A Custom Small Field AAA Beam Model for 6X FFF: Validation and Cross Comparison on Three Truebeams for Dose Delivery Accuracy
Abstract
Purpose
Small field beam models present unique challenges for dose delivery accuracy. In this study we tested a custom AAA beam model for 6XFFF small fields with a millennium MLC and validate its delivery accuracy across three TrueBeams.
Methods
A custom AAA beam model was built for 6XFFF small fields by tuning DLG, effective target spot size on Eclipse 15.6. For each iteration of DLG and target spot size combination a set of clinical plans were calculated and delivery accuracy was tested with PinPoint ionization chamber (PTW 31022), portal dosimetry and SRSMapcheck. After arriving at the optimum custom beam model based on Eclipse calculated vs measured dose difference another set of idealized geometry plans and clinical plans were calculated delivered on three TrueBeams with similar DLG and target size for cross comparison and validation.
Results
A set of DLG ranging 1.5 mm to 0.7 mm was tested. The optimal beam model (calculated vs measured dose difference less than 1% on PinPoint chamber, portal dosimetry and SRSMpacheck) for AAA 6XFFF small fields had DLG 0.9 mm and effective target spot size of 0.1 mm for both x and y directions. Clinical plans with modulation factors ranging from 3 to 6.5 were tested which resulted in delivery accuracy ranging from 0.1% to 1.8%. All clinical test plans had 95% absolute passing with gamma passing criteria of 1mm and 2%. The same clinical plans delivered on two beam matched TrueBeams resulted in dose delivery accuracy of less than 2%.
Conclusion
DLG had larger impact on highly modulated small targets compared to effective target size. DLG difference of 0.1mm had 2% difference in dose delivery for targets less than 1cc. Beam models for small fields require careful tuning and validation.