C. C. Vining1, A. J. Sinnamon1, M. G. Neuwirth1, B. L. Ecker1, R. R. Kelz1, D. L. Fraker1, R. E. Roses1, G. C. Karakousis1 1Hospital Of The University Of Pennsylvania,Department Of Endocrine And Oncologic Surgery,Philadelphia, PA, USA
Introduction:
The management of stage I-III synovial sarcoma is primarily surgical with consideration of adjuvant radiation. However, the role of adjuvant chemotherapy (AC) remains less well-defined limited to small institutional series. Using a large national dataset we sought to identify factors associated with receipt of AC and to evaluate impact on overall survival (OS).
Methods:
Patients with stage I-III synovial sarcoma 2004-2012 undergoing resection were identified in the National Cancer Data Base. Patients were excluded if they received any neoadjuvant therapy or had incomplete grade, size, or adjuvant therapy data. Chi-square and multivariable logistic regression was used to identify factors associated with receipt of AC (univariate p-value<0.05 for inclusion in multivariable model). Clinicopathologic factors and adjuvant therapies associated with improved OS were identified with univariate and multivariable Cox proportional hazard modeling and the Kaplan-Meier method, applied to the overall cohort and to subgroups stratified by stage.
Results:
From 2004-2012, 597 patients underwent resection with evaluable data. Median age was 41 (IQR 29-54) and 302 were female. Four-hundred sixteen tumors were high grade, 204 were 5-10cm, and 97 were >10cm. One hundred eighty-four patients received AC, 311 received adjuvant radiation, and 102 received both. Factors associated with receipt of AC in multivariable analysis included age <40y (OR 2.49), high grade pathology (OR 2.04), size (5-10cm OR 2.45; >10cm OR 3.52), and positive margins NOS (OR 5.67). In multivariable analysis, factors significantly associated with worse OS included age>40y (HR 2.57), Charlson-Deyo comorbidity score ≥ 2 (HR 3.18), monophasic histology (HR 2.89), size 5-10cm (HR 2.06), >10cm (HR 2.12), high grade (HR 3.24), positive lymph nodes (HR 15.9), omission of adjuvant radiation (HR 1.64), and macroscopic surgical margins (n=5, HR 8.53); notably, AC was not significantly associated with improved OS. However, when patients were stratified by stage, AC was associated with improved OS among stage III patients but not in lower stage groups. This association remained significant in multivariable analysis (HR 0.59, p=0.037). The stage III group (n=227) was comprised almost entirely of high grade tumors >5cm, as confirmed LN metastasis were rare (n=3). Ninety-eight (43%) of these patients received AC, compared to 24% of stage II (82/339), and 13% (4/31) of stage I.
Conclusion:
In this large national dataset AC in resected synovial sarcoma was associated with improved OS in patients with stage III disease but not in lower stages. Less restricted use of this therapy may be warranted considering it was administered to less than half of these patients.
