Introduction

Uterine sarcomas account for approximately 3-7% of all uterine cancers [1]. The most common histologic types of uterine sarcomas are leiomyosarcomas (LMS, 63%), endometrial stromal sarcomas (ESS, 21%), adenosarcomas (6%), undifferentiated sarcoma (5%), and smooth muscle tumors of uncertain malignant potential (STUMP) [2]. Because of the low incidence and the lack of prospective studies, it is challenging to reach conclusions as to the best disease management recommendations for uterine sarcoma [3]. It is difficult to discriminate between benign uterine fibroids and uterine sarcomas preoperatively by imaging examinations [4]. Most uterine sarcomas are often found incidentally after primary hysterectomy or myomectomy.

Myomectomy or hysterectomy can be performed by minimal invasive surgery (laparoscopic or transvaginal approach), and through abdominal or hysteroscopic approach [5,6,7,8]. To remove large specimens from the abdominal cavity through the laparoscope or minimally invasive incision, the morcellation approach was developed. Larger specimens are broken into smaller pieces during morcellation [9, 10]. However, the potential risk of malignant disease dissemination associated with power morcellation during surgery for presumed benign leiomyomas has been reported in many studies [5,6,7,8]. Remnant sarcomas and disseminated peritoneal sarcomatosis are often found in a significant number of uterine sarcoma patients after initial myomectomy or hysterectomy surgery, especially for those with uncontained power morcellation [11]. Re-exploration for completion surgery and staging is important when uterine malignancy is found incidentally [6]. For LMS patients who undergo morcellation, remnant sarcomas are found in approximately 15.0 to 64.3% of them and a significant proportion of patients are upstaged as a result of surgical re-exploration [6,7,8]. Surgical re-exploration may be useful for removing any remnant sarcoma and allow for accurate prognostication [12,13,14].

There is limited literature on the management of uterine sarcoma found incidentally after initial surgery for presumed uterine leiomyoma and the sample size of previous studies was small [12, 15, 16]. The exact clinicopathological characteristics and prognosis in these patients was not clear. To address this important knowledge gap, we retrospectively determined the clinicopathological characteristics and prognosis in patients with uterine sarcoma treated following surgery for presumed benign disease.

Patients and methods

Patients

This study was approved by the Institutional Review Board (IRB) (reference number: NCC2594) of the Cancer Hospital, Chinese Academy of Medical Sciences. All methods were performed in accordance with the relevant guidelines and regulations and this study was conducted in accordance with the Declaration of Helsinki. We performed a retrospective analysis of patients who diagnosed with uterine sarcomas incidentally. All patients received initial surgery for presumed benign disease and presented to our institution from January 1, 2004 to January 1, 2021. Patients received re-exploration and adjuvant treatment in the Department of Gynecological Oncology of Cancer Hospital, Chinese Academy of Medical Sciences, National Cancer Center. The inclusion criteria for patients were as follows: (1) presumed benign uterine fibroid disease and pre-operative imaging with computed tomography (CT) scan or ultrasound were performed before primary surgery; (2) diagnosed with uterine confined sarcoma incidentally after primary hysterectomy or subtotal hysterectomy or myomectomy, the primary FIGO stage was stage I; (3) presented to our hospital within three months of the primary surgery date; (4) confirmed by two experienced gynecologic pathologists in our hospital; (5) MRI or CT were performed with contrast before the re-exploration surgery in our hospital; (6) received re-exploration in our center; (7) written informed consent for the staging operation was obtained from all individuals. Criteria for exclusion encompassed patients who: (1) were demonstrated radiological evidence of extrauterine tumor spreading; (2) declined to complete the staging operation; (3) received primary hysterectomy in our center. Myomectomy and subtotal hysterectomy were rarely performed in our hospital. All patients with incidentally uterine confined sarcoma had their primary surgery performed at outside institutions by different surgeons in the present study.

The cohort was divided into two subgroups according to the type of initial surgery: the hysterectomy subgroup and myomectomy subgroup. The patients’ full medical records were included in this study. Clinical and pathologic variables, treatment modalities, and outcomes were assessed. Stage was retrospectively assigned using the International Federation of Gynecology and Obstetrics (FIGO) 2009 staging system for uterine sarcomas [1, 17]. For uterine leiomyosarcomas and endometrial stromal sarcomas, stage I means tumor limited to uterus, stage II means tumor extends beyond the uterus, within the pelvis, stage III means tumor invades abdominal tissues (not just protruding into the abdomen), stage IV means tumor invades bladder and/or rectum or distant metastasis. FIGO staging for uterine adenosarcomas can refer to the stage of leiomyosarcoma, except for stage I [17].

Statistical analyses

The associations between the type of initial surgery and clinicopathologic characteristics were performed using the Pearson χ2 test or the Fisher exact test. For the survival analyses, progression-free survival (PFS) was defined as the time from the date of original surgery (hysterectomy or myomectomy) at the outside center in which they were originally treated to first recurrence (local or distant), or death for any cause [12]. Overall survival (OS) was defined as the time from the date of original surgery (hysterectomy or myomectomy) at the outside center in which they were originally treated to death for any cause. Survival was estimated using the Kaplan–Meier method, and differences were tested for statistical significance using the log-rank test. Two-sided P values less than 0.05 were considered to be statistically significant. Hazard ratios (HRs) and 95% confidence intervals (CIs) for PFS and OS were estimated from multivariable Cox proportional hazards models to find factors associated to higher hazards for recurrence or death. The regressions were adjusted for patient age (<40 year vs. ≥ 40 years), primary surgery method (myomectomy versus hysterectomy), tumor size (≥ 5 cm vs. < 5 cm), remnant sarcomas (Yes. vs. No), disseminated diseases (Yes. vs. No) [5]. All analyses were performed using the SPSS Statistics20.0 software.

Results

Patient characteristics

Overall, we included 95 patients in our study. For the initial surgery, myomectomy was performed in 50 (52.6%, 50/95) patients, hysterectomy was performed in 45 (47.4%, 45/95) patients with 33 (34.7%, 33/95) and 12 (12.6%, 12/95) patients underwent total or subtotal hysterectomy, respectively. Patients median age at diagnosis was 43 years [interquartile range (IQR): 35–47 years]. For patients in myomectomy subgroup, median age at diagnosis was younger than patients in hysterectomy subgroup which was 38 years (IQR: 33–43 years) and 45 years (IQR: 42–50 years), respectively. For the primary treatment in the original center of diagnosis, about 45.3% (43/95) patients presented with no obvious symptoms and were diagnosed because of physical examination, 41.1% (39/95) patients presented with vaginal bleeding. (Table 1)

Table 1 Association of patient/tumor characteristics with the type of primary surgery
Full size table

In the entire cohort, 35 (36.8%) patients had leiomyosarcomas (LMS), 54 (56.8%) patients had low-grade endometrial stromal sarcomas (LG-ESS), 4 (4.2%) patients had adenosarcomas, 1 (1.1%) patient had UUS, and 1 (1.1%) patient had STUMP. (Table 1) All these patients had their primary myomectomy or hysterectomy performed at an outside institution and then referred to our center. The percentage of patients who received laparoscopic or laparotomic surgery was 37.9% and 51.6%, respectively. None of the 95 patients had documented evidence of extra-uterine disease at the time of primary surgery.

Surgical re-exploration treatment

In the entire cohort, all patients were re-explored to complete the staging operation. Patients who diagnosed after myomectomy received total hysterectomy and patients who diagnosed after subtotal hysterectomy received secondary resection of the cervical stump. After the primary and staging surgeries, 83 (87.4%) underwent bilateral salpingo-oophorectomy, 57 (60.0%) pelvic or pre-aortic lymphadenectomy and 74 (77.9%) patients underwent omentectomy. The median time to the staging surgery was 40 days (range 25–70 days). Patients who received hysterectomy were more likely to receive lymphadenectomy (75.6% vs. 46.0%, P = 0.003).

Remnant sarcoma

After re-exploration, there were 29 patients (30.5%, 29/95) had remnant sarcomas, with 17 patients (17/95, 17.9%) on the remaining uterus, 9 patients (9/95, 9.5%) had disseminated diseases, and 4 patients (4/95, 4.2%) had positive lymph nodes. For patients with disseminated diseases (n = 9), 3 patients (33.3%) disseminated to pelvic peritoneum, 4 patients (44.4%) disseminated to ovarian, 1 patient (11.1%) disseminated to abdominal peritoneum, 1 patient (11.1%) disseminated to pelvic peritoneum, residual cervical, and had positive pelvic lymph nodes. (Table 2)

Table 2 Clinical characteristics and survival of patients with disseminated diseases (n = 9)
Full size table

For patients with disseminated diseases (n = 9), myomectomy was performed in 5 patients (10.0%, 5/50), subtotal hysterectomy was performed in 1 patient (8.3%, 1/12), and 3 patients (9.1%, 3/33) underwent total hysterectomy as the initial surgery. (Table 1) Patients with ESS were more likely to have remnant sarcomas than those with LMS, but the difference did not reach a statistical significance (35.2% vs. 25.7%, P = 0.347).

The FIGO 2009 distribution after re-exploration by stage was: stage I in 83 patients (87.4%), stage II in 7 patients (7.4%), stage III in 5 patients (5.3%). (Table 1)

Adjuvant therapy

After re-exploration, about 40 patients (42.1%, 40/95) received adjuvant chemotherapy. The chemotherapy regimens included: doxorubicin/ifosfamide/cisplatin (13), doxorubicin/ifosfamide (9), gemcitabine/docetaxel (7), paclitaxel/platinum (5), paclitaxel/ifosfamide (4), paclitaxel/doxorubicin (1), DTIC/doxorubicin (1). There were 17 (17.9%) patients and 23 (24.2%) patients received adjuvant radiotherapy and adjuvant endocrine therapy after surgery, respectively. More patients in the hysterectomy subgroup received adjuvant radiotherapy than in the myomectomy subgroup (26.7% vs. 10.0%, P = 0.034).

There were 55.2% (16/29) and 36.4% (24/66) patients with/without remnant disease received adjuvant chemotherapy, respectively (P = 0.087). About 41.2% (7/17) patients with remnant disease on the remaining uterus received adjuvant chemotherapy. Patients with disseminated diseases (n = 9) were more likely to receive adjuvant chemotherapy than those without (n = 86) (88.9% vs. 37.2%, P = 0.003).

Survival analysis

The median follow-up duration was 76.7 months (range: 34.8-118.1 months). There were 17 (17.9%) patients experienced recurrence and 8 (8.4%) patients died during follow-up. 5-year PFS and 5-year OS for the entire cohort was 81.7% and 92.1%, respectively. The median PFS and OS were not reached for the entire cohort. Stage-specific 5-year PFS and 5-year OS were as follows: stage I—84.3% and 94.0%, stage II-III—66.7% and 75.0%.

Patients with remnant sarcomas had a tendency towards a worse 5-year PFS and 5-year OS, compared with those without remnant sarcomas (5-year PFS: 75.6% vs. 84.5%, P = 0.224; 5-year OS: 85.5% vs. 95.1%, P = 0.217). (Fig. 1) Patients with disseminated diseases had a worse 5-year OS (62.5% vs. 95.1%, P = 0.007) and non-significantly worse 5-year PFS (64.8% vs. 83.4%, P = 0.153) compared with those without disseminated diseases (Fig. 2). Multivariate Cox regression analysis demonstrated that disseminated diseases was not an independent prognostic marker when adjusted for several factors (Supplementary Table 1).

Fig. 1
figure 1

Kaplan–Meier estimates of PFS and OS according in patients with uterine sarcoma treated following surgery for presumed benign disease with/without remnant sarcomas. A, PFS; B, OS

Full size image
Fig. 2
figure 2

Kaplan–Meier estimates of PFS and OS according in patients with uterine sarcoma treated following surgery for presumed benign disease with/without disseminated diseases. A, PFS; B, OS

Full size image

Patients who received lymphadenectomy had a tendency towards a better 5-year PFS compared with those not received (85.5% vs. 67.1%, P = 0.127), no difference in 5-year OS was found between the two groups (91.8% vs. 88.1%, P = 0.844). In the entire cohort, no difference in 5-year PFS or 5-year OS was found between patients who had received different primary surgery (hysterectomy vs. myomectomy: 5-year PFS, 85.9% vs. 77.8%, P = 0.551; 5-year OS, 93.1% vs. 91.1%, P = 0.974).

Recurrent disease

During follow-up, 17 patients (17.9%) had recurrence following re-exploration surgery. Of these patients with recurrent disease (n = 17), 11 (64.7%) patients had leiomyosarcomas, 6 (35.3%) patients had low-grade endometrial stromal sarcomas.

Median time from diagnosis to first recurrence was 13 months (range 1.9–85.9 months). Of these patients with recurrent sarcoma (n = 17), 3 patients had disseminated diseases and 4 patients had remnant sarcomas on the remaining uterus after re-exploration surgery. The most common site of first recurrence was the abdominal/pelvic peritoneum, diagnosed in 11 (64.7%) patients, followed by lung metastases in 3 (17.6%) patients, abdominal wall metastases in 1 (5.9%) patient, vaginal cuff metastases in 1 (5.9%) patient, and multiple metastases of different organs in 1 (5.9%) patient. (Table 3)

Table 3 Sites of first relapse disease in the entire population
Full size table

There were 8 patients (8/17, 47.1%) with recurrent disease underwent cytoreduction surgery, 7 patients (7/8, 87.5%) completed resection of the tumor and 1 patient (1/8, 12.5%) failed to complete cytoreduction surgery because of the advanced disease. Moreover, 2 patients (2/8, 25.0%) underwent a second cytoreductive surgery for the abdominal wall and vulva metastasis, respectively.

Discussion

We performed a retrospective study to explore the clinicopathological characteristics and prognosis in patients with uterine sarcoma treated following surgery for presumed benign disease. All patients had previously undergone surgery for presumed benign uterine fibroid disease and were diagnosed with uterine sarcoma after surgery. The initial surgery included myomectomy/ subtotal hysterectomy/ total hysterectomy. We found patients with uterine sarcoma treated following surgery for presumed benign disease had a favorable survival. Patients with disseminated diseases had a worse 5-year OS compared with those without. Surgical re-exploration may be valuable for removing remnant sarcomas and disseminated diseases. Our study is one of the largest studies to investigate the clinicopathological characteristics and prognosis in patients with incidental uterine sarcomas.

Uterine sarcomas are rare uterine malignancies that are difficult to diagnose preoperatively. Most uterine sarcomas are often found incidentally after primary hysterectomy or myomectomy. Morcellation is widely used during these surgeries, especially for minimally invasive surgery [9]. But the potential risk of malignant disease dissemination associated with morcellation during surgery have been reported in many studies and may change the recurrence pattern [5, 10]. Morcellation may be associated with a worse survival. However, we were unable to determine whether the morcellation was used during primary surgery, since all patients received the primary surgery at outside institutions. We could not identify the exactly prognosis role of morcellation during myomectomy or hysterectomy surgery in the present study.

Remnant sarcomas and disseminated peritoneal sarcomatosis are often found in uterine sarcoma patients after initial surgery with uncontained power morcellation. Surgical re-exploration may be useful for removing remnant sarcomas or disseminated diseases [12]. Einstein et al. found approximately 15.4% of patients would be found had remnant sarcomas particularly those with LMS who underwent morcellation [6]. In the present study, about 1/3 patients who received myomectomy had remnant sarcomas on the remaining uterus after re-exploration and 9.5% patients had disseminated diseases after re-exploration. We observed no difference in the occurrence rate of disseminated diseases between patients who underwent myomectomy or hysterectomy at the primary surgery. The potential reason might be the sample size of patients with disseminated diseases was too small for firm conclusions to be made.

Surgical re-exploration for completion surgery and staging is important when uterine malignancy is found incidentally. A significant proportion of patients are upstaged as a result of surgical re-exploration, specifically for patients with incidentally identified uLMS [6]. A recent study showed a quarter (10/38, 26.3%) of patients were upstaged during re-exploration and almost all (90.0%) were upstaged to stage III [8]. Consistent with the results of other studies, our study indicated that 12.6% (12/95) patients upstaged. Re-exploration may be a major method to identify patients with remnant sarcomas. In particular, STUMPs are a rare tumor that represent a diagnostic challenge to both clinicians and pathologists. Martina et al. showed that pregnacy may be possible and safe after the diagnosis of STUMP. These patients should undergo rigorous follow-up and consider hysterectomy after completion of their reproductive plan [18].

Findings from re-exploration can contribute to the knowledge of natural history of morcellated sarcomas and allow for accurate prognostication [14]. Most remnant sarcomas may not be identified by ultrasound, computed tomography (CT), and magnetic resonance imaging (MRI) [19]. Re-exploration to surgical staging is valuable to find and remove the remnant sarcomas [6, 13, 20]. It is useful to identify the high-risk patients and improve the survival of these patients. In the present study, patients with remnant sarcomas had a tendency towards worse survival, especially for patients with disseminated disease. In the other hand, patients who are not upstaged appear to have a good prognosis [6]. We highly recommend surgical re-exploration, even when there is no evidence of a metastatic lesion in imaging studies.

Re-exploration may be difficult because of the complex adhesion after primary surgery and disseminated disease. These adhesions can cause a difficulty entering the abdominal cavity at secondary surgery and increase risk of bowel and bladder injury. The difficulty to surgically remove the remnant disease may increase. The time to re-exploration is valuable for prognosis and may alter postoperative treatment [8]. Other studies recommend immediate re-exploration in patients with inadvertently morcellated uterine sarcomas [6, 14]. In our study, the median time to receive re-exploration was 40 days. No significance of 5-year PFS or OS were observed between patients who received re-exploration within 30 days or more than 30 days.

Given the low incidence of lymph node metastases for uterine sarcomas, the value of lymphadenectomy for early-stage uterine sarcomas has been questioned. Emerging evidence has showed lymphadenectomy may be safely omitted for clinically uterus-confined leiomyosarcoma [21]. In the present study, we found 7.0% (4/57) patients had pelvic or para-aortic lymph node metastases and lymphadenectomy associated with a non-significantly better 5-year PFS. In addition, we found no significant association between omentectomy and a better survival. However, omentectomy should be part of the re-exploration surgery, especially for patients with uterine sarcoma found incidentally who underwent morcellation during primary surgery. Surgical re-exploration is likely to alter postoperative treatment because of upstaging. All patients with remnant sarcomas received adjuvant chemotherapy or radiotherapy after re-exploration in our study.

There are two limitations to our study. Because of the rarity of uterine sarcoma found incidentally after primary surgery, the sample size of our study was small. Because the primary surgery was performed outside of our hospital, there were differences in surgical techniques. Therefore, caution is required when interpreting our results. Further prospective studies are warranted to validate the exact role of re-exploration.

Conclusion

In summary, patients with uterine sarcoma treated following surgery for presumed benign disease have a favorable survival. Re-exploration to complete the staging operation have a high likelihood of detecting and removing remnant or disseminated sarcoma. Patients with disseminated diseases had a worse 5-year OS compared with those without.