ESTRO 37 Abstract book

S927

ESTRO 37

EP-1732 Multimodal range verification for proton irradiation using MR and PET imaging A. Runz 1,2 , M. Runz 1,2 , H. Prokopp 1,2 , R. Dal Bello 1,2 , Y. Berker 1,2 , G. Echner 1,2 , P. Mann 1,2 1 German Cancer Research Center DKFZ, Medical Physics, Heidelberg, Germany 2 Heidelberger Institute for Radiooncology, Medical Physics, Heidelberg, Germany Purpose or Objective In proton therapy, various range verification systems in 1D (ionization chamber, IC), 2D (film) and pseudo-3D (IC- array) are implemented in clinical routine. However, the implementation of a systems allowing for a fast, exact and truly 3D proton range measurement would be a desirable tool which will be presented in this work. Material and Methods Proton range verification was realized by means of a modified polymer gel (PG). The gel used is the so-called PAGAT PG. It consists of 88% H 2 O and two different kind of monomers embedded within a gelatine matrix that start to polymerize upon irradiation. This process causes the relaxation time T 2 to locally vary which can be measured in 3D by magnetic resonance imaging (MRI). Furthermore, the H 2 O was enriched with a total of >90% 18 –Oxygen (18-O) causing a nuclear reaction upon proton irradiation. As a result, the β+ emitter 18-Fluor (18-F) is created which can be measured by Positron-Emission- Tomography (PET). The gel was filled inside a PMMA cube (60x60x60mm³) and irradiated 24h after production. A pencil beam with a total dose of 50Gy with a mean range of 34.5mm was planned (syngo RT Optimize Treatment Plan, Siemens). The proton irradiation was performed at the Heidelberg Ion-Beam Therapy center (HIT) and the phantom was measured 3h post irradiation on a hybrid 3T MR-PET Scanner (Siemens Biograph mMR). One major advantage is that both PET and MR images are automatically co-registered by the scanner. A multi-spin echo sequence with 32 equidistant echoes (TE=22.5ms – 720ms) and a resolution of 1x1x1mm³ was used for the quantitative T 2 measurement. For registration purposes with the planned dose, a high-resolution MR image (0.5x0.5x0.5mm³) was also acquired. For the PET acquisition, a total of 2x10 6 counts were measured and a CT-based attenuation map of the phantom was additionally acquired to retrospectively calculate the 18- F distribution within the phantom. This is necessary as the MR-based attenuation map was not able to correctly recognize the PMMA cube. Results The phantom was able to produce a significant amount of 18-F within the gel and the PG showed a sharp dose gradient. Signal profiles of one slice along the particle track of TPS, MR- and PET data are shown in Fig 1. The range of the dose profile, the falloff position of the PG and the maximum of the PET signal were respectively: 34.5mm, 35.2mm and 26.5mm. The PG presents a direct correlation with the dose deposition, whereas the F-18 signal has to be indirectly correlated to the TPS.

for irradiation. EBT3 and EBT-XD films, both cut to 4x4 cm 2 pieces, were aligned in the phantom in two different configurations. The first consisted of 10 films irradiated perpendicular to the beam axis. The second was a combination of a perpendicular and a parallel orientation, i.e. the first six were oriented perpendi- cular, the last one parallel. Parallel orientation was performed in order to get a high resolution at the end of SOBP as well as at the distal dose fall-off region. Seven RW3 slabs were customized for housing 5, 6 or 7 of TLD disks (diameter: 4.5 mm, thickness: 0.89 mm), which allows simultaneous irradiation at various depths along the beam. For absolute depth dose measurements a Roos ionization chamber (Type 34001, PTW, Freiburg) was used. Results Figure 1 shows the response of two sets of 41 TLDs, where one set was evaluated with individual correction factors obtained for each TLD. The uncertainties were reduced on average from 3.7 % down to 1.3 %. Figure 2 shows depth dose measurements of EBT3 films and TLDs. The dashed line represents the measurements from the Roos chamber (errors are below 0.6%). The results from EBT-XD film evaluation are currently processed.

Conclusion The performed measurements show that individually calibrated TLD-100 detectors are well suited for proton dosimetry in a SOBP, as demonstrated by the good agreement with Roos chamber measurements. TLD measurements at larger depths need to be conducted to further study quenching phenomena. EBT3 films showed an under response of up to 11% compared to Roos chamber measurements. Orienting the EBT3 film parallel to the beam proved to be feasible in high dose gradient regions.