Prognostic factors for children and adolescents with surgically resected nonrhabdomyosarcoma soft tissue sarcoma: An analysis of 121 patients treated at St Jude Children's Research Hospital

Purpose: The rarity and heterogeneity of pediatric nonrhabdomyosarcoma soft tissue sarcoma (NRSTS) has precluded meaningful analysis of prognostic factors associated with surgically resected disease. To define a population of patients at high risk of treatment failure who might benefit from adjuvant therapies, we evaluated the relationship between various clinicopathologic factors and clinical outcome of children and adolescents with resected NRSTS over a 27-year period at our institution. Patients and Methods: We analyzed the records of 121 consecutive patients with NRSTS who underwent surgical resection between August 1969 and December 1996, Demographic data, tumor characteristics, treatment and outcomes were recorded. Univariate and multivariate analyses of prognostic factors for survival, event-free survival (EFS), and local and distant recurrence were performed. Results: At a median follow-up of 9.2 years, 5-year survival and EFS rates for the entire cohort were 89% +/- 3% and 77% +/- 4%, respectively. In univariate models, positive surgical margins (P = .004), tumor size greater than or equal to 5 cm (P < .001), invasiveness (P = .002), high grade (P = .028), and intra-abdominal primary tumor site P = .055) adversely affected EFS. All of these factors except invasiveness remained prognostic of EFS and survival in multivariate models. positive surgical margins (P = .003), intra-abdominal primary tumor site (P = .028), and the omission of radiation therapy (P = .043) predicted local recurrence, whereas tumor size greater than or equal to 5 cm (P < .001), invasiveness (P < .001), and high grade (P = .004) predicted distant recurrence. Conclusion: In this largest single-institution analy sis of pediatric patients with surgically resected NRSTS, we identified clinicopathologic features predictive of poor outcome. These variables should be prospectively evaluated as risk-adapted therapies are developed, J Clin Oncol 17:3697-3705. (C) 1999 by American Society of Clinical Oncology.