population. While chemotherapy is a recognized treatment for HG-ESS and UUS, its efficacy in recurrent LG-ESS remains uncertain [ 12 ]. In our study, chemotherapy was received by 10% of patients in recurrent disease and correlated with both lower RFS and OS, suggesting that it may not be an optimal choice for recurrent disease management. Hormonal therapy, on the other hand, showed a favorable association with RFS and OS at first recurrence, suggesting that it may be a preferred approach in recurrent disease. No comparative studies currently exist on the outcomes of different hormonal therapy options in recurrent disease for LG-ESS. In our study, we observed that Megace has comparable RFS to Letrozole, but a superior OS. This survival advantage can offer valuable clinical insights and contribute to patient counseling regarding treatment options in recurrent disease, particularly considering the lack of clinical trials for rare tumors like LG-ESS. The median RFS for Megace was 16.3 years, whereas Letrozole exhibited an RFS of 4.2 years. Although the difference in RFS was not statistically significant (HR=0.64, 95% CI 0.27-1.49, p= 0.298), it is crucial to recognize the potential clinical significance of the observed 12-year difference. Our study has several strengths. Our utilization of the Rare Gynecologic Tumor Registry allowed for the aggregation of well-curated data from a large sample size of 221 patients with LG-ESS which made it feasible to conduct an adequately powered study evaluating and comparing treatment strategies and oncologic outcomes in LG-ESS. Our study also stands out as the first that we know of to undertake a comparative analysis of oncologic outcomes across different hormonal treatment options, providing valuable insights into clinical decision-making. In addition to the inherent limitations of retrospective studies, our comparison groups and our ability to comprehensively explore treatment sequencing were limited by the wide variation in treatment data which stems from the ambiguity in treatment recommendations for LG-ESS. Additionally, with 82% of our respondents being white, underrepresentation of diversity introduces another limitation to our study limiting its generalizability and perpetuating disparities in healthcare outcomes.

Endometrial stromal sarcoma (ESS) is a rare gynecologic tumor accounting for 6-20% of uterine sarcomas and less than 1% of all uterine cancers [ 1 ]. In 2014, the World Health Organization (WHO) classified “endometrial stromal and related tumors” into four tumor types: endometrial stromal nodule (ESN), low-grade ESS (LG-ESS), high-grade ESS (HG-ESS) and undifferentiated uterine sarcoma (UUS) [ 2 ]. In contrast with the latter types, ESN is a benign uterine tumor without metastatic potential [ 3 ]. LG-ESS tends to have a more indolent course and better prognosis compared to HG-ESS or UUS [ 4 ]. The 5-year survival of LG-ESS exceeds 90% [ 1 , 4 ], but late recurrences have been identified in up to 60% of patients [ 5 ]. Extrauterine ESS (EESS) refers to ESS arising at extrauterine sites and presumably from preexisting endometriosis [ 6 ]. Total hysterectomy and bilateral salpingooophorectomy (THBSO) is considered the mainstay of treatment of resectable LG-ESS [ 7 ]. Depending on the stage at diagnosis, patients will be dispositioned to observation or adjuvant endocrine therapy [ 7 ]. Adjuvant therapy involves hormonal treatment with aromatase inhibitors (AIs) (Letrozole) or Megestrol Acetate (Megace). Non-resectable LG-ESS benefit from primary hormonal therapy while recurrent LG-ESS cases may explore second-line hormonal therapy [ 7 ]. Due to the lack of large prospective studies or clinical trials on this topic, there are no formal recommendations regarding the first line agent, dosage, and duration of adjuvant endocrine therapy for primary or recurrent LG-ESS [ 7 ]. The objectives of this study were to (1) evaluate the clinicopathologic features and oncologic outcomes of patients with LG-ESS and (2) compare oncologic outcomes associated with different types of hormonal therapy in primary and recurrent disease.