Prognostic factors and treatment modalities in uterine sarcoma

The aim of this study was to identify the impact of various prognostic factors in the management of uterine sarcoma. Fifty-nine patients with uterine sarcoma were treated at King Faisal Specialist Hospital and Research Center between 1980 and 1997. Forty-three patients (73%) were treated by total abdominal hysterectomy and bilateral salpingo-oophorectomy, 7 (12%) total abdominal hysterectomy and bilateral salpingo-oophorectomy with sampling of pelvic lymph nodes, and 9 (15%) had biopsy only. Nine patients received adjuvant treatment; five had radiation therapy (XRT), two had chemotherapy, one had combined XRT and chemotherapy, and one received hormonal treatment. Leiomyosarcoma cases accounted for 42% of all the uterine sarcomas, carcinosarcoma cases for 34%, and endometrial stromal sarcoma (ESS) for 24%. Fifty (85%) patients had pathologic grade II and III tumor, with only 9 patients grade I. Twenty-seven patients (46%) were classified surgically as stage 1, 7 (12%) as stage H, 17 (29%) as stage III, and 8 (13%) had stage IV tumor. Recurrences developed in 34 patients (71%). The 5- and 10-year overall actuarial survival for all patients was 42%, and the corresponding relapse-free survivals for those who achieved complete response after primary treatment (48 patients) were 27% and 20%. On the univariate analysis, grade I tumors (p = 0.04), ESS (p = 0.02), nonmetastatic stage (p = 0.05), and negative peritoneal cytology (p = 0.04) were associated with better overall survival. Factors associated with better relapse-free survival were premenopausal status (p = 0.048), grade I tumors (p = 0.006) and ESS (p = 0.04). On multivariate analysis, grade I tumors were associated with better overall survival (p = 0.04) and relapse-free survival (p = 0.02), whereas ESS independently influenced relapse-free survival only (p = 0.03). The outlook with uterine sarcomas remains somewhat dismal. Grade I tumors and endometrial stromal sarcoma were associated with good prognosis.