ESTRO 37 Abstract book

S978

ESTRO 37

EP-1815 Comparison of independent Monte Carlo calculations with measurements of spot scanned proton fields A. Fredh 1 , C. Winterhalter 1 , E. Fura 1 , A. Bolsi 1 , S. Safai 1 , D. Weber 1 , A. Lomax 1 1 Paul Scherrer Institute PSI, Center for Proton Therapy, Villigen PSI, Switzerland Purpose or Objective Patient specific QA is time, resource and beam time consuming, all of which are limited in a typical proton therapy facility. Therefore it would be ideal to replace measurements with an independent dose calculation, which in this case is Monte Carlo (MC) based. As a step towards clinical use of MC calculations as a method for patient specific QA, measurements of clinical plans have been compared with independent MC calculations. Material and Methods 104 fields from 29 clinically delivered plans for 16 different patients, all calculated using our in-house treatment planning system, have been simulated in water using TOPAS 3.0.pl (Geant 4.10.02.p01). As part of our patient specific QA program, all those fields have also been measured with a 2D ion chamber array (PTW seven29XDR) in a water phantom. The measurements were done at different depths, all corresponding to the middle of the spread out Bragg peak and they have been compared with the calculated MC dose distributions. After translational corrections of the measured dose distributions to correct for experimental setup inaccuracies (maximum ± 2mm), the mean dose difference of all measured points have been determined and recorded for each field. The results were divided into 3 groups; fields with (IN) and without (OUT) range shifter, and mixed fields (AUTO), in which the range shifter is automatically inserted only for pencil beams within a field with ranges less than 4.2cm (WET). The AUTO mode has recently been introduced in our clinic and has now become the default for most treatments. Results Dose differences between measurements and MC calculations are shown in figure 1. The mean difference is -1.05%±1.00% for all fields. 99 (95.2%) of the 104 fields are within ±2% of this mean value (yellow area in figure 1), and all of the fields are within ±2.25%. No significant differences are found when grouping the results depending on the range shifter: mean value of - 1.14±0.92 % for fields with range shifter IN (red circle), - 0.63±1.17 % for range shifter OUT (blue square) and - 1.19±1.12 % for AUTO range shifter (black diamond)

Conclusion Sources of the systematic differences between measurements and MC calculations of -1% are currently under investigation. Once this has been accounted for, all MC and measurements results are within ±2.25% and the MC model can be used as an independent dose calculation and replace measurements for patient specific QA. The results also show that the modeling of the range shifter is accurate and the differences between fields with no, fixed or automatic range shifter are within measurement uncertainties. EP-1816 Moving from Clinac to TrueBeam for TSET: Beam characteristics and commissioning A. Mambakam Ravindran 1 , S. Esteve Sanchez 1 , R.B. King 1 , M.W.D. Grattan 1 1 Northern Ireland Cancer Centre, Radiotherapy Medical Physics, BELFAST, United Kingdom Purpose or Objective Following the replacement of two Varian Clinac 2100 C/D linear accelerators with Varian TrueBeam (TB) linacs, it was necessary to recommission the total skin electron therapy (TSET) beam on the new TB machines. This work reports on the acquisition of the TSET beam characteristics on the two TB accelerators and the comparison with the Clinac TSET beam data. Material and Methods The 6 MeV high dose rate total skin electron (HDTSe) mode is employed for the modified Stanford dual angle rotational TSET treatments. The optimal gantry angle pair was determined by moving the gantry in increments of 0.5° in order to achieve the position closest to where the meter reading for each of the dual fields was half of a single direct beam. The following dosimetric characteristics for the dual angle beams were measured: (a) A percentage depth ionization curve was measured both at the central axis and off axis with a parallel plate (NACP) chamber and converted to a depth dose curve by applying appropriate stopping power ratios (b)Lateral and vertical beam profiles were measured with the NACP chamber in a solid water phantom mounted on a beam mapping board where the phantom assembly can be moved to any point in the measurement plane (c) Composite PDDs resulting from six dual fields were measured with both Gafchromic TM EBT3 film and thermoluminescent dosimeters (TLD) sandwiched between a wax insert in a Rando phantom (d) Overlap factor - which is the ratio of dose at depth maximum for a single pair of beams to dose on the surface of the phantom from all six pairs of beams when maintaining the same MUs was measured using TLDs (e) The monitor chamber linearity and reproducibility were measured and (f) The beam output was calibrated. Results A comparison of the data for the Clinacs and two TB linacs is given in Table 1. The depth dose measured on the central axis and off-axis were in agreement to better than 1%. The dose profile uniformity and PDD characteristics were in line with the European Organization for Research and Treatment of Cancer (EORTC) guidelines. The monitor linearity and reproducibility were found to be within 0.4% and 0.1 % respectively. It was found that to deliver the same dose of 78 cGy at d max , requires 750 MU for the Clinac and 2112 MU for the TB. This is due to the difference in sensitivity of the monitor chambers between the two machines