GCIG Consensus Review: Uterine and Ovarian Leiomyosarcomas

Abstract

Objective

The GCIG aimed to provide an overview of uterine and ovarian leiomyosarcoma management.

Methods

Published articles and author experience were used to draft management overview. The draft manuscript was circulated to international members of the GCIG for review and comment, and appropriate revisions were made.

Results

The approach to management of uterine and ovarian leiomyosarcoma management is reviewed.

Conclusions

Uterine and ovarian leiomyosarcomas are rare, aggressive cancers that require specialized expertise for optimal management.

Introduction

Uterine sarcomas represent about 8% of uterine cancers, with an incidence of about 0.4 per 100,000 women¹. Leiomyosarcomas are the most common subtype; most are high grade malignancies with a high risk for recurrence and progression. Overall survival is dependent on stage with 5-year survival estimates of stage I: 76%, stage II: 60%, stage III: 45%, and stage IV: disease 29%². Uterine leiomyosarcomas are staged using the FIGO 2009 uterine sarcoma staging system, although anatomic staging systems perform poorly in terms of survival prognostication³. Other factors that have been evaluated for their potential prognostic impact include tumor morcellation⁴, mitotic index⁵,⁶, and tumor grade. A nomogram that includes additional non-anatomic prognostic factors such as patient age, tumor grade, and mitotic rate provides better estimates of overall survival⁷,⁸.

Epidemiology

Most patients with uterine leiomyosarcoma have no identifiable risk factors. Patients who carry a germline p53 gene mutation (Li Fraumeni syndrome) have an increased risk of soft tissue sarcoma, including uterine LMS, as well as other cancers⁹. Patients with Rb mutations who are survivors of childhood retinoblastoma, and survivors of childhood rhabdomyosarcoma, or other childhood cancers whose treatment involves radiation, have an increased risk secondary cancers, including uterine LMS¹⁰. The familial syndrome hereditary leiomyomatosis with renal cell carcinoma (HLRCC), in which there are germline mutations in fumarate hydratase, has also been associated with an increased risk of uterine LMS¹¹. Some studies have suggested an increased risk for uterine sarcoma among women with a history of obesity and diabetes¹², and among women exposed to tamoxifen¹³.

Pathology

Stanford criteria are commonly used to diagnosis uterine LMS, incorporating histologic atypia, tumor cell necrosis, and mitotic rate¹⁴. There is incomplete consensus regarding the grading of uterine leiomyosarcomas¹⁵. Immunohistochemistry for smooth muscle differentiation markers such as SMA and caldesmon may be used to support the diagnosis. Histologic subtypes of uterine LMS such as epithelioid and myxoid LMS may have different histologic criteria. Because of the nuances of the histologic diagnosis of uterine LMS, expert review by gynecologic pathologists and/or sarcoma pathologists is recommended.

Molecular biology and genetics

No single driving mutation has been identified in uterine LMS. Most tumors show multiple somatic chromosomal abnormalities. Genetic profiling is investigational in LMS, but could potentially elucidate treatment targets¹⁶,¹⁷. Genetic profiling may be able to improve prognostication by identifying gene signatures that differentiate indolent uterine LMS tumors from clinically aggressive tumors¹⁸.

Diagnosis

Presenting symptoms may include pelvic pain or pressure, or abnormal vaginal bleeding. Sonogram, CT, or MRI imaging may reveal a uterine mass. No single imaging criterion can reliably distinguish a benign uterine tumor from a malignant one. One small study of pre-operative MRI for patients with uterine mesenchymal neoplasms showed poor accuracy in distinguishing leiomyomas with atypical features from malignant mesenchymal neoplasms¹⁹. A separate study (19 patients with uterine mesenchymal lesions, 3 of which were LMS) suggested that MRI may be able to distinguish benign from malignant disease²⁰. Intrauterine tumors that continue to increase in size after menopause should raise suspicion for malignancy. In most patients the diagnosis of uterine LMS is made at the time of myomectomy or hysterectomy for presumed benign disease²¹,²².

Staging

Uterine sarcomas are staged using the FIGO 2009 staging system.

LEIOMYOSARCOMAS
Stage		Definition
I		Tumor limited to uterus
	IA	≤5 cm
	IB	>5 cm
II		Tumor extends beyond the uterus, within the pelvis
	IIA	Adnexal involvement
	IIB	Involvement of other pelvic tissues
III		Tumor invades abdominal tissues (not just protruding into the abdomen).
	IIIA	One site
	IIIB	>one site
	IIIC	Metastasis to pelvic and/or para-aortic lymph nodes
IV	IVA	Tumor invades bladder and/or rectum
	IVB	Distant metastases

Initial treatment

Surgery

For patients whose disease appears limited to the uterus, hysterectomy is recommended. If there is suspicion of malignancy prior to surgery, we recommend against morcellation hysterectomy because of concern for intra-operative spread of malignant tissue²³. Morcellation of malignant tumors are associated with poorer survival outcomes²⁴. Routine lymph node dissection is not generally required; however, it is recommended that lymph nodes that appear enlarged/suspicious for malignant involvement should be resected²⁵. Bilateral salpingo-oophorectomy (BSO) is reasonable in peri-menopausal and post-menopausal women, although there are no data to show that oophorectomy improves survival outcomes²⁶. Estrogen receptors and/or progesterone receptors have been reported to be positive in 40 to 70% of uterine leiomyosarcomas²⁷,²⁸, and may have prognostic significance²⁹, suggesting that oophorectomy may be reasonable even in pre-menopausal women. However, retrospective data have not shown survival differences among women under age 50 with uterus-limited disease who did or did not undergo BSO³⁰.

For disease that appears locally advanced but potentially completely resectable, an attempt to resect all disease is reasonable. Retrospective data have shown longer overall survival among women whose disease is completely resected compared to those with residual disease at end of the resection attempt³¹.

For women who present with multi-site metastatic, unresectable disease, there is not generally a role for hysterectomy. Palliative hysterectomy may be appropriate for patients with metastatic disease who have poorly controlled uterine bleeding.

Laparoscopic re-evaluation of the pelvis after morcellation hysterectomy should be considered to evaluate for, and resect, any residual malignant tissue. Resection of the cervix, and consideration of BSO if not yet performed, is reasonable for patients who had only a supracervical hysterectomy.

Post-resection management of uterus-limited disease

Although it is recognized that the risk for recurrence after resection of uterus-limited high grade LMS exceeds 50%³², no adjuvant intervention has been shown to improve progression-free or overall survival outcomes. Standard management after complete resection of uterus-limited disease is observation. Nearly one-third of patients who are found at time of hysterectomy to have uterine LMS will have evidence of metastatic disease on post-resection imaging³³ ; therefore CT and/or PET/CT and/or MRI is recommended to rule out distant metastases. PET imaging has not been shown to be superior to CT or MRI for detection of recurrent disease³⁴. PET imaging may not detect small volume lung metastases³⁵.

Adjuvant pelvic radiation was evaluated in a prospective randomized trial for women with uterine carcinosarcoma, leiomyosarcoma, or endometrial stromal sarcoma. Survival outcomes were not improved by adjuvant radiation. Among the patients with uterine LMS, there was no difference in local recurrence rates between patients assigned to adjuvant pelvic RT and those assigned to observation³⁶.

A small prospective randomized trial of adjuvant doxorubicin versus observation for uterine LMS and carcinosarcoma did not show a survival benefit for doxorubicin. A prospective phase II study of adjuvant gemcitabine-docetaxel, followed by doxorubicin, demonstrated a 2-year progression-free survival rate of 78%; however at 3 years only 58% remained progression-free³⁷. In a small study of 81 patients with a variety of uterine sarcoma histologies and stages (52 stage I, 16 stage II, 13 stage III; 53 leiomyosarcomas, 9 undifferentiated sarcomas, 19 carcinosarcomas) chemotherapy with doxorubicin plus ifosfamide plus cisplatin followed by radiation was superior to radiation alone at 3 years for disease-free survival (55% v. 41%) but not for overall survival³⁸. These data cannot be used to support a recommendation for adjuvant chemotherapy as standard treatment given the heterogeneity of the tumor types and stages and the very small sample size and no overall survival benefit. An international randomized, phase III trial of observation versus adjuvant chemotherapy (gemcitabine-docetaxel for four cycles followed by doxorubicin for four cycles) is ongoing (GOG 0277/IRCI study 001).

For patients with locally advanced, completely resected uterine LMS, there are no prospective data upon which to base management recommendations. Choices may include observation, adjuvant radiation, adjuvant hormone blockade, or adjuvant chemotherapy. The location of the disease, histologic grade, estrogen receptor and progesterone receptor status, patient preferences, organ function, and comorbidities would be incorporated into the decision.

Metastatic disease

Patients found to have metastatic disease should be evaluated to determine whether resection of metastases may be appropriate.

Potentially resectable metastatic disease

Retrospective data show that survival may be prolonged among patients who undergo resection of metastatic disease. These data have inherent patient selection bias, but nevertheless support consideration of metastasectomy for selected patients. Outcomes are more favorable for those patients who have had a long disease-free interval, have a paucity of metastatic sites, and for whom the resection is likely to render them measurably disease-free³⁹,⁴⁰,⁴¹,⁴². Radiofrequency ablation and other non-surgical, interventional radiology techniques may be appropriate for certain patients⁴³. There are no prospective studies of these interventions, nor randomized trials comparing ablation outcomes with surgical outcomes. There are no data evaluating adjuvant systemic treatment after metastasectomy. The standard approach is surveillance.

Systemic treatment options for unresectable metastatic disease

Objective response rates can be achieved with systemic treatment for metastatic uterine LMS; in patients with symptomatic disease, chemotherapy may provide palliation of symptoms. There is no established superior first line chemotherapy regimen. Treatment recommendations for an individual patient should take into consideration the patient’s preferences for the treatment schedule, drug side effects, venous access, co-morbidities, disease burden, and organ function. Reasonable regimens to consider for first-line therapy include doxorubicin, doxorubicin plus ifosfamide, gemcitabine, gemcitabine plus docetaxel. Other treatment options, used as second-line therapy or after, include pazopanib, trabectedin, dacarbazine or temozolomide. Enrollment on clinical trials is highly recommended.

Doxorubicin 60 mg/m2 every 3 weeks achieved objective response in 19% of patients with uterine sarcoma whether given as a single agent or combined with cyclophosphamide. Median overall survival was 12 months⁴⁴.

Doxorubicin plus ifosfamide achieved objective response in 30% of patients with uterine LMS⁴⁵. The choice between single agent doxorubicin versus doxorubicin plus ifosfamide should incorporate the disease burden and the patient’s risk for toxicity from dual-agent treatment.

Gemcitabine 1000 mg/m² IV over 30 minutes on a three-week on/one-week off schedule achieved objective response in 20% of patients with uterine LMS in a phase II trial⁴⁶.

Fixed dose-rate gemcitabine plus docetaxel achieved objective response in 27% of patients with uterine LMS when given as second-line therapy (90% of patients had progressed on or after doxorubicin) in a phase II trial⁴⁷. The objective response rate was 36% in the phase II trial as first-line therapy⁴⁸. A randomized trial in patients with metastatic soft tissue sarcoma showed superior objective response rates, progression-free, and overall survival among patients treated with gemcitabine plus docetaxel compared to those assigned to gemcitabine alone⁴⁹. Another randomized trial did not find a difference between gemcitabine v. gemcitabine-docetaxel but the very small sample size, and the imbalance in the treatment arms for important variables make these data difficult to interpret⁵⁰. The toxicity of gemcitabine plus docetaxel is greater than that of single agent gemcitabine.

Ifosfamide 1.5 g/m² IV for five days with Mesna, every three weeks achieved objective response in 17% of patients with uterine LMS⁵¹.

Pazopanib 800 mg oral daily achieved objective response in about 6% of patients with metastatic soft tissue sarcoma in a phase III trial. The PFS was 20 weeks with pazopanib versus 7 weeks with placebo. There was no difference in overall survival⁵².

Trabectedin 1.5 mg/m² IV over 24 hours every three weeks achieved objective response in 10% of patients with uterine LMS as first line therapy⁵³. The study was closed for failure to meet the objective response rate goal. Among the 20 patients treated, the median PFS was 5.8 months. In a retrospective study, among patients with uterine LMS who had had prior treatment, trabectedin was associated with a 16% response rate but only a 3 month PFS⁵⁴.

Trabectedin by three hour infusion plus doxorubicin yielded an objective response in 57% of patients with leiomyosarcoma of either uterine or soft tissue origin.⁵⁵

Dacarbazine and Temozolomide have modest activity in soft tissue sarcomas, and in uterine LMS, although prospective data are limited for these agents in the uterine LMS population⁵⁶,⁵⁷,⁵⁸.

Special considerations for patients with small volume and indolently paced uterine LMS

Although there are no specifically established histologic criteria by which to recognize them, clinically, there are patients whose uterine LMS disease pace is indolent⁵⁹ and/or in whom the disease burden is very low. Such patients may not need systemic cytotoxic chemotherapy. Observation may be appropriate in order to keep the toxicity of the treatment from being worse than the burden of the disease.

Hormonal blockade may be considered for these low-disease-burden/indolent disease-pace patients, if their tumors are ER and/or PR positive. In a retrospective study, aromatase inhibition treatment was associated with objective response in fewer than 10% of patients. The relatively prolonged PFS that was observed could be attributed to the inherent biology of the uterine LMS in these cases rather than to the hormonal intervention⁶⁰. A small prospective study of letrozole in ER and/or PR positive uterine LMS patients showed a 12-week progression-free survival rate of 50% with median duration of treatment being 2.2 months⁶¹.

Ovarian Leiomyosarcomas

Ovarian leiomyosarcomas arise from the smooth muscle component of the ovary or are of vascular origin⁶². These are generally considered to be high-risk cancers⁶³. There are no prospective studies upon which to base management recommendations for ovarian leiomyosarcomas. In the absence of data specific to ovarian LMS, it is reasonable to adapt recommendations from data that exist for uterine LMS.

Surgery for disease limited to the ovary

Total hysterectomy and BSO is recommended. For patients who have not had lymph node dissection or omentectomy, a second operation is not considered necessary since the probability of occult metastatic disease is likely low.

Post-resection management

For disease that is limited to the ovary, there are no prospective data to support the routine use of adjuvant chemotherapy⁶⁴. For locally advanced, completely resected disease, post resection therapy may be considered; however there are no data to show that treatment will improve survival outcomes.

Management of metastatic disease

Extrapolation of data for treatment of uterine LMS to the ovarian LMS setting is reasonable. Agents with demonstrated activity in uterine LMS and in soft tissue sarcoma are reasonable to consider as treatment options for metastatic ovarian LMS.