ESTRO 36 Abstract Book

S792 ESTRO 36 2017 _______________________________________________________________________________________________

Communications Department, Alcala de Henares-Madrid, Spain Purpose or Objective In this study, we present a new method for portal dosimetry. CT images of the Electronical Portal Imaging Device (EPID) were used as phantom images for dose calculation. The clinical beam model and beam energy, in the treatment planning system, were used to calculate dose over the EPID. Material and Methods The method was developed for a Varian Clinac 21-EX (Varian Medical Systems, USA), with a nominal photon energy of 6 MV, equipped with a Varian aS1000 EPID. Pinnacle 8.0m (Philips Medical Systems, NL) was used for treatment planning calculations. Matlab® v2012a (Mathworks, USA) was employed to develop code for calculations involving backscatter and output correction factors. The EPID was calibrated, following the manufacturer procedure, and then unmounted from the linear accelerator and scanned to acquire CT images of the EPID (Fig. 1) on an Aquilion LB (Toshiba Medical Systems, Japan). These CT images were imported into the Pinnacle planning system. The imported images were used as a quality assurance phantom to calculate dose on the image plane, which was considered as the predicted portal dose. Two sliding-window IMRT treatment plans, a prostate and a head and neck case, were delivered, measured and analyzed with both with the EPID and with MatriXX (IBA Dosimetry, Germany), as an independent measurement method. Matlab code was used to calculate EPID arm backscattering and output factor corrections. Gamma index comparison (3 %, 3 mm) was made for the EPID and MatriXX dose planes versus the calculated dose planes with OmniPro ImRT (IBA Dosimetry).

The obtained results show the validity of the method presented here. This method can be easily implemented into clinic, as no additional modeling of the clinical beam is necessary. The main advantage of this method is that portal dose prediction is calculated with the same algorithm and energy beam model used for patient treatment planning dose distribution calculations. EP-1497 Dosimetric effect of the Elekta Fraxion cranial immobilization system and dose calculation accuracy Purpose or Objective Devices external to the patient may cause an increase in the skin dose, as well as modify the dose distribution and hence the tumor dose. This study describes the effect on this parameters caused by the Elekta Fraxion cranial immobilization system. The effect of the inclusion of Fraxion in ElektaMonaco treatment planning system (v. 5.00.00) was also checked. Material and Methods To study the dose attenuation a cylindrical phantom was placed over the Elekta Fraxion with a CC13 Scanditronix- Wellhofer ionization chamber located in the central insert at the linac isocenter. Dose measurements were performed for two open fields, 10x10 cm and other smaller 5x5 cm, as Fraxion is used mainly for radiosurgery treatments. The gantry angles were the ones which cross Fraxion (135º - 225º, 5º-10º increment, IEC gantry angles). Calculated and measured doses are the average doses of symmetrical angles from 180º. Reference dose without Fraxion was the average dose at 0º, 90º, and 270º. 100 MU were delivered at each angle. All measured doses were compared with the ones calculated with Monaco. To measure the skin dose and the dose distribution in the Build-up region, several radiochromic Films EBT3 were placed at linac CAX between the slabs of a RW3 phantom placed over Fraxion (SSD= 90 cm) and read using FilmQA Pro software. Films were situated at the surface, 0.5 cm, 1.5cm depth and the linac isocenter. 200 MU were delivered for 10x10 and 5x5 open field sizes and 0º gantry angle. Once irradiated and removed, another set of films were placed under the phantom, in contact with Fraxion, and at 0.5 cm and 1.5 cm from Fraxion, as well as at the linac isocenter. Additional films were located 1 cm away from CAX as in this section Fraxion is wider. Same field sizes and MU at 180º were employed. Results Table 1 shows the comparison between measured and calculated transmitted dose with and without Fraxion in the calculation. Measurements show a 1% attenuation for 180º gantry angle as stated on the Fraxion manual, but this attenuation can be as high as 5 % (5x5 open field) or 6 % (10x10 open field) for 150º gantry angle, as with this angle, the beam traverses the thickest part of the Fraxion. If Fraxion is not included in the calculation, Monaco calculation can result in a 7 % difference between measured and calculated doses, while with Fraxion in the calculation, the maximum difference is 1.5% (10x10, 150º). Table 2 shows the evaluated skin dose increment caused by Fraxion, and compares calculated and scanned values. Fraxion increases 3.8 times the surface dose, and by 17% at 0.5 cm depth, which can be calculated by Monaco with a difference lower than 1% if Fraxion is included in the C. Ferrer 1 , C. Huertas 1 , R. Plaza 1 , A. Serrada 1 1 Hospital universitaria La Paz, Radiofísica y Radioprotección, Madrid, Spain

Figure 1. Acquired CT images of the Varian aS1000 EPID. Results For plans verified with EPID, Gamma index pass rate were 98.6% and 96.5% for prostate (Table 1) and head and neck case, respectively. Dose differences (EPID vs planned) were -0.7% and -0.4%. For MatriXX measurements, the results are very similar: gamma pass rate of 97.2% for prostate and 97.9% for head and neck, and dose differences (MatriXX vs planned) of - 1.4% and -0.8%, respectively.

Gamma (3 %, 3 mm)EPID

Dose diff EPID (%)

Gamma (3 %, 3 mm)MatriXX

Dose diffMatriXX (%)

Field

1 2 3 4 5 6 7

98.5% 98.6% 98.9% 98.6% 99.0% 98.7% 98.0%

-2.0 -1.0 -0.5 -0.6 -0.3 -0.1 -0.6 -0.7

96.5% 98.1% 96.0% 96.5% 99.0% 97.3% 97.2%

-1.6 -1.1 -1.5 -1.1 -1.5 -1.2

Average 98.6% -1.4 Table 1. Gamma index and dose difference results for prostate treatment. Conclusion