ESTRO 2022 - Abstract Book

S709

Abstract book

ESTRO 2022

is needed after automatic optimization. The suggested approach is under clinical implementation, fully replacing manual plan optimization for left whole-breast tomotherapy with TD. The study is supported by an AIRC grant (IG23150)

MO-0791 Towards a novel way to optimize the DLG using plan complexity metrics for SBRT/SRS using FFF beams

M. Oellers 1 , A. Swinnen 2 , A. Vaniqui 2 , F. Verhaegen 2

1 MAASTRO , Medical Physics, MAASTRICHT, The Netherlands; 2 MAASTRO, Medical Physics, MAASTRICHT, The Netherlands

Purpose or Objective There is a trend towards hypo-fractioned stereotactic treatments (SBRT/SRS) which often consist of highly modulated flattening filter free (FFF) treatment fields. Varian’s MLC model in Eclipse allows tuning of only 2 parameters, leaf transmission and dosimetric leaf gap (DLG). Recent literature [1] showed that a single DLG value, determined using a modified sweeping gap technique [2], may lead to relevant dose discrepancies between measured and calculated dose for stereotactic plans. Adjustments are done iteratively by measuring and verifying dose correspondences to obtain a tuned DLG. In this work, a method based on plan complexity metrics is described that predicts when dose discrepancies may start to arise and effectively help to determine the optimal DLG. Materials and Methods Modulated Complexity Scores (MCS) [3] were calculated for 6 VMAT plans based on a simple target geometry with decreasing MLC modulation using Eclipse’s Aperture Shape Controller (ASC) and for 21 clinical SBRT cases of prostate, kidney, spine and liver cancer, planned with 10MV FFF. Another 33 clinical SRS treatments of single and multiple brain lesions planned with 6MV FFF were also included. All plans were calculated using the Acuros algorithm (v15) . Dose verification was performed with PTW Octavius phantom and 1000 SRS detector. A gamma pass rate (GPR) of 2%/2mm with a 80% isodose threshold, representing the high dose region, was used for dose evaluation. A 90% pass rate of pixels with a gamma below 1 was clinically accepted. Results For 6 and 10MV FFF, respectively, 8 (24%) and 13 (62%) plans failed to meet the GPR above 90% (Fig1). For 6MV FFF no correlation could be observed between MCS and GPR, indicating a near optimal DLG. For the 10MV FFF plans a trend was seen between increasing MCS and GPR, which was also observed in the phantom experiment (Fig2). A MCS below 0.4 did not meet the GPR of 90%. Adaption of our 10MV FFF DLG value lead to GPR above 90% even for the high dose regions.

Fig1

Fig2

Conclusion