Y. Shi1, W. Zhang4, E. Beierle5, J. Doski6, A. Goldin7, M. Goldfarb8, K. Gow7, M. Langer9, J. Nuchtern1,3, S. A. Vasudevan1,3 1Baylor College Of Medicine,Department Of Surgery,Houston, TX, USA 3Texas Children’s Hospital,Department Of General Pediatric Surgery,Houston, TX, USA 4Texas Children’s Hospital,Outcomes And Impact Service,Houston, TX, USA 5University Of Alabama,Department Of Surgery,Birmingham, AL, USA 6University Of Texas Health Science Center – San Antonio,Department Of Surgery,San Antonio, TX, USA 7University Of Washington,Department Of Surgery,Seattle, WA, USA 8University Of Southern California,Department Of Surgery,Los Angeles, CA, USA 9Tufts University,Department Of Surgery,Portland, ME, USA
Introduction: Embryonal sarcoma of the liver (ES) is a rare tumor that primarily afflicts children. With an incidence of 1 per million and less than 100 cases reported in the literature since 1955, few data exist to help guide therapy and counsel patients. The aim was to identify factors that correlate with survival in patients with ES.
Methods: The National Cancer Data Base was queried for primary cases of ES diagnosed between 1998 and 2011. Data were examined using log-rank comparisons of Kaplan-Meier survival curves. Multivariate analysis was performed using a Cox proportional hazards model. Chemotherapy and surgery together are referred to as combined therapy. Survival is reported as 5-year overall survival (OS).
Results: There were 156 total patients, 107 pediatric (<18y) and 49 adult (≥18y). The OS of the entire study group was 72%. On univariate analysis, the OS of pediatric patients was significantly better than that of adults (82% vs. 52%; p<0.01). Regarding treatment, the OS of patients who had combined therapy was greater than that of patients who had one of the other treatment modalities, which were surgery only, chemotherapy only, and no treatment (75% vs. 48%, 29%, 20%, respectively) (Figure A). On subgroup analysis based on age, the OS of children who had received combined therapy was greater than that of adults (88% vs. 32%; p<0.01) (Figure B). When other factors were examined, greater OS was associated with high vs. low income (79% vs. 18%; p=0.02) and private vs. public insurance (51% vs. 12%; p<0.01) in adults. This difference based on income (86% vs. 78%; p=0.22) and insurance status (84% vs. 80%; p=0.49) was not seen in children. Metastatic disease, tumor size, race, and ethnicity were not significant factors in either age group. For all patients, factors that improved OS in the multivariate analysis (shown as hazard ratio; 95% confidence interval) were: surgical resection (0.094; 0.015 – 0.62), combined therapy (0.083; 0.017 – 0.41), negative surgical margins (0.16; 0.37 – 0.68), and age <18y (0.14; 0.057 – 0.35). No factors significantly affected OS in the adult group. In children, combined therapy (0.044; 0.007 – 0.28) was associated with greater OS.
Conclusion: Superior OS was associated with age <18y, negative surgical margins, surgical resection, and combined therapy. The optimal treatment for ES appears to be combined therapy. Patients with ES seem to respond differently to treatment depending on age; survival benefit is greater in children than in adults. Further study with a larger patient population may identify additional prognostic factors related to survival in these patients.
