ESTRO 2021 Abstract Book

S1369

ESTRO 2021

both energies. The final model reduced discrepancies in high dose regions blow 1% for all plans but for C- Shape Hard plan 10MV which was 1,47%.

Conclusion MLC parameters optimization was the most difficult part due to poor documentation of the process at the TPS manual. Two different methods were applied to adjust MLC parameters and the final parameter selection was decided based on the agreement between planned dose distributions calculated by the treatment planning system (TPS) and measured dose distributions of the clinical TG119 plans.

PO-1649 Correcting isocentric shifts in plans for cerebral targets S. Wegener 1 , R. Schindhelm 1 , O.A. Sauer 1 1 University of Wuerzburg, Radiation Oncology, Wuerzburg, Germany

Purpose or Objective Small cerebral targets are treated by stereotactic irradiations on conventional Linacs. Any Linac will show non-ideal alignment of the gantry, collimator and table axes and gantry flex or wobble to a different degree, influencing the target coverage. We observed differences between our three matched machines in the past. Here, we studied how taking into account these shifts in treatment planning influences the agreement between measured and expected dose. Materials and Methods Isocentric shifts were quantified for an Elekta Synergy Agility Linac using the QualiForMed ISO-CBCT+ test objects and software, yielding the shift between kV and MV isocenters, the gantry flex and wobble as well as the positions of the table and collimator rotation axes. The obtained shifts were included into a script that corrects the position of the isocenter for each field in the Philips Pinnacle treatment planning system. The script was applied to the beam configuration prior to dose calculation. 15 stereotactic treatment plans of cerebral metastases (0.34 cm³ to 26.53 cm³) comprising 9 to 11 beams were investigated; 54 different gantry and table combinations in total. Unmodified plans and modified plans, taking the isocentric shifts into account were measured using the Sun Nuclear SRS-MapCHECK in combination with the Stereophan phantom. The mean gamma was obtained for each plan. The unmodified plans were also measured on a second Linac, known to show the smallest geometric deviations of the three available machines. Results Shifts of up to 0.8 mm between kV and MV isocenter, a table rotation axis 0.9 mm off the kV isocenter and gantry flex with an amplitude of 1.1 mm were found. For eight of the usually smaller PTVs, D99 values declined more than 5% by simulating these shifts. The average gamma (2%/2mm, abs, 20% threshold) was reduced from 0.53 for the unmodified plans to 0.31 when including the isocentric corrections (Figure 1). Thus, Linac 1 reached the accuracy level of Linac 2, the one with the smallest geometric shifts.

Conclusion It is feasible to correct geometrical deviations of a Linac during the planning process. The produced treatment plans showed improved agreement between measurement and calculation compared to unaltered plans. For clinical use, changes in the geometrical shifts of the Linac as well as the correct application of the script for individual patients need to be closely monitored. PO-1650 Investigating dosimetric accuracy of SRS treatment plans for brain metastases G. Mok 1 , J.H. Phua 1 , H.Q. Tan 2 , K.W. Ang 2 , S.Y. Park 2 , C.L.J. Lee 2 1 National Cancer Centre Singapore, Division of Radiation Oncology (DRO), Singapore, Singapore; 2 National Cancer Centre Singapore, Division of Radiation Oncology (DRO), Singapore, Singapore

Purpose or Objective National Cancer Center Singapore (NCCS) has recently begun Stereotactic radiosurgery (SRS) treatment with