Demographic and Treatment Variables Influencing Outcome for Localized Paratesticular Rhabdomyosarcoma: Results From a Pooled Analysis of North American and European Cooperative Groups

Abstract

Purpose Treatment recommendations for localized paratesticular rhabdomyosarcoma (PT RMS) differ in North America and Europe. We conducted a pooled analysis to identify demographic features and treatment choices that affect outcome. Patients and Methods We retrospectively analyzed the effect of nine demographic variables and four treatment choices on event-free survival (EFS) and overall survival (OS) from 12 studies conducted by five cooperative groups. Results Eight hundred forty-two patients with localized PT RMS who enrolled from 1988 to 2013 were included. Patients age ≥ 10 years were more likely than younger patients to have tumors that were > 5 cm, enlarged nodes (N1), or pathologically involved nodes ( P ≤ .05 each). With a median follow-up of 7.5 years, Kaplan-Meier estimates for 5-year EFS and OS were 87.7% and 94.8%, respectively. Of demographic variables, cooperative group, era of enrollment, age category, tumor size, Intergroup Rhabdomyosarcoma Study group, and T stage affected EFS ( P ≤ .05 each). Surgical assessment of regional nodes, which was performed in 23.5% of patients—usually in those age ≥ 10 years or with suspicious or N1 nodes—was the only treatment variable associated with EFS by univariable and multivariable analyses ( P ≤ .05 each) in patients age ≥ 1 year. A variable selection procedure on a proportional hazards regression model selected era of enrollment, age, tumor size, and surgical assessment of regional nodes as significant ( P ≤ .05 each) in the EFS model, and era of enrollment, age, tumor size, and histology ( P ≤ .05 each) in the OS model. Conclusion Localized PT RMS has a favorable prognosis. Age ≥ 10 years at diagnosis and tumor size larger than 5 cm are unfavorable prognostic features. Surgical assessment of regional nodes is important in patients age ≥ 10 years and in those with N1 nodes as it affects EFS.