An Introduction to Uterine Leiomyosarcomas
Uterine sarcomas are a rare and aggressive form of uterine cancer. They arise from the endometrial lining or the myometrium in the uterus. Compared to the more common endometrial carcinomas, uterine sarcomas behave more aggressively and are associated with a poorer prognosis.
In this review, we shall focus on one subset - uterine leiomyosarcomas (ULMS), although it is noteworthy to mention that leiomyosarcoma can arise from other gynecological primary sites. ULMS are rare smooth muscle tumors accounting for approximately 1% of patients with uterine cancer1 with an estimated annual incidence of 0.64 per 100,000 women.2 ULMS are considered neoplasms of high metastatic potential with 5-year overall survival rates varying between 0 and 73%.3-5 These discrepancies may be attributable to inconsistent definitions and variable sample sizes for diagnostic criteria. Additionally, this adds to the dilemma of addressing survival rates given the variable time periods of these studies.
ULMS occur primarily in women 40 to 60 years of age. The most frequent presenting symptoms are abnormal vaginal bleeding and pelvic or abdominal pain. The amount of bleeding ranges from spotting to menorrhagia and is often associated with foul-smelling vaginal discharge. Less common symptoms include weight loss, weakness, lethargy, and fever.5-7 On pelvic examination, the uterus is often enlarged, and in some cases part of the tumor may prolapse through the cervical os and into the vaginal canal. Diagnosis is usually not made before surgery, thus many patients present with advanced disease.
The rarity of these tumors has prevented the performance of large epidemiological studies to identify risk factors. Data regarding parity, onset of menarche, or age at menopause as risk factors are inconclusive. Based on available United States data, there is approximately a two- to three-fold higher incidence of ULMS among African-American women compared to Caucasian women.2,8,9 A history of pelvic irradiation is noted in 5-10% of patients.10
Variable size, usually 3-5cm
Large, often >10cm
Firm, whorled surface
Soft, fleshy cut surface
Yellow or tan
Hemorrhage and necrosis infrequent
Hemorrhage and necrosis frequent
The belief that the risk of ULMS is elevated among women with a "rapidly growing" uterus or leiomyoma was proven false in a study of 1322 women admitted to two community hospitals for hysterectomy or myomectomy. Fibroids rarely, if ever, degenerate into ULMS.12
Tissues and Cells
The Gynecologic Oncology Group (GOG) uses a classification scheme for uterine sarcomas that divides them into five categories:
- Mixed homologous mullerian sarcoma
- Mixed heterologous mullerian sarcoma
- Endometrial stromal sarcoma
Homologous refers to similarity to endometrial stroma or myometrium, while heterologous indicates similarity to other cell types, including fat, muscle, etc. Malignant mixed mullerian tumors, now called carcinosarcomas, arise from endometrial adenocarcinoma, but resemble sarcoma on histology.
The typical gross appearance is a large (>10cm), poorly circumscribed mass with a soft, fleshy consistency and a variegated cut surface that is grey-yellow to pink, with foci of hemorrhage and necrosis.13 The histologic classification of uterine sarcomas is based upon homology to normal cell types and include ULMS (analogous to myometrium), stromal sarcoma (analogous to endometrial stroma), and other heterologous cell types (i.e., osteosarcoma, liposarcoma). See Figures 2 and 3.
Microscopically, most ULMS are overtly malignant, with hypercellularity, coagulative tumor cell necrosis, abundant mitoses (>10 to 20 mitotic figures (mf) per 10 high power fields (hpf)), atypical mitoses, cytologic atypia, and infiltrative borders. Mitotic rate is the most important determinant of malignancy, but is modified by the presence of necrosis and cytologic atypia. The diagnosis of ULMS may be made in the presence of tumor necrosis and any mitoses. In the absence of tumor necrosis, the diagnosis can be made with moderate to severe cytologic atypia and a mitotic index greater than 10mf/10hpf. Without tumor necrosis and significant atypia, a high mitotic index is compatible with a benign clinical course, however, data is limited.14 See Definitions Sidebar. See Figures 4, 5, and Table 2.
One study looked at the expression of particular markers that are of interest in gynecological cancers (p53, Epidermal Growth Factor, and Platelet Derived Growth Factor) in tissue samples from patients who had ULMS or benign leiomyomas. Their data demonstrated significant and molecular differences between benign and malignant smooth muscle tumors of the uterus. The study also suggested a prognostic interrelationship between expression of p53 and stage in ULMS.15
A subset of smooth muscle tumors will not be easily classified based on the criteria and are designated as smooth muscle tumors of uncertain malignant potential (STUMP). The literature is unresolved on whether special studies such as proliferation index or stains for p53 add to the discriminating power of the basic criteria of mitoses, necrosis and cytologic atypia in determining the malignant potential of STUMP lesions. In practice, however, these stains are uncommonly used, and definitive diagnosis of sarcoma is never reported based on these stains alone.16,17
Limited data has allowed some tumors, formerly classified as STUMP, into the leiomyoma category and should be distinguished from their sarcomatous counterparts.
Tumors now in the leiomyoma category include: mitotically active, cellular, epithelioid, myxoid, atypical (pleiomorphic, bizarre, or symplastic) tumors. Mitotically active leiomyomas can occur in pre-menopausal women and have the typical macroscopic and histologic appearance of a leiomyoma with the exception that they have > 5mf/hpf. Cellular leiomyomas tend to have hypercellularity and can suggest the diagnosis of ULMS, but they lack tumor cell necrosis, cytologic atypia and mitotic figures. Epithelioid leiomyomas are yellow or grey and may contain visible areas of hemorrhage and necrosis, and tend to be solitary and softer than the usual leiomyoma. Myxoid leiomyomas have myxoid material separating the tumor cells. They are soft and translucent with circumscribed margins with neither cytologic atypia nor mitotic figures. Atypical leiomyomas lack all the other components with the exception of atypia and have little recurrence potential.14 See Figure 6.
|Standard smooth muscle Differentiation||Epithelioid differentiation||Myxoid differentiation|
|Histology||Cigar-shaped spindled cells with scanty to abundant eosinophilic cytoplasm||Rounded cells with central nuclei, and clear to eosinophilic cytoplasm||Spindle shaped cells set within an abundant myxoid matrix|
|Criteria for LMS||Any coagulative tumor cell necrosisIn the absence of tumor cell necrosis, the diagnosis required diffuse, moderate to severe cytological atypia and a mitotic index of > 10mf/10hpf*. If the mitotic index is < 10mg/10hpf, the chance of recurrence is low (less than 2-3%). In the absence of coagulative tumor cell necrosis and significant atypia, a high mitotic index is compatible with a benign clinical course. *mf/hpf = mitotic figures/high power fields||Any coagulative tumor cell necrosis In the absence of tumor cell necrosis, the diagnosis requires diffuse, moderate to severe cytological atypia and a mitotic index of >5mf/10hpf.||Any coagulative tumor cell necrosis In the absence of tumor cell necrosis, the diagnosis requires diffuse, moderate to severe cytological atypia and a mitotic index of >5mf/10hpf.|
Unlike smooth muscle tumors at other sites, uterine smooth muscle tumors are generally not graded. Rather, clinical behavior is defined by the designation to categories of ULMS, leiomyoma, or STUMP. The distinction is important since grading ULMS based on criteria at other body sites is misleading.
Patients with abnormal uterine bleeding or a suspicious uterine lesion should undergo endometrial sampling. Imaging studies and/or clinical findings are not specific for ULMS versus other uterine tumors. Ultrasound examination, magnetic resonance imaging (MRI), or computed tomography (CT) do not reliably distinguish between sarcoma, leiomyoma, endometrial cancer, lymphoma, intravenous leiomyomatosis, or adenomyosis.18
MRIs and Diagnosis
The utility of MRI for diagnosis is being addressed in case reports. Contrast resolution in soft tissues (better than ultrasonography) and lack of ionizing radiation show great promise as an imaging tool to evaluated LMS. The findings of atypical degeneration with irregular contours should bring LMS into the differential when evaluating leiomyomas (or other pelvic masses).19-21 One study looked at patients (including nine patients with pathologically proven LMS and three with STUMP) in order to study the magnetic resonance characteristics of non-benign uterine smooth muscle tumors. Additionally, they analyzed twelve cases of benign leiomyomas in which the gynecologists had suspected LMS. Size, location, signal intensity, and contrast enhancement of the tumors were studied on an individual basis. With some exceptions, the authors concluded that more than 50% of high signal on T2-weighted images and the presence of any small high-signal areas on T1-weighted images with un-enhanced pockets were considered MRI suggestive for STUMPS and LMS.22
Staging is based on surgical, not clinical findings. Extensive local growth is a hallmark of ULMS and spread of these tumors occur by local, lymphatic, and hematogenous routes (see Figure 2). Metastasis frequently involves the lung. If the diagnosis of ULMS is known preoperatively, chest imaging is necessary to evaluate for metastatic disease.
Surgical staging for ULMS is the same as for endometrial carcinoma (see Figure 7). The surgery includes peritoneal washings for cytology, extrafascial total abdominal hysterectomy, bilateral salpingo-oophorectomy, removal of enlarged lymph nodes, and biopsy or any suspicious areas. Some oncologists recommend omentectomy and pelvic and paraaortic lymph node sampling. See Figure 7.
Stage I Tumor confined to corpus uteri
IA Tumor limited to the endometrium
IB Tumor invades up to or less than 50% of the myometrium
IC Tumor invades more than 50% of the myometrium
Stage II Tumor invades cervix but does not extend beyond uterus
IIA Endocervical glandular involvement only
IIB Cervical stroma invasion
Stage III Local and/or regional spread
IIIA Tumor involves uterine serosa and/or adnexa (direct extension or metastasis)
IIIB Vaginal involvement (direct extension or metastasis)
IIIC Metastasis to the pelvic and/or para-aortic lymph nodes
IVA Tumor invades the bladder mucosa and/or bowel mucosa
IVB Distant metastasis(excluding metastasis to vagina, pelvic serosa, or adnexa. Including metastasis to intra-abdominal lymph nodes other than para-aortic, and/or inguinal lymph nodes)
LMS should be grouped with regard to the degree of differentiation as follows:
G1 5 percent or less of a nonsquamous or nonmorular solid growth
G2 6 percent to 50 percent of a nonsquamous or nonmorular solid growth
G3 More than 50% of a nonsquamous or nonmorular solid growth
The importance of lymph node dissection is controversial.23 Though the involvement of lymph nodes is of prognostic significance, lymphadenectomy has not been shown to be therapeutic. Outcomes have been comparable among similarly staged patients who did or did not undergo lymphadenectomy.23 Based on these results, most perform lymph node dissection only in patients with clinically suspicious nodes. Patients with ULMS confined to the uterus have a low risk of occult nodal disease (2.4%).24 When ULMS is diagnosed postoperatively, re-exploration for surgical staging is probably unnecessary since this risk of metastasis to lymph nodes and beyond is minimal.
Ovarian conservation may be an option for premenopausal women who wish to retain ovarian function. Two studies have suggested that retention of the ovaries may not adversely affect prognosis in women with Stage I ULMS.23 Informed consent as to the uncertainty of outcome with conservative surgery and close follow-up is clearly needed.
Treatment for Localized Disease
At a minimum, surgical treatment of a patient with a ULMS of the uterus should include a total hysterectomy and removal of the cervix.
The benefit of postoperative adjuvant radiotherapy (RT) in ULMS is unclear. The European Organization for Research and Treatment of Cancer (EORTC) randomly assigned 222 patients with stage I or II uterine sarcoma (including 103 patients with ULMS) to pelvic external beam radiation or observation. The preliminary report in 2003 suggested a lower rate of local recurrence in the irradiated group but no improvement in overall survival.25
Retrospective studies provide conflicting data. Most studies that group all uterine sarcomas together note better pelvic control with adjuvant RT. In GOG protocol 20, women with stage I or II uterine sarcoma were randomized to a trial of adjuvant doxorubicin with or without adjuvant RT. The majority of these patients had carcinosarcoma. Results showed that the irradiated group of patients had a significantly lower rate of pelvic failure, but no improvement in overall survival.26 Smaller retrospective studies suggest the possibility of a survival benefit as well as an improvement in local control.27-30 The largest series evaluated 103 women with stage I-IV uterine sarcoma (42% ULMS) who received RT at the discretion of their physician. Irradiated patients had a significantly better five-year pelvic control (76% versus 36%) and overall survival (73% versus 37%). The significance of improved survival and pelvic control remained in multivariate analysis after controlling for stage, histology, tumor grade, and presence of lymphvascular invasion.29
A major obstacle with ULMS is that even if pelvic control is achieved, the majority of women develop distant extraabdominal metastases.31
Guidelines from the National Comprehensive Cancer Network (NCCN) suggest that adjuvant RT can be considered for all women with resected stage I or stage II ULMS. For stage III ULMS with positive lymph nodes, the NCCN recommends consideration of adjuvant chemotherapy and pelvic RT, vaginal brachytherapy, and/or adjuvant chemotherapy.32
NCCN: The National Comprehensive Cancer Network
NCCN is an alliance of twenty cancer centers that work together to, among other things, develop treatment guidelines for most cancers. NCCN is also dedicated to research that improves the quality, effectiveness, and efficiency of cancer care. The NCCN Clinical Practice Guidelines in Oncology™ are viewed by many as the standard for clinical policy. See the article, Options and Follow-up Care for Women with Uterine Sarcomas, by Suzie Siegel which appeared in ESUN.
The use of RT needs to be balanced with the negative effects of therapy. Short term or immediate side effects include vaginal bleeding, vaginal discharge, skin reactions, hair loss, urinary problems, diarrhea and pain. Long term side effects include changes in bowel/bladder function and sexual function.
With the high rate of distant metastatic spread in ULMS, adjuvant systemic therapy is controversial. Some observational studies suggest a benefit,33, 34 while most do not.32 35-38 To date, no prospective studies are available that focus on patients with ULMS, and there is no definitive evidence that adjuvant chemotherapy improves overall survival. Therefore, it cannot be recommended as the standard of care, and should be considered in individual circumstances.
Three observational studies suggest that the combined use of postoperative RT and chemotherapy may provide benefit after resection of uterine sarcoma. Two of the studies involved patients with carcinosarcomas. The third study had 41 patients with uterine sarcoma who received either pelvic RT or RT plus adjuvant chemotherapy. Three-year survival rates were significantly better in the chemotherapy group (66% versus 36%).39
Neoadjuvant chemotherapy can be used to improve respectability of advanced disease, in the appropriate setting. The data is limited, at best.
Treatment for Recurrent, Advanced or Metastatic Disease
Recurrent ULMS is diagnosed by the new development of symptoms. Most relapses occur in the pelvis, followed by the lung and abdomen. Bone and brain metastases are uncommon.23 Surgical resection should be considered in patients with localized single foci recurrences, either local or metastatic. In a report of 41 women who underwent resection for recurrent uterine ULMS (29% pulmonary, 41% pelvis), two-year survival was 71 percent among those who had a disease-free interval between resection of the primary and the development of metastatic disease of 12 months or longer.40-42 In a study evaluating metastatic disease, Lenvenback et al showed that 71% had unilateral lesions, 51% had one lesion, and 70% had nodules greater than 2 cm. After pulmonary resection, unilateral versus bilateral disease was a significant predictor of survival (p = 0.02). Size, number of metastases, disease-free interval, and patient age were not significant.42
Radiofrequency Ablation (RFA) and Video Assisted Thoracic Surgery (VATS)
There are a few case reports in the literature of RFA and VATS for metastatic lesions to the liver and lung. Most of these studies have small numbers and, within those, even fewer with lesions from a uterine primary. Nevertheless, these approaches provide an alternative local therapy for metastatic lesions, and more studies will be needed to establish its role in LMS.43-47
Though unproven in the adjuvant setting, single agent doxorubicin is an effective drug for advanced ULMS. Objective response rates are between 16 and 25 percent, lasting generally less than 6 months.48-52
Two randomized trials have examined the benefit of doxorubicin single agent therapy versus in combination. Doxorubicin alone was compared to doxorubicin plus cyclophosphamide. Response rates were similar in both arms for patients with measurable disease (19%), as was the progression-free and median overall survival (median 11.6 versus 10.9 months).49
The second trial compared doxorubicin with and without dacarbazine. Although combined therapy was associated with a significantly higher response rate overall, there were no significant differences between the two groups in terms of progression-free survival or overall survival (7.7 versus 7.3 months). Combination therapy was associated with more hematologic and gastrointestinal toxicity.48
Ifosfamide has limited activity as a single agent with a response rate of 17%.53 Its combination with doxorubicin increased the objective response but added substantial toxicity.54,55
The combination of gemcitabine and docetaxel is the most effective chemotherapy regimen for ULMS patients with advanced disease described to date. In one report, patients with unresectable uterine or other primary site ULMS received gemcitabine plus docetaxel and granulocyte colony stimulating factor. Of the 34 patients in the study, complete response was seen in 3 patients and partial response in 15, for an overall response rate of 53%. Seven patients had stable disease. Despite the use of granulocyte colony stimulating factor, grade 3 or 4 neutropenia and febrile neutropenia developed in 21 and 6 percent, respectively. The toxicity profile was otherwise mild.53 A second series with 35 patients reported 7 of 12 patients (2 with ULMS) had a response.56
To date, there has not been a phase III trial comparing doxorubicin plus ifosfamide versus gemcitabine plus docetaxel. Historical comparison shows at least equivalent response, with improved toxicity with gemcitabine plus docetaxel. Therefore gemcitabine plus docetaxel can be considered for first line use in the appropriately selected patient.
Temozolomide is also modestly active. In an observational series with 12 patients (most of whom had received two prior chemotherapy regimens), one patient had a prolonged partial response and one a near complete response after 13 months.57 In a second study, responses were seen in 5 of 11 patients with gynecologic ULMS.58 Response lasted for longer than one year in four patients.
A potentially new agent being investigated in the treatment of soft tissue sarcomas is Trabectedin (ecteinascidin), or ET-743. It is the active component of extracts from Caribbean tunicate, Ecteinascidia tubinata. ET-743 binds to the guanine residue within the minor groove of DNA causing a bend in the major groove which interferes with the DNA binding proteins and transcription factors in the cancer cell. Several phase II studies have demonstrated some activity in advanced soft tissue sarcomas, including ULMS. Response rates range between 4-17%.59 This drug and its potential future combination with additional active agents will be investigated in a future GOG phase II trials.
Certain uterine tumors are responsive to hormonal therapy because they express estrogen and/or progesterone receptors. However, this is not the case in ULMS and adjuvant hormonal therapy is not recommended for any stage of ULMS.
Patients with metastatic ULMS have limited options with regards to chemotherapy and enrollment in clinical trials is appropriate. Chemotherapy is palliative and should be used to relieve symptoms. Options include single agent doxorubicin, doxorubicin and ifosfamide, single agent gemcitabine, and gemcitabine and docetaxel. Considering that there is no survival benefit with our current chemotherapeutic options, toxicity versus symptom management should be evaluated on a case by case basis with full informed consent.
Summary and Recommendations
ULMS are rare tumors with a limited body of literature to help guide treatment. Patient care should be individualized. Further investigation is needed to improve the treatment options for our patients with this disease.
- Physical exam every 3 months for 2 years, then every 6-12 months.
- Chest imaging every 3-6 months for 2 years, then annually.
- CT/MRI as clinically indicated.
- Patient education regarding symptoms.
- Extrafascial total abdominal hysterectomy with bilateral salpingo-opherectomy and formal surgical staging.
- Fertility-sparing in young women who wish to preserve childbearing potential with low grade ULMS may be considered.
- RT appears to improve local control while it is unclear whether it provides survival benefits. Adjuvant chemotherapy is of uncertain benefit.
Questions, Comments & Counterpoints
Two letters about this article
We have received two letters regarding this article. The first letter was received from Dr. Ian Judson of Royal Marsden Hospital, London, and reads:
To the authors: This is an excellent and comprehensive review of a highly aggressive sarcoma and it is important that awareness be drawn to this disease entity. However, I would caution against the blanket statement that ULMS is unresponsive to oestrogen. I know of cases of ULMS that express oestrogen and progesterone receptors and have responded to oestrogen withdrawal measures. It has been assumed that this phenomenon is restricted to low grade endometrial stromal sarcoma. This appears not to be the case and there are grounds for believing that pathologists should examine all ULMS cases for ER and PgR expression.
The second letter was received from Shirley Collings, a leiomyosarcoma survivor who lives in the UK. It reads:
Dear Bruce and ULMS survivors,
First of all may I refresh your memories about the article in the ESUN Apr 07 on Uterine Leiomyosarcomas. In particular the three following extracts: "Benign leiomyomas (fibroids) and ULMS often coexist in the same uterus, but are genetically distinct entities. ULMS are much less common and not hormonally driven. "
"Fibroids rarely, if ever, degenerate into ULMS12"
"Certain uterine tumors are responsive to hormonal therapy because they express estrogen and/or progesterone receptors. However, this is not the case in ULMS and adjuvant hormonal therapy is not recommended for any stage of ULMS."
I am sure I speak for a number of women on this and the Sarcoma UK lists, who disagree vehemently with these three statements. Through personal experience with diagnoses from fibroids, ER/PR positivity and subsequent treatment with aromatase inhibitors many of us can prove these statements wrong. I am shocked and disappointed that four such eminent doctors can attach their names publicly to these misleading statements in an otherwise very interesting article.
Bruce, would you kindly contact Drs Gosh, Hecht, Ferzandi and Awtry and ask them if they would be willing to respond to and comment on this email in the next edition of ESUN.
The Authors' Reply
We thank the Dr. Ian Judson for his thoughts and Mrs. Collings for her passionate comments and offer further data from the literature to clarify the issues raised.
The distinction between benign leiomyomata (fibroids) and smooth muscle sarcomas in the uterus is clear.1 The genetic changes in sarcomas are more complex than in fibroids [many references over 20 years, available on request]. Interestingly, the specific genetic aberration detected may vary from cell to cell suggesting a high level of genomic instability. This instability can also be detected by techniques that measure allelic imbalance such as comparative genomic hybridization, and is not present in benign fibroids.2 Some changes are unique to sarcomas such as a heterozygosity for the long arms of chromosomes 10 and 13 that is found in more than half of sarcomas.
Although hormone manipulation with agents such as leuprolide (a gonadotropin-releasing hormone –GnRH agonist) or aromatase inhibitors (an agent that reduces circulating estrogen) have a clear role in treatment of benign leiomyoma, the response of sarcomas is highly variable, and is not part of the primary therapy offered to women with sarcoma [treatment recommendations for sarcoma based on stage are available at the NCI website].
We could not find publications describing even anecdotal experiences with anti-estrogens for treatment of leiomyosarcoma. We suspect that such options are used when conventional therapy fails. There is no evidence that hormone receptor status correlates with response and those stains are not routinely performed. Rare reponses to antiprogestins has been reported. [Koivisto-Korander R, Leminen A, Heikinheimo O. Mifepristone as treatment of recurrent progesterone receptor-positive uterine leiomyosarcoma. Obstet Gynecol. 2007;109:512-4.]
Many cases of leiomyosarcoma express hormone receptors and the association between hormone receptor expression and sarcoma growth has been examined. The largest study has com from Memorial Sloan-Kettering Cancer Center. They show that the expression of estrogen and progestin receptor was less frequent in uterine sarcoma compared with leiomyoma. Progesterone and androgen receptor expression appeared to be associated with disease-free survival but were not found to correlate with overall survival. [Leitao MM, Soslow RA, Nonaka D, Olshen AB, Aghajanian C, Sabbatini P, Dupont J, Hensley M, Sonoda Y, Barakat RR, Anderson S. Tissue microarray immunohistochemical expression of estrogen, progesterone, and androgen receptors in uterine leiomyomata and leiomyosarcoma. Cancer. 2004 Sep 15;101(6):1455-62.]
Another category of uterine sarcoma, uterine stromal sarcoma, may have some overlap in microscopic appearance and be confused with leiomyosarcoma. In addition, rare sarcomas have mixed differentiation (smooth muscle and stromal). Stromal sarcomas are more likely to express progesterone receptors and progesterone therapy may have some role in treating recurrent stromal sarcomas. [Katz L, Merino MJ, Sakamoto H, et al.: Endometrial stromal sarcoma: a clinicopathologic study of 11 cases with determination of estrogen and progestin receptor levels in three tumors. Gynecol Oncol 26 (1): 87-97, 1987.]
Dr. Judson's Response
Thank you for giving me the opportunity to respond to the comments both of Mrs. Collings and the authors of the review on uterine sarcomas.
Over the last 20 years there has been a gradual change in the approach to the hormonal manipulation of gynecological sarcomas. Progestogens have been used for more than 25 years, to my knowledge, in the management of the rare condition of metastasising leiomyoma, which is really low grade leiomyosarcoma. I first met a patient with the disease who had responded to medroxyprogesterone acetate (MPA) in 1983. She required pulmonary metastasectomy in the 1990s for progressive disease that was no longer responding but remains alive and well. I have other patients with the same condition whose disease fluctuates but remains essentially stable. We recently saw a case of grade I leiomyosarcoma of the vulva that had grown dramatically during pregnancy. Such lesions are acknowledged to be hormonally driven. We know that leiomyosarcomas represent a broad spectrum of malignancy from STUMP to high grade disease. It is assumed that hormonal dependence is lost with increasing grade. While there is no literature on the treatment of frankly malignant leiomyosarcoma with estrogen deprivation therapy, clearly in some cases it can be effective. What is particularly important is that doctors and patients alike should be aware of the potentially detrimental effect of hormone replacement therapy (HRT), suggesting that it might be helpful to know the ER/PgR status of all such tumours.
As acknowledged in the article, the position regarding endometrial stromal sarcoma (ESS) is clearer, although the literature is still sparse. Progestogens used to be the mainstay of hormonal therapy for these tumours. Since the advent of potent aromatase inhibitors it is much more logical, as well as highly effective, to use these agents that are better tolerated and provide a reliable way of reducing oestrogen levels in the menopausal / post-oophorectomy state. A brief review of our own experience of treating metastatic endometrial stromal sarcoma reveals 4 patients with response / prolonged stable disease on MPA, ranging from 4-10 years, 3 patients responding to withdrawal of HRT or oophorectomy lasting 18+months, 3 and 6 years respectively and 6 patients responding to an aromatase inhibitor (usually letrozole) of 1-5 years in duration. Recurrent well-differentiated ESS that expresses ER and PgR can be expected to respond to withdrawal of HRT (which should not be given in the first place), oophorectomy if this has not been performed, or administration of an aromatase inhibitor with a high degree of confidence. Tamoxifen should not be used since it is pro-estrogenic in the uterus and I have seen exacerbation of the disease with this agent.
It would appear that further research is required to address the true response rate of ESS to aromatase inhibitors and to examine their potential role in the treatment of a proportion of less agressive uterine leiomyosarcomas.
The Authors' Closing Reply
We appreciate the depth and insight of Dr. Judson's comments. As Dr Judson mentioned, the literature on this issue is sparse. In this review, we have attempted to outline guidelines for evidence-based standard of care for uterine LMS. We do recognize that in individual cases and case-series, there has been reponse to hormonal therapy, dependent on grade. Although in rare circumstances that the therapy may be indicated, until sufficient data exists, we were careful not to project this as routine treatment. However, the physician has a duty to serve his/her patients with the data available and that the patient must be given full informed consent thereof.