Introduction

Uterine sarcomas account for about 3–7% of uterine cancers. Because of their rarity and heterogeneity, no consensus has yet emerged on prognostic factors for clinical outcome and treatment strategy. Their histopathologic classification was revised by the World Health Organization in 2003. Application of the more restrictive diagnostic criteria allows for the exclusion of smooth-muscle tumours of uncertain malignant potential, as well as several leiomyoma variants (i.e. cellular, mitotically active, and atypical leiomyomas) often previously misdiagnosed as leiomyosarcomas. Carcinosarcoma is considered as a metaplastic form of endometrial carcinoma. Endometrial stromal sarcomas have been long subdivided into low-grade and high-grade endometrial stromal sarcomas. Some investigators, however, have recommended a change in the classification of high-grade endometrial stromal sarcomas into undifferentiated or poorly differentiated endometrial sarcomas, as these tumours have distinct pathologic and clinical properties compared with low-grade endometrial stromal sarcomas.

Uterine sarcomas have long staged according to a modification of the 1988 International Federation of Gynecology and Obstetrics staging system for endometrial carcinoma. Recently, a new International Federation of Gynecology and Obstetrics classification and staging system has been specifically designed for uterine sarcomas in an attempt to reflect their different biological behaviours. Three classifications have been developed: staging for leiomyosarcomas and endometrial stromal sarcomas; staging for adenosarcomas; and staging for carcinosarcomas. Stage I leiomyosarcomas and endometrial stromal sarcomas are subdivided according to tumour size, whereas subdivision of stage I adenosarcomas takes into account myometrial invasion, and carcinosarcomas are staged as endometrial carcinomas.

Tumour stage is the strongest prognostic factor for all uterine sarcomas, whereas the prognostic value of histological type has been long debated. Most investigators have failed to detect significant differences in survival according to this variable, excluding low-grade endometrial stromal sarcomas, which show an excellent clinical outcome. Conversely, multivariate analysis of results from other studies has revealed that the survival of women with leiomyosarcoma, when adjusted for stage, age, grade, and mitotic count, is poorer than that of women with carcinosarcoma. For instance, in an Italian multicentre retrospective study of 249 women, the risk of death was significantly lower for low-grade endometrial stromal sarcoma (RR 0.257; 95% CI 0.071 to 0.931) and carcinosarcoma (RR 0.509; 95% CI 0.324 to 0.799) compared with leiomyosarcoma, whereas no survival difference was found between leiomyosarcoma and high-grade endometrial stromal sarcoma.

In this chapter, we aim to assess the prognostic relevance of different clinico-pathological and biological variables in uterine sarcomas. Most data have been collected from retrospective studies, including limited numbers of cases, and only a few investigators have analysed larger groups of women treated in different institutions and with different treatment modalities. Therefore, the quality of the information available is relatively poor.

Leiomyosarcoma

The main histological criteria for the diagnosis of uterine leiomyosarcoma are the presence of cytologic atypia, high mitotic count, and coagulative tumour cell necrosis. Moreover, leiomyosarcoma often shows irregular, infiltrating borders as well as vascular invasion. Bell et al. advocated the diagnosis of leiomyosarcoma for any smooth-muscle tumour showing coagulative tumour cell necrosis in the presence of cytological atypia and any mitotic activity as well as for those tumours with coagulative tumour cell necrosis and 10 mitoses/10 high power field (HPF) or more even in the absence of cytologic atypia. None of these histological findings, however, is valid as single criterion, and the diagnosis is based on the cumulative assessment of all these features. For instance, mixoid leiomyosarcoma is characterised by low cellularity caused by an abundant mixoid material, low mitotic count, and presence of mild or moderate atypia. In epithelioid leiomyosarcoma, necrosis may be absent. In the absence of severe cytologic atypia and high mitotic count, both tumours are diagnosed as sarcomas based on their infiltrative borders.

Disease stage is the most important prognostic variable in leiomyosarcoma ( Table 1 ). For instance Kapp et al., who reassessed 1396 cases of uterine leiomyosarcoma identified in the Surveillance, Epidemiology, and End Results (SEER) database from 1988 to 2003, found that stage was an independent prognostic factor for disease-specific survival ( P < 0.001).

Reported survival rates are about 50% for stage I–II disease, whereas dismal results have been reported in advanced-stage disease.

A review of the literature on lymph-node metastases in uterine leiomyosarcoma showed a low incidence of occult retroperitoneal involvement (about 4–11%) in women without macroscopic spread of the disease and a limited prognostic relevance for lymph-nodal status. These findings are in agreement with the low risk of lymph-node involvement (2.7–4%) in patients with leiomyosarcomas of all soft-tissue sites.

In a retrospective review on 208 women with uterine leiomyosarcoma, Giuntoli et al. found positive nodes in only four out of 36 (11%) women who underwent lymphadenectomy, and three of these had concomitant extra-uterine disease. No statistical difference in disease-specific survival emerged between women diagnosed as stage I with histologically negative lymph nodes and women diagnosed as stage I whose lymph nodes were assumed to be uninvolved. In a study of 63 women, Ayhan and Aksan reported that median disease-free survival and overall survival were similar in women who had undergone lymphadenectomy and in those who had not (2.51 v 2.36 years; P = not significant; and 2.44 v 3.16 years; P = not significant, respectively). In the study by Kapp et al., 5-year disease-specific survival was 26% for women who had positive lymph nodes compared with 64% for those who had negative nodes ( P < 0.001, but lymph-node status failed to achieve a statistical significance on multivariate analysis; HR for lymph-node metastases 1.168, 95% CI, 0.947 to 1.440).

Advancing age has been associated with unfavourable clinical outcome in most studies. Patient age was related to survival on univariate analysis ( P = 0.001) but not on multivariate analysis in the study of D’Angelo et al., whereas some investigators have reported that older age at diagnosis is an independent poor prognostic variable. In an Italian retrospective study, women aged 50 years or older had 2.073-fold increased risk of progression ( P = 0.0048), and in the study by Wu et al., women aged over 50 years had 11.07 increased risk of death ( P = 0.017) compared with younger women. Conversely, other investigators have failed to detect a prognostic relevance for patient age.

Mitotic count has been found to affect clinical outcome in most studies ( Table 2 ). In the Gynecologic Oncology Group study on prognostic factors on early stage uterine sarcomas, mitotic count was the only variable significantly related to progression-free interval in patients with leiomyosarcoma. ( P = 0.03). Pelmus et al. observed that mitotic activity was associated with metastasis-free interval ( P = 0.03) in a study of 108 women with stage I–II disease. Some investigators, however, have failed to detect an effect of mitotic count on survival. The reason for this discrepancy may be differences in handling of the surgical specimen, thickness of sections, size of the HPF, and the interpretation of mitotic figures.

Tumour size has been reported to be of prognostic value, especially in stage I leiomyosarcoma. Abeler et al. reported that, among women with disease confined to the uterus, tumour size was related to crude 5-year survival on both univariate analysis (<50 mm: 64.0% v 50–100 mm; 56.4% v >100 mm; 29.3%, P < 0.0001) and multivariate analysis (with 100 mm as cut-off, RR of death 2.7; P < 0.01). When combining tumour size and mitotic count (with 10 mitoses/10 HPF as cut-off), women could be stratified into three risk groups: a low-risk group with size 100 mm or less and mitotic count 10 or less; a medium-risk group with either size greater than 100 mm or mitotic count greater than 10 with a 1.9-fold increased risk of death; and a high-risk group with size greater than 100 mm and mitotic count greater than 10 with a 5.3-fold increased risk of death. In the study by Wu et al. that included 51 women with stage I–IV leiomyosarcoma, tumour size was an independent predictor of death (>11 cm v ≤11 cm: HR: 11.63, 95% CI 2.14 to 63.12; P = 0.004).

High tumor grade has been associated with worse survival by some investigators but not by others. Kapp et al. found that high tumour grade was an independent predictive variable for poor disease-specific survival (HR 1.831, 95% CI 1.432 to 2.340; P < 0.001), and Blom et al. reported that tumour grade was the only independent prognostic factor for survival in stage I disease. Conversely, histological grade did not correlate with progression-free survival in the Gynecologic Oncology Group study.

Mayerhofer et al. found that vascular space involvement was an independent poor prognostic factor for overall survival among women with stage I–IV leiomyosarcoma (RR 4.12, 95% CI 1.61 to 10.54; P ≤ 0.0003), and Pelmus et al. observed that this finding was associated with a shorter metastasis-free interval in stage I–II disease ( P = 0.04). Conversely, vascular space involvement was not related to the clinical outcome in other studies.

The prognostic relevance of vascular endothelial-growth factor (VEGF) and its kinase receptors VEGFR-1 (flt-1) and VEGF R- (flk-2) in uterine sarcoma has not yet defined. Arita et al. reported that VEGF and its kinase receptors stained significantly stronger in 39 uterine sarcomas than in normal uterine smooth muscle, with no significant differences among the different histological types. No significant survival difference was detected according to VEGF immunoreactivity, whereas high intensity of immunostaining for phosho-flt-1 was associated with a poorer prognosis.

Oestrogen receptor and progesterone receptor positivity ranged from 26–87% and from 17–80%, respectively, in uterine leiomyosarcomas. In particular, the assessment of progesterone receptor status seems to have diagnostic and prognostic relevance. Mittal and Demopoulos found immunohistochemical expression of progesterone receptor in 17% of 12 leiomyosarcomas, 100% of seven smooth-muscle tumours of unknown malignant potential, and 93% of 15 cellular leiomyomas. Progesterone receptor positivity seems to be associated with a lower risk of recurrence and a better survival in women with leiomyosarcoma. In a recent retrospective investigation, Watanabe and Suzuki reported that all seven tumours with excellent clinical outcome expressed progesterone receptor, often intensely, and also frequently expressed oestrogen receptor, whereas all seven tumors with poor prognosis lacked them completely. Conversely other investigators failed to detect any significant correlations between steroid receptor status and survival.

Some investigators showed that DNA ploidy and S-phase fraction (SPF) influenced clinical outcome Blom et al. found that diploidy and a SPF lower than 10% were independent good prognostic variables for survival ( P = 0.045 and P = 0.041, respectively). Nordal et al., who reviewed 58 women, observed that DNA ploidy was significantly related to prognosis on univariate analysis ( P = 0.043) but not on multivariate analysis. Conversely in the study of Wu et al., DNA ploidy correlated with neither the risk of recurrence nor the risk of death on univariate analysis.

Uterine leiomyosarcomas have significantly higher Ki-67 index and p53 expression than benign smooth muscle tumors.

In the study of Mayerhofer et al., median disease-free survival was 12.6 months for women with Ki-67 positive tumours compared with 52.6 months for women with Ki-67 negative tumours ( P = 0.01). In the subgroup of women diagnosed with stage I, the difference in disease-free survival between Ki-67 positive and Ki-67 negative cases remained statistically significant ( P = 0.02). Similar results were observed by Aykan et al. Conversely, in the study by Zhai et al., Ki-67 positivity was not related to overall survival, whereas positive staining for cyclin A ( P = 0.008) and diffuse staining for cyclin E ( P = 0.038) were associated with shorter survival.

Most investigators reported that p53 expression had a prognostic relevance. For instance Blom et al. reported that p53 status was the only independent prognostic variable for disease-free survival among stage I-II patients ( P = 0.001), and that all p53 positive cases recurred within 28 months. Conversely other authors found no relationship between p53 status and patient survival.

p16 overexpression has been demonstrated in 57–100% of uterine leiomyosarcomas compared with about 13% or less of leiomyomas. The assessment of p16, in combination with p53 and Ki-67, may be of value as an adjunct to morphological examination in the differential diagnosis of problematic uterine smooth muscle tumors. D’Angelo et al. observed that immunoreactions for Ki-67, p53, and p16 had similar staining diffuse distribution and strong intensity in leiomyosarcomas and undifferentiated endometrial sarcomas.

Twist is an oncogene that inhibits apoptosis and appears indistinguishable from bcl-2. Leisler et al. found that positive immunostaining for bcl-2 was significantly associated with longer time to recurrence ( P = 0.02) in uterine leiomyosarcoma. In the study by D’ Angelo, Twist expression showed a trend to an inverse relationship with overall survival on univariate analysis ( P = 0.07), p16 and Ki-67 expression had a negative effect on disease-free survival ( P = 0.007 and P = 0.021, respectively), and strong bcl-2 immunostaining was associated with longer overall disease-free survival on multivariate analyses ( P = 0.001). These authors reported that all the three women, without evidence of recurrence on long-term follow up, had tumours that were negative/low expressors of Ki-67, p53, p16, and Twist, with strong bcl-2 expression. Therefore, it seems to be possible to identify a subset of leiomyosarcomas with a favorable prognosis based on a panel of immunomarkers for cell proliferation and apoptosis.

Several investigators reported that ovarian preservation did not adversely affect the clinical outcome of women with leiomyosarcoma. For instance, in a case- control study, Giuntoli et al. found no significant difference in recurrence-free survival and disease-specific survival between 25 women with ovarian preservation and 25 women matched by stage, grade who underwent bilateral salpingo-oophorectomy.

Leiomyosarcoma

The main histological criteria for the diagnosis of uterine leiomyosarcoma are the presence of cytologic atypia, high mitotic count, and coagulative tumour cell necrosis. Moreover, leiomyosarcoma often shows irregular, infiltrating borders as well as vascular invasion. Bell et al. advocated the diagnosis of leiomyosarcoma for any smooth-muscle tumour showing coagulative tumour cell necrosis in the presence of cytological atypia and any mitotic activity as well as for those tumours with coagulative tumour cell necrosis and 10 mitoses/10 high power field (HPF) or more even in the absence of cytologic atypia. None of these histological findings, however, is valid as single criterion, and the diagnosis is based on the cumulative assessment of all these features. For instance, mixoid leiomyosarcoma is characterised by low cellularity caused by an abundant mixoid material, low mitotic count, and presence of mild or moderate atypia. In epithelioid leiomyosarcoma, necrosis may be absent. In the absence of severe cytologic atypia and high mitotic count, both tumours are diagnosed as sarcomas based on their infiltrative borders.

Disease stage is the most important prognostic variable in leiomyosarcoma ( Table 1 ). For instance Kapp et al., who reassessed 1396 cases of uterine leiomyosarcoma identified in the Surveillance, Epidemiology, and End Results (SEER) database from 1988 to 2003, found that stage was an independent prognostic factor for disease-specific survival ( P < 0.001).

Reported survival rates are about 50% for stage I–II disease, whereas dismal results have been reported in advanced-stage disease.

A review of the literature on lymph-node metastases in uterine leiomyosarcoma showed a low incidence of occult retroperitoneal involvement (about 4–11%) in women without macroscopic spread of the disease and a limited prognostic relevance for lymph-nodal status. These findings are in agreement with the low risk of lymph-node involvement (2.7–4%) in patients with leiomyosarcomas of all soft-tissue sites.

In a retrospective review on 208 women with uterine leiomyosarcoma, Giuntoli et al. found positive nodes in only four out of 36 (11%) women who underwent lymphadenectomy, and three of these had concomitant extra-uterine disease. No statistical difference in disease-specific survival emerged between women diagnosed as stage I with histologically negative lymph nodes and women diagnosed as stage I whose lymph nodes were assumed to be uninvolved. In a study of 63 women, Ayhan and Aksan reported that median disease-free survival and overall survival were similar in women who had undergone lymphadenectomy and in those who had not (2.51 v 2.36 years; P = not significant; and 2.44 v 3.16 years; P = not significant, respectively). In the study by Kapp et al., 5-year disease-specific survival was 26% for women who had positive lymph nodes compared with 64% for those who had negative nodes ( P < 0.001, but lymph-node status failed to achieve a statistical significance on multivariate analysis; HR for lymph-node metastases 1.168, 95% CI, 0.947 to 1.440).

Advancing age has been associated with unfavourable clinical outcome in most studies. Patient age was related to survival on univariate analysis ( P = 0.001) but not on multivariate analysis in the study of D’Angelo et al., whereas some investigators have reported that older age at diagnosis is an independent poor prognostic variable. In an Italian retrospective study, women aged 50 years or older had 2.073-fold increased risk of progression ( P = 0.0048), and in the study by Wu et al., women aged over 50 years had 11.07 increased risk of death ( P = 0.017) compared with younger women. Conversely, other investigators have failed to detect a prognostic relevance for patient age.

Mitotic count has been found to affect clinical outcome in most studies ( Table 2 ). In the Gynecologic Oncology Group study on prognostic factors on early stage uterine sarcomas, mitotic count was the only variable significantly related to progression-free interval in patients with leiomyosarcoma. ( P = 0.03). Pelmus et al. observed that mitotic activity was associated with metastasis-free interval ( P = 0.03) in a study of 108 women with stage I–II disease. Some investigators, however, have failed to detect an effect of mitotic count on survival. The reason for this discrepancy may be differences in handling of the surgical specimen, thickness of sections, size of the HPF, and the interpretation of mitotic figures.

Table 2

Only gold members can continue reading. Log In or Register to continue

Share this:

• X
• Facebook

Related

Related posts:

1. Staging of uterine sarcomas
2. Role of radiotherapy treatment of uterine sarcoma
3. Corrigendum to Uterine Prolapse In Young Women YBEOG 25/2 (2011) 157–165
4. Colposcopy and cervical pathology: Answers to multiple choice questions for Vol. 25, No. 5

Stay updated, free articles. Join our Telegram channel

Join