at SciVerse ScienceDirect

International Journal of Surgery 11 (2013) 518e523

ORIGINAL RESEARCH

Contents lists available

International Journal of Surgery

journal homepage: www.thei js .com

Original research

Uterine sarcomas: Review of 26 years at The Instituto Nacional deCancerologia of Mexico

David Cantú de León a,*,e, Heliodoro González a,e, Delia Pérez Montiel b, Jaime Coronel c,Carlos Pérez-Plasencia c,d, Verónica Villavicencio-Valencia c, Ernesto Soto-Reyes c,Luis Alonso Herrera c,d,**

aDepartamento de Ginecología Oncológica, Instituto Nacional de Cancerología de México, Av. San Fernando # 22 Col. Sección XVI, México Distrito Federal14080, MexicobDepartamento de Patología Quirúrgica, Instituto Nacional de Cancerología de México, Av. San Fernando # 22 Col. Sección XVI, México Distrito Federal14080, MexicocUnidad de Investigaciones Biomédicas en Cáncer, Instituto Nacional de Cancerología, Instituto de Investigaciones Biomédicas, Universidad NacionalAutónoma de México (UNAM), Tlalpan, Mexicod Instituto Nacional de Cancerología de México, Av. San Fernando # 22 Col. Sección XVI, México Distrito Federal 14080, Mexico

a r t i c l e i n f o

Article history:Received 19 August 2012Received in revised form13 April 2013Accepted 27 April 2013Available online 7 May 2013

Keywords:UterineSarcomaTreatmentDisease Free Survival

* Corresponding author. Tel./fax: þ52 55 56 28 04** Corresponding author. Unidad de Investigaciones BNacional de Cancerología, Instituto de InvestigacioNacional Autónoma deMéxico (UNAM), Tlalpan, Mexico

E-mail addresses: dcantude@yahoo.com, madeliapde León), herreram@biomedicas.unam.mx (L.A. Herre

e These authors contributed equally to the article.

1743-9191/$ e see front matter � 2013 Surgical Assohttp://dx.doi.org/10.1016/j.ijsu.2013.04.013

a b s t r a c t

Uterine sarcomas are a group of uncommon tumors that account for approximately 1% of malignantneoplasms of the female genital tract and between 3 and 8.4% of malignant uterine neoplasms.Objective: To evaluate the factors associated with the clinical behavior of uterine sarcomas.Materials and methods: In the period from October 1983 to December 2009, clinical files of patients witha confirmed diagnosis of uterine sarcoma at the National Institute of Cancerology of Mexico (INCan) werereviewed and evaluated.Results: We identified 77 cases with complete information; average age at presentation was 51.6 years(range, 14e78 years); most frequent histology was leiomyosarcoma (LMS) in 53/77 (68.8%) cases; mostfrequent symptom reported at the time of diagnosis was abnormal vaginal bleeding in 36/77 (46.7%)cases, and the most frequent clinical stage was clinical stage (CS) I in 31/77 (40.2%) cases. Initial treat-ment was total abdominal hysterectomy (TAH) and bilateral salpingo-oophrectomy (BSO) in 53/77(68.9%) cases. Disease-free period was 27.8 months (range, 0e184 months), with disease recurrence in33/77 (42.85%) cases, most frequent site as lung in 13/33 (39.39%) cases. Management of recurrences wassurgery and chemotherapy (CT) in 5/33 (15.15%) and CT in 10/33 (30.30%) of cases. At present, 40.3% ofthe patients (31/77) are found to be Disease-free.Conclusion: Notwithstanding that uterine sarcomas are aggressive neoplasms, most accepted manage-ment to date is TAH þ BSO, observing that the fact that this procedure is not performed by oncologistsdoes not affect the DFP nor OS, contrary to what occurs in other gynecological neoplasms.

� 2013 Surgical Associates Ltd. Published by Elsevier Ltd. All rights reserved.

1. Introduction

Uterine sarcomas are a group of infrequent neoplasms thatoccupy approximately 1% of malignant gynecologic neoplasmsand represent 3e8.4% of all malignant uterine neoplasms.1,2

25.iomédicas en Cáncer, Institutones Biomédicas, Universidad. Tel./fax:þ52 55 56 28 04 25.mg@yahoo.com.mx (D. Cantúra).

ciates Ltd. Published by Elsevier Lt

Previously, uterine sarcomas were classified by the WorldHealth Organization (WHO) as mesenchymal tumors included inthe group of the endometrial sarcoma and smooth muscle tu-mors.3 The initial classification included carcinosarcomas, whichrepresent 40% of cases, Leiomyosarcomas (LMS) in 40%, Endo-metrial stromal sarcomas (ESS) in 10e15% of cases, and undiffer-entiated uterine sarcomas in 10e15% of cases.3 In 2009, theclassification of uterine sarcomas changed modifying their stagingand histologic classification, consequently their proportional dis-tribution changed.4e8

Despite that the behavior of these tumors is aggressive; the lowpresentation of this entity and its histopathological diversity has

d. All rights reserved.

Table 1General characteristics of patients.

N ¼ 77 %

Age (years)Mean 51.6Range 14e78MenarcheMean (years) 13.2Range (years) 9e18Family history of cancer 12 (77) 15.5First grade 12Second grade 3FPMNone 58 75.3IUD 6 7.7Hormonal 7 9.3SPC 6 7.7

HRT 0 0PregnanciesMean 4Range 0e15DeliveryMean 3.92Range 0e13Cesarean sectionsMean 1.6Range 0e3AbortionsMean 1.88Range 0e5AFP (years)Mean 20.62Range 15e43

FPM: family planning method; IUD: intrauterine device; SPC: salpingoclasy; HRT:hormone replacement therapy; AFP: age at first pregnancy.

D. Cantú de León et al. / International Journal of Surgery 11 (2013) 518e523 519

ORIGINAL RESEARCH

contributed to the lack of a consensus on the optimal treatment toperform.13

At present, there is scarce evidence on the management ofuterine sarcomas due to their low incidence and histopathologicaldiversity; furthermore, the recent change in their classification hasgiven rise to the need for conducting new studies to evaluate thebest treatment alternatives in this group of uterine tumors.9e13

In Mexico, to date there are no reports related with thisneoplasm; therefore, it is of the utmost importance to show theexperience acquired in the management of this particular group ofpatients in a single reference center.

2. Materials and methods

This is an observational, retrospective study inwhich we reviewed the files of allcases diagnosed with uterine sarcomas as well as the pathology reports of the Na-tional Institute of Cancerology of Mexico (INCan) for the period fromOctober 1983 toDecember 2009. The data were obtained through an information search of patientswith a confirmed diagnosis of uterine sarcoma. The study was evaluated andauthorized by the internal review board of the Institute.

2.1. Patients

One hundred and nine patients were identified at the INCan in a 26-year period;we excluded 32 cases due to that the new uterine-sarcoma classification excludesthose with the diagnosis of carcinosarcomas; consequently 77 cases were suitablefor evaluation.

We assessed the following variables: age; family history of cancer; smokinghabits; age of menarche; parity; age at first delivery; related symptoms; clinicalstage; histological type; treatment received; disease-free period (DFP), and time andsite of recurrence, as well as treatment of the latter and overall survival (OS).

2.2. Tumor morphology

The pathology slides of cases reported as uterine sarcomas were retrieved fromthe pathology laboratory and re-reviewed by a gynecologic pathologist. Evaluationof the following variables was performed: histological type; lymphovascular inva-sion (LVI); myometrial invasion, and spread to other organs. In the case of samplesidentified lymph node tissue, the number of lymph nodes with and withoutmetastasis was evaluated. Tumor size was assessed in the majority of patients bymeasuring the surgical specimen; and cases of not possessing this information dueto the patients’ having been operated on outside of our Institute, the measurementsof the neoplasm were obtained from the pathological report sent by the outsideinstitution and/or image reports such as ultrasound or CT-scans.

2.3. Treatment

All the informationwas obtained by reviewing the clinical file, whether throughthe report of the surgical procedure performed at our Institute or by the report of thesurgical procedure conducted outside of the Institute; additionally, we reviewed thecases of patients who did not have surgery as primary treatment.

In cases in which patients received adjuvant treatment with RT, this was re-ported on the data collection sheet in detail, specifying the technique utilized, tel-etherapy, brachytherapy, or the combination of both, and the total doseadministered.

Adjuvant management with CT was evaluated in cases of patients who receivedthis, identifying the type of CT, with one or multiple drugs, and the time ofadministration.

2.4. Follow-up

Follow-up data were obtained by reviewing the patients’ clinical file. TheDisease-free period (DFP) and overall survival (OS) were defined with standardparameters. In patients lost to follow-up (whether due to a death report in themajority of cases or to the patient’s withdrawal from follow-up) or those who werefollowed-up at other institutions, this information was obtained by telephonecontact.

3. Statistical analysis

We carried out descriptive and inferential statistics includingthe Student t test, the Chi square test, or the Fisher exact test ac-cording to the case. Survival analysis was obtained by means of theKaplaneMeier test and log-rank comparison. We performedmultivariate analysis with a logistic regression test or Cox

multivariate analysis according to the case for independent factorsin the appearance of recurrence and for evaluation of OS. A value ofp <0.05 was considered statistically significant. We used the SPSSv17 software program for these purposes.

4. Results

4.1. General characteristics of the patients

General demographic characteristics of the 77 evaluated pa-tients are summarized in Table 1.

4.2. Symptoms

The most frequent symptom was abnormal genital bleeding in33/77 (46.7%) patients, other symptoms associated are shown inTable 2.

4.3. Histology, tumor size, and clinical stages

The most frequently identified histological type was LMS, pre-senting in 53/77 (68.9%) patients, followed by ESS in 11/77 patients(14.3%) and another type of uterine sarcoma in 13/77 (16.8%) pa-tients. Affectation of the lymphovascular space (LVI) was reportedin four cases (5.19%).

Tumor size was determined on all cases by the pathologicalstudy, reporting an average size of 11.9 cm (range, 1e26 cm).Regarding myometrial invasion, this was absent in 6/77 (7.8%) pa-tients, with invasion in <50% of the myometrium in 10/77 (12.9%)patients, invasion of>50% but without affectation of the entire wallin 7/77 (9.1%) patients, with invasion of the entire wall in 18/77(23.4%) patients, and in 36/77 cases (46.8%), it was not possible to

Table 2Clinical characteristics of the study population.

N %

SymptomsTVB 36 46.7Pelvic Pain 15 19.4Pelvic or abdominal mass 24 31.1Other 2 2.8Clinical stageI 31 40.3II 3 3.9III 5 6.5IV 18 23.3Unknown 20 25.9Histological typeLMS 53 68.9ESS 11 14.3Another type 13 16.8Histological gradeI 11 14.3II 3 3.9III 35 45.4Unknown 28 36.4Uterine size (cm)Average 12.9Range 7e26Tumor size (cm)Average 11.9Range 1e26Myometrial invasionWithout invasion 6 7.8<50% invasion 10 12.9>50% invasion 7 9.1Entire wall invasion 18 23.4Unknown 36 46.8

TVB: transvaginal bleeding; LMS: leiomyosarcoma; ESS: endometrial stromalsarcoma.

Table 3Treatment characteristics of the study population.

LMS ESS Other Total %

Surgical procedure (n)None 7 0 1 8 10.3TAH þ BSO 36 7 10 53 68.9Staging 4 0 1 5 6.5RH 1 1 0 2 2.6Exenteration 1 2 1 4 5.2Debulking 4 1 0 5 6.5RadiotherapyNo 33 8 5 46 59.8Teletherapy 3 2 2 7 9Brachytherapy 0 0 0 0 0Teletherapy and brachytherapy 17 1 6 24 31.2ChemotherapyNo 30 10 9 49 63.7Monodrug 3 0 0 3 3.9Multidrug 20 1 4 25 32.4

Total 53 11 13 77 100

LMS: leiomyosarcoma; ESS: endothelial stromal sarcoma.TAH D BSO: total abdominal hysterectomy þ bilateral salpingo-oophrectomy; RH:radical hysterectomy. Staging: as for endometrial carcinoma staging procedure.

D. Cantú de León et al. / International Journal of Surgery 11 (2013) 518e523520

ORIGINAL RESEARCH

determine myometrial invasion. Distribution by Clinical stage (CS)are summarized in Table 2.

4.4. Treatment

We recorded the different treatments to which the patientswere initially submitted, it is noteworthy that eight (10.3%) patientsreceived no treatment, this due to that these patients opted not toreceive any.

Initial treatment in 69/77 (89.6%) patients was surgery. Totalabdominal hysterectomy and bilateral salpingo-oophrectomy(TAH þ BSO) was performed in 53/77 (68.9%) cases; in five cases(6.5%) pelvic lymphadenectomy (LDN) plus additional para-aorticlymphadenectomy was carried out; in two cases (2.6%), Radicalhysterectomy (RH) was performed; in four cases (5.2%), total pelvicexenteration was performed, and in five cases (6.5%), tumordebulking was conducted, as shown in Table 3.

In cases were LND was performed, none had metastatic disease.

Fig. 1. Overall survival.

4.5. Adjuvant treatment

Thirty one patients were administered RT as adjuvant treat-ment, in 28/77 (36.6%) patients, CT was administered, as shown inTable 3. Schemes had changed during time, thus, we found thatfrom 1983 to 1995, the most common utilized scheme was dacti-nomycin/adriamycin, and after that period, the schemes have var-ied, with the currently most utilized scheme being the combinationof paclitaxel/gemcitabine, while the ifosfamide/adriamycin schemeis the second most employed. It is noteworthy that one patient(1.2%) received hormonal therapy after being reported with diseaseprogression after CT administration; this case was diagnosed as ESS

and was managed with anastrazole for 17 months until diseaseprogression.

On performing a comparison among the different treatmentsthat have been administered over time in relation to the differenthistological strains, it was not possible to find a statistically sig-nificant difference among them (p ¼ 0.376).

4.6. Follow-up

Average follow-up was 35.1 months (range, 1e184 months)(Fig. 1) and the average Disease-free period (DFP) was 27.8 months(range, 0e184 months) (Fig. 2).

Recurrent disease presented in 33 cases (42.85%), the mostfrequent sites of recurrence presentation were lung in 13/33(39.39%) patients, retroperitoneum in 4/33 (12.12%), liver in 4/33(12.12%), and abdominal cavity in 2/33 (6%) patients, and in 7/33(21.21%) patients disease recurrence presented at other sites.

Treatments administered for recurrence were heterogeneousdifferent CT schemes were used in 10/33 (30.30%) cases; 5/33(15.15%) patients were treated with residual disease resection

Fig. 2. Disease free survival.

Fig. 4. Overall survival by histology.

D. Cantú de León et al. / International Journal of Surgery 11 (2013) 518e523 521

ORIGINAL RESEARCH

(debulking); 3/33 (9%) patients were treated with RT, and 15/33(45.45%) cases received no treatment because the patients decidednot to attend to the hospital again.

We found 31/77 (40.3%) patients who were alive and withoutevidence of disease at follow-up, 8/77 (10.3%) patients who werealive with disease, and 38/77 (49.3%) patients who had died due tothe disease.

The DFP was similar in relation to the histology of the neoplasm,as shown in Fig. 3 (log rank, p ¼ 0.789), as OS was similar in thedifferent histologies, as depicted in Fig. 4 (log rank, p ¼ 0.063).

It is noteworthy to evaluate and mention the group of patientswho had surgery (TAH þ BSO) at other institutions in whom thepresence of a mesenchymal neoplasm at the time of evaluationwasnot suspected. Our interest in this group of patients was to seewhether the recurrence rate was greater in comparison to that of

Fig. 3. Disease free survival by histology.

patients initially treated at an oncological center. In Fig. 5, DFP isshown by clinical stage, finding that there is no statistical differenceamong the different groups assigned as early, locally advanced, andmetastatic, or unknown stages, the latter of these having been thegroup of patients treated at other institutions (log rank, p ¼ 0.742),this allocationwas considered because the few cases on every strataof the FIGO staging. In Fig. 6, a figure is presented that demonstratesOS, which found that there is a statistical difference among thedifferent groups when these are analyzed by clinical stage, with

Fig. 5. Disease free survival by clinical stage.

Fig. 6. Overall survival by clinical stage.

D. Cantú de León et al. / International Journal of Surgery 11 (2013) 518e523522

ORIGINAL RESEARCH

early stages demonstrating better survival, as expected (log rank,p ¼ 0.002).

We carried out Cox multivariate analysis in an attempt toidentify whether there are variables that can exert an influence onthe Disease-free period (DFP), any of the variables studied prove tobe independent factors.

5. Discussion

Uterine sarcomas comprise a rare and heterogeneous group ofneoplasms arising in the mesenchymal cells of the uterus thatexhibit aggressive behavior; their estimated incidence is 1.55 per100,000 women, and they occupy 1e3% of gynecological tumorsand 3e7% of all uterine tumors.14e16 To date, the data obtained arederived from studies conducted at other institutions with a smallnumber of patients; our study is based on the experience of areferral center during 26 years and includes 77 patients. This studycontributes with relevant information on the result of initialtreatment with or without adjuvant treatment in our milieu,reflecting our experience in the management of these tumors.17,18

In Israel in 2011, a study was reported in which 40 patients withuterine sarcomas were analyzed, finding that the most importantprognostic factor for this neoplasm is the clinical stage, which issimilar to our findings; that is, patients found at the most advanced

Table 4Treatment employed in different series.

Patients (n) Age (years) Surgical procedure (TAH þ BSO

Geller 2004 28 54.4 21Kapp 2008 1396Cheng 2011 74 43.5 38D’Angelo 2011 84 51 68Naaman 2011 40 53 39This study 77 51.6 53

TAHD BSO: total abdominal hysterectomy plus bilateral salpingo-oophrectomy; LDN: lymtherapy; DFP: disease free period; NS: not shown.

stages correspond to patients who in the end died within a shorttime period.18

Standard procedure for staging uterine sarcomas includes Totalabdominal hysterectomy (TAH) with Bilateral salpingo-oophrectomy (BSO), peritoneal cytology, and biopsy of suspiciousareas of affectation.7 Staging is currently the most importantprognostic factor among all histological types; the most contro-versial aspect of surgical treatment is the role of optimal cytor-eduction in advanced disease, the need for performing lymph nodeLymphadenectomy (LDN), and that of preserving the ovaries. Insome studies, the effect of cytoreduction has been evaluated inpatients with disseminated disease; to date, the results are incon-sistent.19 In our study, we coincide with standard staging man-agement, because the great majority of our patients were treatedwith TAH þ BSO. On the other hand, performing routine LDN iscontroversial; at our center, this is not carried out routinely, only 5/77 (6.49%) cases were submitted to LDN, and no case reportedpositive lymph nodes. Thus, lymphadenectomy is not performedroutinely in cases of uterine sarcomas; this is supported by the factthat no evidence has been found that demonstrates a benefit inperforming the procedure.

The cornerstone of treatment for uterine sarcomas is surgery,while the use of Radiotherapy to date is controversial, in our study,we analyzed 31 patients who received RT, of whom 24 receivedcombined therapy (teletherapy and brachytherapy); the dosesadministered ranged between 35 and 50 Gy. However, improve-ment in OS was not shown in patients who received RT because ofthe great heterogeneity, with these results similar to those in re-ports in the literature.11,20

Management with systemic therapy in uterine sarcomascurrently has fallen in terms of its level of evidence, this is due tothat the carcinosarcoma has been eliminated from the newestclassification and has come to be treated as endometrial epithelialtumor, resulting in the low frequency of uterine sarcomas. There-fore, we are able to observe that in the past two decades, theoptimal scheme has not been established for treating these tumors.What has been observed is that schemes that employ adriamycinprevail over other schemes, whether as monodrug or in combina-tion, which coincides with our study.21,22

In Table 4, it is possible to observe that the series reported in theliterature do not included large numbers of patients.6,18,23e25 Whenour series is compared with those of other authors, we are able toobserve that age at disease presentation is very similar, lympha-denectomy is not part of the surgical procedure, even though itcontinues to be performed in nearly one third of the reported cases;at our Institute, the procedure is carried out in no more than 5% ofcases diagnosed with this neoplasm.

It is noteworthy that these neoplasms are considered as highlyaggressive and that the follow-up to be applied is broad, reachingup to several years; in our study, average DFS is 27 months.

On analyzing the cases with regard to DFP, we found no signif-icant difference on comparing clinical stage and histology; how-ever, this was not so when analyzing OS because, as expected,

) LDN CT RT HRT DFP

13 (46%) 3 14 e 9 months (range, 8e52 months)348 (25) NS18 (24) 3 6 25 168 months12 (14%) 18 13 e 3.5 years (3 monthse8 years)11 (27%) 10 16 44 months3 (3.8%) 28 31 1 27.8 months (0e184 months)

phadenectomy; CT: chemotherapy; RT: radiotherapy;HRT: hormonal replacement

D. Cantú de León et al. / International Journal of Surgery 11 (2013) 518e523 523

ORIGINAL RESEARCH

better survival presents in sarcomas with better prognosis, such asthe endometrial stromal sarcoma (ESS), with the worst prognosisfor leiomyosarcomas (LMS).

In conclusion, it is considered that uterine sarcomas are neo-plasms with aggressive behavior. Diagnosis of these is difficult toestablish prior to surgery due to that there are no studies thatreport characteristic findings. Standard management to date con-tinues to be Total abdominal hysterectomy plus Bilateral salpingo-oophorectomy. Lymph node sampling is not carried in routinefashion. Adjuvant management with chemo- or radiotherapy is notestablished routinely because, up to the present, this has not shownimprovement in overall survival, this perhaps due to that in pastdecades and to the present day, there has not been a scheme thatshows these benefits, in turn probably due to the low frequency ofuterine sarcomas, due to which it is necessary to conduct moreprospective studies.

Ethical approvalSince this is a retrospective study no ethical issues are violated.The study was approved by the internal review board of the

institution.

FundingNo funding was necessary for conducting the study.

Author contributionDavid Cantú de León: Study Design, Data analysis and Writing.Heliodoro González: Study Design, Data analysis and Writing.María Delia Pérez Montiel: Writing and Data Collection.Jaime Coronel: Writing and manuscript Review.Carlos Pérez-Plasencia: Manuscript Review and Data review.Verónica Villavicencio-Valencia: Data analysis.Ernesto Soto-Reyes: Data collection.Luis Alonso Herrera: Study Design and Manuscript review.

Conflict of interestNo conflicts of interest.

References

1. Seddon BM, Davda R. Uterine sarcomas e recent progress and future chal-lenges. Eur J Radiol 2011;78:30e4.

2. Ueda SM, Kapp DS, Cheung MK, Shin JY, Osann K, Husain A, et al. Trends indemographic and clinical characteristics in women diagnosed with corpuscancer and their potential impact on the increasing number of deaths. Am JObstet Gynecol 2008;198:218e26.

3. World Health Organization classification of tumours. In: Tavassoli FA, Devilee P,editors. Pathology and genetics of tumours of the breast and female genital organs.Lyon, France: IARC Press; 2003.

4. Amant F, Moerman Neven P, Timmerman D, Van Limbergen E, Vergote I.Endometrial cancer. Lancet 2005;366:491e505.

5. Amant F, Coosemans A, Biec-Rychter M, Timmerman D, Vergote I. Clinicalmanagement of uterine sarcomas. Lancet Oncol 2009;10:1188e98.

6. D’Angelo E, Prat J. Uterine sarcomas: a review. Gynecol Oncol 2010;116:131e9.7. Gadducci A, Cosio S, Romanini A, Genazzani AR. The management of patients

with uterine sarcoma: a debated clinical challenge. Crit Rev Oncol Hematol2008;65:129e42.

8. Lin JF, Slomovitz BM. Uterine sarcoma 2008. Curr Oncol Rep 2008;10:512e8.9. Nam JH, Park JY. Update on treatment of uterine sarcoma. Curr Opin Obstet

Gynecol 2010;22:36e42.10. O’Cearbhaill R, Hensley ML. Optimal management of uterine leiomyosarcoma.

Expert Rev Anticancer Ther 2010;10:153e69.11. Reed NS. The management of uterine sarcomas. Clin Oncol (R Coll Radiol)

2008;20:470e8.12. Zagouri F, Linardou H, Dimopoulos AM, Papadimitriou CA. Management of

advanced stage uterine sarcomas: a bone of contention. Eur J Gynaecol Oncol2009;30:483e92.

13. Giuntoli II RL, Metzinger DS, DiMarco CS, Cha SS, Sloan JA, Keeney GL, et al.Retrospective review of 208 patients with leiomyosarcoma of the uterus:prognostic indicators, surgical management, and adjuvant therapy. GynecolOncol 2003;89:460e9.

14. Brooks SE, Zhan M, Cote T, Baquet CR. Surveillance, epidemiology, and endresults analysis of 2677 cases of uterine sarcoma 1989e1999. Gynecol Oncol2004;93:204e8.

15. Giuntoli II RL, Bristow RE. Uterine leiomyosarcoma: present management. CurrOpin Oncol 2004;16:324e7.

16. Toro JR, Travis LB, Wu HJ, Zhu K, Fletcher CD, Devesa SS. Incidence patterns ofsoft tissue sarcomas, regardless of primary site, in the surveillance, epidemi-ology and end results program, 1978e2001: an analysis of 26,758 cases. Int JCancer 2006;119:2922e30.

17. Pautier P, Genestie C, Rey A, Morice P, Roche B, Lhommé C, et al. Analysis ofclinicopathologic prognostic factors for 157 uterine sarcomas and evaluation ofa grading score validated for soft tissue sarcoma. Cancer 2000;88:1425e31.

18. Naaman Y, Shveiky D, Ben-Shachar I, Shushan A, Mejía-Gómez J, Benshushan A.Uterine sarcoma: prognostic factors and treatment evaluation. Isr Med Assoc J2011;13:76e9.

19. Leath III CA, Huh WK, Hyde Jr J, Cohn DE, Resnick KE, Taylor NP, et al. A multi-institutional review of outcomes of endometrial stromal sarcoma. GynecolOncol 2007;105:630e4.

20. Reed NS, Mangioni C, Malmström H, Scarfone G, Poveda A, Pecorelli S, et al.Phase III randomised study to evaluate the role of adjuvant pelvic radiotherapyin the treatment of uterine sarcomas stages I and II: a European Organizationfor Research and Treatment of Cancer Gynaecological Cancer Group study(protocol 55874). Eur J Cancer 2008;44:808e18.

21. Odunsi K, Moneke V, Tammela J, Ghamande S, Seago P, Driscoll D, et al. Efficacyof adjuvant CYVADIC chemotherapy in early-stage uterine sarcomas: results oflong-term follow-up. Int J Gynecol Cancer 2004;14:659e64.

22. Hensley ML, Ishill N, Soslow R, Larkin J, Abu-Rustum N, Sabbatini P, et al.Adjuvant gemcitabine plus docetaxel for completely resected stages I-IV highgrade uterine leiomyosarcoma: results of a prospective study. Gynecol Oncol2009;112:563e7.

23. Geller MA, Argenta P, Bradley W, Dusenbery KE, Brooker D, Downs Jr LS, et al.Treatment and recurrence patterns in endometrial stromal sarcomas and therelation to c-kit expression. Gynecol Oncol 2004;95:632e6.

24. Kapp DS, Shin JY, Chan JK. Prognostic factors and survival in 1396 patients withuterine leiomyosarcomas: emphasis on impact of lymphadenectomy and oo-phorectomy. Cancer 2008;112:820e30.

25. Cheng X, Yang G, Schmeler KM, Coleman RL, Tu X, Liu J, et al. Recurrencepatterns and prognosis of endometrial stromal sarcoma and the potential oftyrosine kinase-inhibiting therapy. Gynecol Oncol 2011;121:323e7.