Uterine sarcoma is a rare mesenchymal-origin gynecological tumor with a high recurrence rate and poor prognosis. Preoperative diagnosis is usually difficult due to the lack of specific symptoms and diagnostic techniques [1]. So far, most researches on the relationship between risk factors, adjuvant treatment and survival were conducted in western countries [12-16], but much less is known about such association in China [17]. Our retrospective study provided insights into the clinicopathological features and prognosis factors of 75 Chinese uterine sarcoma patients over 10 years.
Our study confirmed three major histopathological subtypes of uterine sarcoma: ESS, LMS and adenosarcoma, which was in accordance with previous study [18]. However, our study showed that the most common type was ESS (48%), which was contrary to previous studies showing LMS as the most common type [4]. One explanation may be the different samples from various studies, which need to be tested by further multicenter studies.
The univariate analysis showed that over 50 years, post-menopause, advanced stage, and ≥1/2 myometrial invasion were significantly associated with poorer survival; while multivariable analysis demonstrated that post-menopause and advanced stage were independent prognostic factors for survival of the total cohort and the LMS group. These findings were consistent with previous studies showing higher incidence and poorer prognosis of uterine sarcoma in postmenopausal women, women aged >50 [19, 20] and with advanced tumor stages [14, 21]. Our study showed that the 5-year survival rate of ESS was as high as 84.9%, which was similar to the reported survival rate of 75% in an Iran study with 42 patients [21]. These high survival rates might be related to an early stage when diagnosed. No statistically significant differences were observed in the PFS and OS by various histological types. However, the LMS group had a significantly higher recurrence and a shorter 5-year survival rate than ESS, which was in agreement with other studies [22, 23]. In our study, 26 out of 75 patients (34.67%) had recurrence, which was slightly lower than the reported recurrence rate range of 50-70% by previous studies [24-26]. In addition, our study showed that all relapses occurred in the pelvis, although Maria Ruiz-Minaya [27] reported multiple metastases were common.
Diagnosing uterine sarcoma is generally difficult due to unspecific early symptoms, such as irregular vaginal bleeding, abdominal pain, and pelvic mass which were also common in other uterine lesions [4]. In our study, 56% of cases exhibited irregular vaginal bleeding as an early symptom, which was observed for all histological types. This supports the finding of Nagai [28] who reported that abnormal vaginal bleeding was the most common clinical symptom in uterine sarcoma patients. Our study also showed that the early symptom of abnormal vaginal bleeding was associated with longer survival including PFS and OS in LMS patients, but the associations were only marginally significant.
Ultrasound is a useful diagnostic tool for uterine disease among women with uterine bleeding. In our research, 62.9% of patients had UBFS during the examination. In the LMS group, patients with UBFS had significantly shorter PFS and marginally significantly shorter OS. This suggests the potentially important role of UBFS in the evaluation of the malignant transformation of uterus myoma, a finding similar to those reported by Yang Hua [29]. Asim Kurjak reported that transvaginal color doppler ultrasound could distinguish uterine sarcoma from uterus myoma by using a cutoff of 0.4 of the resistance index of tumoral blood vessels [30]. This cutoff showed a diagnostic value of 90.91% for sensitivity, 99.82% for specificity, 71.43% for positive predictive value and 99.96% for negative predictive value, respectively [30]. However, a major limitation of the study lies in its small sample size (n=10). Further studies are needed to figure out the clinical significance of the resistance index of tumor blood flow for uterine sarcoma.
Surgery is another important independent prognostic factor for uterine sarcoma [14, 31]. NCCN guideline (https://www.nccn.org/) recommends that TH-BSO is the most effective method to treat uterine sarcoma, even in premenopausal women with stage I of ESS. As described in our study, 97.33% of patients received primary surgery, among who 93.15% had TH-BSO and 6.85% had a hysterectomy as their first-line treatment within 1 month after diagnosis. This finding was in line with other studies reporting surgical treatment rates ranging from 87.6% to 98% [27]. In our research, the effect of surgery with ovarian function preservation on survival was not analyzed due to the small sample size (n=2). However, previous studies confirmed that surgery without preserving ovarian function was more effective in decreasing the risk of recurrence than surgery with ovarian function preservation [11, 32]. Lymph node dissection remains controversial because of infrequent metastasis [16, 26], noted in less than 10% [33]. Researchers suggested that lymphadenectomy was only performed when the lymph nodes were suspected of metastasis by imaging or during surgery. In our study, 32.88% of patients received lymphadenectomy. Only one LMS patient with stage Ⅳ had positive lymph node metastasis, with an OS of 9 months. A study showed that 70% of lymph node metastasis patients were at the FIGO stage Ⅳ [34].
The role of adjuvant chemotherapy is also controversial in the management of uterine sarcoma. Some studies reported that regimens of ifosfamide [11, 35, 36] or doxorubicin [11, 32, 36] were effective for uterine sarcoma and decreased the risk of metastasis. Some researchers found that adjuvant chemotherapy was associated with decreased survival [12, 37], while others showed better pelvic control and survival rate in adjuvant chemotherapy than those who did not receive either adjuvant chemotherapy or radiotherapy after TH-BSO [11, 14, 33]. In our study, 76.7% of patients received chemotherapy or chemoradiotherapy after surgery, yet adjuvant therapy didn’t affect their survival, which was similar to previous studies [38-40]. This controversy may be due to the large proportion of FIGO I stage patients and the small sample size of our cohort. In the future, more clinical trials are needed to verify the effect of adjuvant chemotherapy on the survival of uterine sarcoma.
The prognosis of uterine sarcoma is generally poor. In previous studies, the overall 5-year survival rate was reported within a range of 27-51% [24, 25, 41]. According to our results, the 5-year survival rate was slightly higher (71%). This difference is probably due to the fact that 81.33% of them were at FIGO stage I. The following factors were found to independently contribute to the poor prognosis of uterine sarcoma: post-menopause and advanced FIGO stage in the total cohort and LMS, age >50 years, high histologic grade, and LVSI in ESS.
Our study had some limitations. First, the small sample size of adenosarcoma did not allow us to make any subgroup analyses on the adenosarcoma group. Second, the retrospective nature of the study design makes it subject to recall bias. Third, the study was performed in a single center and may lack representativeness. Last, the role of lymphadenectomy with ovarian function preservation was unclear in our study due to the limited sample.