[P292] Dosimetric verification and clinical evaluation of PRIMO as an independent Monte-Carlo-based dose verification tool

• Published in: Physica medica 2018-08, Vol.52, p.184-184
• Main Authors: Berthes, Sarah, Großmann, Sascha, Schmidberger, Heinz, Brualla, Lorenzo, Rodriguez, Miguel, Karle, Heiko
• Format: Article
• Language: English
• ISSN: 1120-1797; 1724-191X
• DOI: 10.1016/j.ejmp.2018.06.566

Radiotherapy treatment depends on the accuracy of the dose delivery in patients. The purpose of this study is to test an independent Monte-Carlo dose calculation program (PRIMO) to verify dose distributions of 6-MV-FF VMAT plans. A Varian TrueBeam accelerator in 6-MV-FF mode was commissioned in PRIMO (V.0.3.1.1577) using Varian’s phase space files. Field sizes from 3 × 3 cm2 up to 40 × 40 cm2 were simulated in PRIMO and compared to the measured basic beam data. Pass rates greater 99% were achieved using a gamma criterion of 2% and 2 mm. Furthermore six VMAT plans (one glioblastoma, two prostate and three head neck plans) were initially calculated with Eclipse (V.13.1) using the Acuros-XB algorithm for dose calculation (dose to medium). These plans were transferred to the Octavius verification phantom (PTW Freiburg) geometry for all verification purposes. Phantom dose calculations with PRIMO were performed using the dose planning mode (DPM). For dose verification the dose distributions from Eclipse were imported in PRIMO via DICOM-RT and γ3mm,3%, γ2mm,2% and γ1mm,1% were applied. Additionally the results were compared to 2D ionisation chamber array (PTW Octavius 729) measurements and portal dosimetry measurements using the MV imaging unit of the linac (aS1200). The predicted portal dose was calculated with the Portal Dose Image Prediction algorithm (V.13.0.26). The comparison of the calculated 2D dose distributions in the ionization chamber plane of the Octavius detector 729 resulted in an average accordance between PRIMO and Eclipse of (99.9 ± 0.1)%, (98.3 ± 1.9)% and (82.5 ± 7.0)%, γ3mm,3%, γ2mm,2% and γ1mm,1% respectively. The 2D array dose measurements showed an average agreement to the Acuros-XB calculations of (98.1 ± 0.6)%, (89.9 ± 2.2)% and (53.7 ± 4.9)% (same gamma criteria as previously). The portal dosimetry analysis showed an average compliance between simulations and measurements of (99.0±1.5)%, (96.4 ± 2.8)% and (67.3 ± 10.0)% (gamma criteria as mentioned above). PRIMO is a Monte-Carlo-based dose simulation program which gives the possibility of independent dose verification in clinics. In future cumbersome array measurements could be replaced by simulations. Further investigations on the sensitivity of PRIMO for MLC misalignment and dose errors have to be done.