Abstract
Aim: To (i) determine prognostic factors after initial surgery for uterine leiomyosarcomas (LMS) and their recurrence and (ii) assess the effectiveness of re-resections. Patients and Methods: All cases that underwent surgery for uterine leiomyosarcomas at the Fundeni Clinical Institute, Bucharest between 2002 and 2013 were reviewed. Twenty-six patients with primary uterine leiomyosarcomas were introduced in our study. Sixteen of them were re-addressed to the same hospital for recurrence. Results: At the moment of initial surgery the most important prognostic factors were age <60 years, International Federation of Gynecology and Obstetrics (FIGO) stage I, tumor dimension <15 cm and mitotic index <15/10 high-power fields (HPF). The five-year overall survival was 40%. Sixteen patients were re-operated for recurrence; the most important prognostic factors being a late recurrence (>12 months) and initial FIGO stage I. The five-year overall survival was 12.5%. Conclusion: Uterine leiomyosarcoma is an aggressive malignancy with a high rate of recurrence. In selected cases surgery may be attempted for re-recurrence.
Leiomyosarcomas (LMS) of the uterus is a rare malignancy accounting for approximately 1% of gynecological cancers and carries an extremely poor prognosis. (1) Survival is dependent upon stage at diagnosis (2), but overall results are poor with a 5-year overall survival ranging between 52 and 75% for stage I and 0-39% for stages II-IV (3, 4).
For early-stage LMS simple hysterectomy is warranted. The incidence of ovarian metastases is low (3.4-3.9%) (5, 6, 7) and preservation of ovarian tissue does not seem to increase the risk of recurrence (8, 9, 10). Lymph node metastases are not common (3.5-11%) (5, 11) and lymphadenectomy does not seem to affect survival (12). The major problem with LMS, even in early stages, is relapse. Radiation therapy, at least when used alone, does not seem to affect survival; the European Organization for Research and Treatment of Cancer (EORTC) randomized control trial for early-stage sarcoma (radiation versus no further treatment) failed to show improvement of survival or local control for patients with LMS (13). Several other studies showed improvement in pelvic control after adjuvant radiotherapy (5).
The effect of chemotherapy is modest with single-agent response ranging from 9 to 25% (14, 15); as far as combination is concerned, the best response is obtained by a combination of gemcitabine and docetaxel (27% response rate) (16).
Treatment of recurrent disease is mainly based on resection of both local and systemic recurrence with pulmonary metastasectomy (17, 18, 19), liver metastasectomy (20, 21) and even re-resection (17, 22).
Due to the rarity of disease, most studies are retrospective, with a small number of patients making, thus, prospective randomized studies difficult. We aimed to review cases of uterine LMS treated at a tertiary referring center (Fundeni Clinical Institute) between 2002 and 2014 from primary surgery to quaternary cytoreduction.
Associated resections at the time of initial surgery.
Patients and Methods
We retrospectively reviewed the hospital medical records of Dan Setlacec Center of General Surgery and Liver Transplantation, Fundeni Clinical Institute, Bucharest between January 2002 and April 2014. Twenty-six patients were identified to have histopathological findings of primary uterine LMS and sixteen of them were re-operated for recurrent tumors at the Fundeni Clinical Institute. Survival curves, overall survival and disease-free survival, were generated using the the Kaplan-Meyer method. A p-value <0.05 was considered statistically significant.
After approval by the Fundeni Clinical Institute Ethics Committee, the files of the patients diagnosed with uterine LMS between January 2002 and April 2014 were retrieved. The diagnosis of uterine LMS was confirmed by histopathological exam. Date of death was confirmed with the National Register of Population. Statistical analysis was performed using the application SigmaPlot version 12.1 (link or supplier with address).
Results
Initial surgery. A total of twenty-six patients with uterine sarcomas were eligible for our study. The median age at initial surgery was 53.26 years (range=32-83). Nineteen (73%) of the patients had a post-menopausal status. The most common symptoms were vaginal bleeding, 19 patients (73%); abdominal pain,15 patients (57.6%); and palpable abdominal mass, 11 patients (42.3%). Only five patients (19.3%) were diagnosed preoperatively through bioptic curettage with uterine sarcomas. Associated comorbidities were diabetes mellitus in four cases and arterial hypertension in ten cases. Thirteen patients were assigned to stage I (50%), nine cases to stage II (34.6%) and four cases to stage IV (15.3%), according to the 2009 FIGO classification for uterine sarcomas. None of the patients presented ascites at the moment of diagnosis. Neo-adjuvant chemo-irradiation was associated in three cases.
In twenty-five patients, surgical treatment consisted in total hysterectomy with bilateral salpingo-oophorectomy. Lymph node dissection was associated in five cases (pelvic lymph node dissection); para-aortic lymph node dissection was associated in four cases. In none of the cases lymph nodes were invaded. Other associated resections are summarized in Table I.
The main characteristics of patients after initial surgery influencing survival at univariate analysis.
The median hospitalization stay was 8.5 days (range=6-25 days). The postoperative mortality was 3.8% (abdominal sepsis after enteral fistula, 1 case). Postoperative morbidity was also 3.8%-wound infection in one case .
Six patients (14.28%) with advanced stage and high grade sarcoma underwent adjuvant chemotherapy. In one case radiotherapy was also associated.
Overall survival after initial surgery was 51 months (range=1-133 months). When classified on FIGO stages, overall survival was 75 months (range=20-133) for stage I, 36 months (range=7-98) for stage II and 6.4 months (range=1-22) for stage IV. The 5-year overall survival was 40%.
The main characteristics of patients with uterine LMS after initial surgery influencing survival at univariate analysis are presented in Table II. Survival analysis was performed on the remaining lot of twenty-five patients (after excluding the early postoperative death after enteral fistula).
The most important factors associated with an improved overall survival after initial surgery were age <60 years (p=0.007), FIGO stage I (p=0.00191 when compared to FIGO stage II and p=0.0000235 when compared to FIGO stage IV, respectively) (Figure 1), tumor size <15 cm (p=0.015) and mitotic count <15 (p=0.001). Although other factors like the tumoral grade of differentiation were not statistically significant in terms of survival, important differences between the three groups (G1 versus G2 or versus G3) were found. Overall survival in the G1 group was 99 months and significantly decreased in the other groups (75 months for G2 and 35 months for G3, respectively).
Associated resections at the time of secondary surgery.
Surgery for recurrence. A total of sixteen patients underwent surgery for abdominal or pelvic recurrence. The mean age at secondary surgery was 56.6 years (range=38-73). Disease-free survival after initial surgery was 35 months (range=3-117 months).
Single-tumor recurrence was found in thirteen patients (81.25%), while the other three patients had at least two lesions. Eleven patients were diagnosed solely with pelvic recurrence, while in other three cases both pelvic and upper abdominal recurrence was found. One case presented only an upper abdominal metastasis and another one presented an isolated pulmonary metastasis. The overall survival at two and five years was 43.75% and 12.5%, respectively.
Complete cytoreduction was performed in 14 cases (87.5%) and included abdominal, pelvic and thoracic resections. In two cases incomplete R1 resections were performed. Isolated resections were performed in one case with pulmonary metastasis and in another case with isolated liver metastases. All other patients benefited from extended combined abdominal and pelvic resections. Associated resections at the time of secondary surgery are noted in Table III.
The mean dimension of resected tumor was 13 cm (range=2-30 cm). The distribution according the tumoral grade revealed: G1 tumor in four cases, G2 in eight cases and G3 in three cases. None of the patients presented changed type of tumor differentiation when compared with primary surgery.
Pelvic lymph node dissection was associated in four cases but nodal involvement was found in a single patient with G3 LMS. Para-aortic lymph node dissection was also associated in five cases, while two patients with G1 and G2 LMS presented positive lymph nodes, respectively. The median hospital stay was 10 days (between 5 and 21 days).
Postoperative mortality was (6.25%): one patient died in the early postoperative course due to colic necrosis-abdominal sepsis. Postoperative morbidity was 25%, with four cases presenting postoperative complications: wound infection in 2 cases and urinary incontinence also in 2 situations. In one case Clavien Dindo grade III complications occurred: surgical re-exploration was needed for gastric perforation through gastric ulcer. The postoperative course after re-exploration was uneventful, with a reported survival of 12 months.
The most important factors influencing survival in cases in which complete R0 resections were performed.
Associated visceral resections at the time of tertiary cytoreduction.
The overall survival after secondary surgery was 33 months (range=7-116 months).
After resection for recurrence, the most important factors associated with an improved survival are FIGO stage I at initial surgery (stage I vs. stages II-IV, p=0.02) (Figure 2) and time to recurrence (disease-free interval above twelve months is associated with better prognosis, p=0.035) (Figure 3). There were important differences in terms of survival when comparing the groups with single versus multiple recurrences (61 months vs. 23 months), pelvic versus abdominal localization of the tumor (53 months vs. 29 months), association of adjuvant chemo-irradiation (54 months in patients with adjuvant chemoirradiation and only 30 months in those with no oncologic treatment) or the different histopathological grades (54 months for G1, 36 months for G2 and 16 months for G3); however, none of these factors proved to be statistically significant. The overall survival at 2 years was 43.75%, while the 5-year overall survival was 12.5%.
The most important factors influencing survival in cases in which complete R0 resections were performed are shown in Table IV.
Eight patients recurred after secondary surgery, with mean age at tertiary surgery being 52.2 years (range=39-61). Disease-free survival after secondary surgery was 15.3 months (range, 7-42).
Six patients presented pelvic recurrence, while isolated upper abdominal recurrence was found in two cases: hepatic recurrence in one case and pancreatic recurrence in the other.
Six of the eight patients (75%) underwent complete cytoreduction, while in two cases only biopsy was performed. Associated visceral resections at the time of tertiary cytoreduction are mentioned in Table V.
The mean hospitalization stay was 14 days (range, 6-30). No postoperative death was reported. Two of the eight patients presented postoperative complications: Clavien Dindo grade I (one case, wound infection) and Clavien Dindo grade III (one case, enteral fistula), which required surgical re-exploration and ileostomy. The overall survival after tertiary surgery was 17 months (range=1-35).
Two patients were diagnosed with abdomino-pelvic recurrences after tertiary surgery. Their mean age was 60 years (range=55-65). In both patients colic and segmentary enteral resections were associated. One of the two patients died in the early postoperative period, two weeks after surgery, due to neurological disorders, i.e. ischemic vascular stroke. Survival for the other patient was 14 months.
Discussion
Uterine LMS carries a grim prognosis; the body of evidence concerning this disease is scarce, mostly consisting of small retrospective studies. There is a need to better-establish prognostic factors and principles of management of the recurrence. Also, benefit of surgery beyond first recurrence needs to be assessed.
Medical charts of Dan Setlacec Center of General Surgery and Liver Transplantation, Fundeni Clinical Institute, Bucharest between 2002 and 2014 (twelve years) were reviewed; twenty-six patients with uterine LMS were identified. The vast majority were stage I and II, the rest of them being assigned to stage IV. Standard procedure at primary surgery was total hysterectomy with bilateral oophorectomy. Despite the early stages, most of them (15/25, 60%) realpsed, which is consistent with current literature indicating a relapse often above 60% (11, 23).
One of the most important factors associated with outcome after surgery for uterine LMS is stage; the overall survival for stage I is 103 months, for stage II 47 and only 5 months survival for stage IV. The survival in stage I has statistical significance when compared to both stage II and IV. This data is consistent with conclusions of other investigators; Kapp et al. (24) on a SEER analysis of 1,396 cases of uterine LMS established stage as an independent prognostic factor for survival.
Age (>60) proved to be a statistically significant factor for worse prognosis, consistent with already existing evidence (10, 11, 25, 26).
Tumor size under 15 cm and mitotic count below 15 mitoses proved statistically significant for better survival as previously described by most investigators (5, 27-30).
Although not reaching statistical significance, a difference was noted regarding tumor grading with an overall survival of 99 months for G1 group versus 35 months for G3 group.
Management of recurrent disease is an important matter in a malignancy that, as pointed above, carries a strong tendency for relapse even in early stages. Randomised control prospective studies are lacking due to the rarity of the disease. Sixteen of the initial 26 patients with uterine LMS underwent surgery for relapse in the same centre. The overall survival at two years after secondary cytoreduction was 43.75% with 12.5% alive subjects at 5 years.
The most important prognostic factors after resection for recurrence in our study was a disease-free interval of over 12 months after initial surgery that is statistically associated with better survival; this probably reflects a more indolent course in tumor biology.
FIGO stage I at initial surgery represents a different subset of patients with better outcome after surgery for recurrence. Hoang et al. (31) have made similar observations. Although not statistically significant, we found a difference in survival favouring patients with single versus multiple recurrence (similar to studies by Hoang and Giuntoly) (23, 31) and pelvic versus upper abdominal recurrence (confirmed by Giuntoly) (23).
Interestingly, patients who underwent chemo-irradiation after initial surgery had better prognosis after resection of recurrence (54 versus 30 months), although statistically significance has not been reached.
The use of radio-chemotherapy in the recurrence after LMS brings little survival benefit as reflected by most studies. On the other hand, there is important evidence to support extended use of surgery in management of both local recurrence and distant metastases, including lung (17-19, 32) and liver (20, 21). It would, thus, seem logical to give the same credit to surgery beyond the first recurrence, although existing data is extremely scarce.
Eight patients from the same lot underwent tertiary surgery in the same centre with a good resecability rate (75%) and acceptable complication rate (25%). The overall survival after tertiary surgery was 17 months (range=1-35). Beyond tertiary surgery, there is a problem of individual evolution, patient and surgeon's choice. Nevertheless, one patient with quaternary surgery survived 14 months.
Survival analysis on FIGO stages after the initial surgery.
Survival after resection for recurrence according to FIGO staging.
Conclusion
Uterine LMS is a rare but very aggressive uterine malignancy with high risk of recurrence even after complete initial resections. In our study the most important prognostic factors after initial surgery were age <60 years, FIGO stage I and smaller tumors (<15 cm) with a lower mitotic index (<15/10 high-power fields (HPF)). Although complete R0 resection was performed in all cases at the moment of initial surgery, in sixteen patients the disease relapsed. Our retrospective study showed that, at the moment of resection for recurrent tumors, the most important factors improving survival were the moment of recurrence (>12 months) and initial FIGO stage I. In fourteen cases complete resections were performed. The overall survival after recurrence resection was 43.75% at 2 years and 12.5% at 5 years, respectively. These results sustain the role of surgery in treating recurrent LMS, showing an important benefit in terms of overall survival when compared with patients who were only chemotreated and in whom the median survival reported in literature ranges between 9 and 17.9 months.
The influence of the time to recurrence on overall survival.
Beyond resections for first recurrence, we further identified patients with secondary and even tertiary relapses who benefited from complete resection of the re-recurence. We report an overall survival of 17 months after re-resetion of the recurrences and a survival of 14 months in the case of a patient who underwent quaternary cytoreduction. Although improved survival after resection was obtained in almost all the cases, our study was limited by the small number of patients who benefited from tertiary or even quaternary resections.
Footnotes
-
This article is freely accessible online.
- Received September 13, 2014.
- Revision received December 17, 2014.
- Accepted December 23, 2014.
- Copyright© 2015 International Institute of Anticancer Research (Dr. John G. Delinassios), All rights reserved
