An Evaluation of Beam Matching for Unique Linear Accelerator Models
Abstract
Purpose
Beam matching allows RT delivery using a shared beam model for identical linear accelerator models. However, LINACs of differing models may also match depending on specific LINAC performance. One such configuration was analyzed in this report: a Varian TrueBeam against a Varian Clinac iX.
Methods
Primary concerns for beam matching include open field diagonal profiles, penumbras, and large and small fields output factors. The following tests were used highlight variations in LINAC performance: Clinical TPS model comparison Annual QA performance TPS model differences in PTV/OAR metrics for 20 clinical patients Patient specific QA measurements on MapCHECK3 diode-array for 100 volumetric modulated arc therapy (VMAT) clinical patients. Beam model and patient specific QA evaluated per AAPM MPPG5.b and TG-218. iX QA results evaluated against TrueBeam baselines using AAPM TG-142.
Results
TPS Model Comparison: TrueBeam and Clinac iX output factors agreed within 0.8%, with a mean difference of 0.1%. Percent depth doses agreed within 0.3% past dmax. Open field profiles agreed with a mean difference of 0.05% and diagonal profiles agreed with a mean difference of 0.01%. Annual Comparison: Profiles agreed within .6%. Output factors agreed within 0.7%, with a mean of -0.13%. Output factors agreed within 0.7%, with a mean of -0.13%. PDD comparison agreed within 0.45% with a mean of -0.13%. Patient Plan Comparison: 3D and VMAT patients plans between TPS models for structures agreed within 1% across several metrics. Patient Specific QA: All MapCHECK3 measurements had a pass rate >90% with an average of 99.38% and 97.44% for the TrueBeam and Clinac iX, respectively.
Conclusion
A high degree of agreement was found between the two distinct LINACs across all tested metrics. To consider beam matching for other disparate LINACS, consider if any performance differences are significant enough to outweigh the benefits of a beam matched environment.