Clinical outcomes of women with recurrent or persistent uterine leiomyosarcoma.

OBJECTIVES: This study aimed to identify prognostic factors influencing the outcome of recurrent or persistent uterine leiomyosarcoma (ULMS). METHODS: All patients with recurrent or persistent ULMS who underwent treatment at the participating institutions between January 2000 and December 2010 were identified from the tumor registry. The Kaplan-Meier method was used to generate overall survival data. Factors predictive of outcome were compared using the log-rank test and Cox proportional hazards model. RESULTS: One hundred fifteen (68.8%) patients who had recurrent/persistent disease were identified, 40 (34.8%) had persistent disease, and 75 (65.2%) had a recurrence. Median follow-up time was 24.9 months. The 5-year postrelapse survival rate was 15% and was not significantly different between women with recurrent or persistent disease (16% vs 13%; P = 0.1). Variables identified affecting the 5-year postrelapse survival rate included low number of mitosis at the time of diagnosis (<25, 25% vs 5%; P = 0.002), time to relapse from original diagnosis (≤6 vs >6 months, 8% vs 22%; P = 0.003)), and surgical treatment (17% vs 12%; P = 0.01). Age, stage, chemotherapy at time of original diagnosis or at the time of relapse, site of recurrence, and single versus multiple sites of recurrence were not associated with survival. In a multivariate Cox regression model, only low number of mitosis (hazard ratio, 0.5; 95% confidence interval, 0.3-0.8, P = 0.02) was identified as a predictor of overall survival. CONCLUSIONS: The prognosis of patients with recurrent/persistent ULMS is, in general, poor. Women who have low number of mitosis at the time of diagnosis seemed to have better postrelapse survival.

Uterine leiomyosarcoma: an updated series.

OBJECTIVE: The aim of this study was to analyze and compare the clinicopathologic characteristics, treatment, and survival in patients with uterine leiomyosarcoma (ULMS) during the last 10 years in 3 referral academic centers. METHODS: All patients with ULMS who underwent treatment at the participating institutions between January 1, 2000, and December 31, 2010, were identified from the tumor registry database. In each case, the diagnosis was confirmed by a dedicated gynecologic pathologist following postsurgery pathology review. The Kaplan-Meier method was used to generate overall survival (OS) data. Factors predictive of outcome were compared using the log-rank test and Cox regression analysis. RESULTS: Analysis of 167 women with ULMS with adequate follow-up was performed. One hundred twenty-eight patients (77%) were initially managed at the participating institutions, and 39 (23%) were referred after initial management at a different institution. Ninety-two (55%) had stage I disease, 7 (4%) had stage II, 18 (11%) stage III, and 50 (30%) had stage IV disease. The median OS for women with stage I was 75 months, for stage II 66 months, stage III 34 months, and stage IV 20 months (P < 0.001). For patients with early stage (I and II), race, lower grade, smaller tumor size (<11 cm), low number of mitosis (<25/10 high-power field [HPF]), lymphovascular space invasion, and presence of necrosis were identified as variables with prognostic influence on survival in the univariate analysis. A Cox proportional hazards model identified size 11 cm or greater (hazard ratio, 5.9; P < 0.001) and mitotic count of 25/10 HPF or greater (hazard ratio, 2.3; P = 0.05) as independent predictors of OS. For patients with late stage (stage III and IV), race, stage III versus IV, lower grade, smaller tumor size (<11 cm), and low number of mitosis (<25/10 HPF) were all associated with significantly improved OS. A Cox proportional hazards model identified mitotic count of 25/10 HPF or greater (P = 0.01) as independent predictor of OS. CONCLUSIONS: In early stage, size of the tumor and number of mitosis were associated to survival. In contrast to late stage, only mitotic count was associated to survival.