ESTRO 38 Abstract book

S47 ESTRO 38

modifications described below, can accurately convert portal images of small unflattened fields to dose distributions in patients and phantoms, both for static

Purpose or Objective Microbeam radiation therapy (MRT) is a still preclinical technique in radiation oncology, which uses arrays of micro planar beams. High doses are delivered in the areas of the beams (peak dose, several 100 Gy) and a very low dose in between the beams (valley dose). This technique was suggested in the treatment of lung cancer as it promises less healthy tissue damage but equal tumor control when compared to conventional radiation therapy. However, as typical sizes of micro-cavities in the lung (alveoli, bronchioles, etc.) are in the same order of magnitude as microbeam widths and spacings the lung microstructure may have an effect on the dose distribution inside the lung tissue. The aim of this study is to quantify these differences in MRT dose distributions between the inhomogeneous lung tissue and the assumption of a homogeneous water-air mixture in Monte Carlo simulations. Material and Methods All simulations were performed with the Geant4 tool kit (vers. 10.0 patch 2). Two different lung models were developed, model 1 consisting of homogeneous tissue (r=0.26 g/cm 3 ) and model 2 of a water filled volume with small spherical air cavities (r=100 mm) in a cubic face centered packing to reach the same mean density as in model 1. Shape and size of the simulated volumes correspond to the geometry of an ongoing dosimetric validation using radiochromic films. Experimentally a slab of gel foam imitates the porous lung material. For reference the lung tissue and lung model in the phantom were replaced by pure water in a third simulation. A 20x20 mm 2 radiation field was used for all simulations comprising of 50 microbeams with a beam width of 50 mm and a center to center distance of 400 mm. Results Simulation results show that in the inhomogeneous lung model the valley dose is up to 41% higher than for the homogeneous model. There is also an up to 2% higher peak dose observed in the inhomogeneous lung model. Subsequently the peak to valley dose ratio (PVDR) is reduced in the inhomogeneous lung tissue (mean PVDR = 48.2) when compared to the homogeneous lung model (mean PVDR = 60.5).

fields and VMAT arcs. Material and Methods

The back-projection model in Ref. [3] was modified to render it applicable for small fields dosimetry as follows: (i) New reference measurements using the PTW microdiamond detector (type 60019) were used as input data for commissioning to reduce the volume effect. (ii) The blurring EPID scatter kernel ([3] Eq. 5.11A), which mimics the ion chamber’s volume effect, was removed for commissioning and dose reconstruction. (iii) Square fields from 10x10cm 2 down to 1x1cm 2 were used to extract the primary signal, instead of the 23x23cm 2 down to 3x3cm 2 data in [3], better capturing the small fields regime. (iv) Alignment errors between the portal images and the CAX microdiamond data arising from limitations in the accuracy of the EPID positioning system were corrected in the commissioning software. Results Fig. 1 compares TPS dose and portal dosimetry for small fields using the virtual reconstruction method [4], for a 1x1cm 2 square field (a-b) and a VMAT plan (c-d). The standard reconstruction model (SRM) yields underdosage of about 8% compared to the TPS. Fig. (e) shows box plots of the dose difference at the isocenter for 15 VMAT plans expressed as a percentage. The small field model (SFM) reduces the discrepancy for all treatments to less than 3%, the tolerance value for clinical use. The mean dose error (not plotted) reduces from -2.9% to -0.6%.

Conclusion The EPID back-projection model in Ref. [3] was successfully extended to the small fields regime. With the modifications here described, portal dosimetry (in vivo as well as pre-treatment) can be used for accurate verification of both static fields and VMAT plans of small unflattened beams. References [1] The New York Times, February 24. [2] IAEA (2017) Technical Reports Series No. 438. Vienna, Austria. OC-0093 Microcavities in the lung affect the dose distribution in microbeam radiation therapy G. Hombrink 1,2,3 , J.J. Wilkens 1,2 , S.E. Combs 1,3 , S. Bartzsch 1,3 1 Klinikum rechts der Isar, Department of Radiation Oncology, Munich, Germany; 2 Technical University of Munich, Physics Department, Munich, Germany; 3 Helmholtz Zentrum German Research Center or Environmental Health, Institute of Innovative Radiotherapy, Neuherberg, Germany [3] Medical Physics 33, 259–273 (2006). [4] Physica Medica 37, 49 – 57 (2017).

Conclusion The results show that the lung microstructure has a strong impact on the MRT dose distribution and cannot be neglected. At equal mean density the simple assumption of a homogeneous air-water mixture overestimates the PVDR by 25% on average in our simulations. The effect can be explained by secondary electrons that are produced in