173 - Interplay of Radiation and Cyclophosphamide Doses on Local Control in Children with Group III Rhabdomyosarcoma: An Analysis of Studies from the Childrens Oncology Group

M. Kozak¹, W. Xue², N. Collins³, A. Qumsera², E. Rudzinski⁴, D. Rodeberg⁵, S. L. Wolden⁶, R. Venkatramani⁷, and J. C. Breneman⁸; ¹Department of Radiation Oncology, University of Texas Southwestern Medical Center, Dallas, TX, ²University of Florida, Gainesville, FL, ³Dana Farber Cancer Institute, Boston, MA, ⁴Seattle Childrens Hospital, Seattle, WA, ⁵University of Kentucky, Lexington, KY, ⁶Memorial Sloan Kettering Cancer Center, New York, NY, ⁷Baylor College of Medicine, Texas Children’s Hospital, Houston, TX, ⁸Department of Radiation Oncology, University of Cincinnati, Cincinnati, OH

Purpose/Objective(s): To determine whether there is an interplay effect between radiation dose and alkylating agent dose for the endpoint of local control in patients with Group III RMS. Materials/

Methods: The study population consisted of patients with Group III RMS enrolled on Children’s Oncology Group (COG) studies D9602, D9803, ARST0331 or ARST0531. Data collected included patient, gender, histology, tumor site, TNM stage, radiation (RT) dose and fractionation, and local, regional or distant failure. RT BED using an a/ß = 10 was calculated for each patient, and cyclophosphamide equivalent dose (CED) was assigned to patients based on the study treatment arm to which they were assigned. Local failure (LF) was defined as progression or relapse at the primary tumor site and was calculated from time of study entry. BED was segmented into 5 “bins” (<42, 42-46, 47-52, 53-58, and >58) and CED was segmented into 4 bins (<=4.8, 8.4, 16.8, and >20). Chi square or Fisher exact tests were used to compare patient characteristics. Logistic regression analysis was used to assess the impact of a variety of BED, CED, and clinicopathologic factors on LF.
Results: There were 792 patients who met inclusion criteria. Multivariable analysis of tumor/demographic factors showed tumor size >5cm to be predictive of local failure as seen in prior reports (p<0.01). No trend for local control was observed for combinations of BED and CED.
Conclusion: This analysis failed to show a relationship between local control and various combinations of BED and CED for patients with Group III RMS. Tumor size >5 cm was again shown to be predictive of local failure. Additional analysis including results of the recently completed ARST1431 study will be performed to further test the hypothesis that BED and CED both contribute to local control in patients with Group III RMS.

Abstract 173 – Table 1: Local Failures

	CED
Total BED	<=4.8	8.4	16.8	>20	p-value
< 42	1 (16.67%)	0	0	4 (16.67%)	0.1878
42-46	0	0	0	1 (10.00%)	0.5605
47-52	0	0	1 (25.00%)	5 (19.23%)	0.3014
53-58	12 (10.62%)	3 (33.33%)	1 (33.33%)	4 (12.50%)	0.0993
>58	2 (10.53%)	38 (26.21%)	31 (20.67%)	46 (19.74%)	0.3085
p-value	0.0123	0.0851	0.3604	0.0011