Abstract Book

S1153

ESTRO 37

month after SBRT. The dose-response semeed two-phasic. A linear decline in hepatocyte function was observed with doses below 30 Gy. At higher doses the decline seemed less pronounced. A 50% reduction in liver function was quantified at 17.8 Gy (CI 95%: 12.0-25.8) delivered in three fractions. The mean recovery rate was highest in the volumes of liver receiving doses below 20 Gy. A linear correlation was observed between Kmet and SUV of volumes receiving varying doses. Conclusion A radiation dose dependent decrease in liver function was observed in patients treated with SBRT for liver tumors one month after treatment. Recovery of liver function was observed with radiation dose below 20 Gy. The dose- response relationship was the same for parametric images of Kmet and SUV. EP-2098 Bioluminescence tomography-guided radiation therapy for GBM in vivo Z. Deng 1 , X. Xu 1 , Z. Belcaid 2 , T. Garzon-Muvdi 2 , A. Luksik 2 , R. Maxwell 2 , I. Iordachita 3 , J. Yu 4 , M. Lim 2 , J.W. Wong 1 , K.K.H. Wang 1 1 Johns Hopkins University School of Medicine, Department of Radiation Oncology and Molecular Radiation Sciences, Baltimore- MD, USA 2 Johns Hopkins University School of Medicine, Department of Neurosurgery, Baltimore- MD, USA 3 Johns Hopkins University, Laboratory for Computational Sensing and Robotics, Baltimore- MD, USA 4 Shaanxi Normal University, School of Physics and Information Technology, Shanxi, China Purpose or Objective Current treatment outcome for GBM is dismal. Preclinical study is needed to develop new radiation therapy(RT) options. CBCT-guided small animal radiation research platform(SARRP) has gained its popularity. However, CBCT is limited in guiding RT for GBM because of low imaging contrast. Alternatively, we developed bioluminescence tomography(BLT) to guide SARRP irradiation. The BLT can accurately reconstruct the center of mass(CoM) of the GBM model. We further utilized contrast-enhanced CBCT to delineate GBM volume. The relationship between the GBM volume and time of growth provides an estimated irradiation volume to be used in BLT-guided RT. In this study, we will validate the BLT-guided RT and devise the margin and beam collimation based on the BLT-reconstructed CoM and the estimated tumor volume. Material and Methods Our BLT system consists of an optical assembly, a mobile cart and a moveable mouse bed. SARRP CBCT image is acquired for planning and to generate anatomic mesh for BLT reconstruction. To establish GBM model, GL261-Luc cells were injected into left striatum. Mice were subject to multispectral bioluminescence imaging, followed by SARRP CBCT and BLT reconstruction, 1-4 weeks after the implantation. Contrast-enhanced CBCT imaging was used to delineate GBM and to verify the localization accuracy of the BLT system. To reduce the uncertainty of BLT reconstruction, we will apply differential evolution algorithm to optimize the optical properties of mouse brain by minimizing the distance between the CoMs of BLT and contrast imaging. The residual uncertainty will be included in the margin for irradiation. The volume of contrast-labeled GBM vs. time of growth will be established to provide the irradiation volume. For the 1- week time point, it is still challenging to reveal the GBM volume with the contrast imaging. MRI will be used to assist the tumor delineation. Equivalent sphere of the average tumor volume at different time points of tumor growth will be used to design beam collimation. For irradiation, we will localize the GBM with BLT reconstructed CoM and apply the estimated volume and margin to collimate the radiation beam. H&E and γH2AX

staining will be used to validate the coverage of irradiation. Results Fig 1 shows a case of the BLT reconstruction. The 3D rendering (Fig 1d) shows 0.6mm difference between the CoMs of the contrast-enhanced and BLT volume.

We further studied if the BLT localization accuracy depends on tumor size by assessing the CoM location for a GBM that has grown 1-4 weeks (Fig 2, tumor size:12– 113mm 3 ). We found the average BLT CoM is within 1.3mm with that of contrast-enhanced volume and no volume dependence was found. The efficacy of using such margin for irradiation is being validated.

Conclusion Our results demonstrate the BLT-SARRP can be a novel system to guide radiation in vivo. By utilizing the BLT reconstructed CoM and the contrast-labeled GBM volume, we expect to provide optimal margin and collimation to deliver conformal irradiation for GBM model.