ESTRO 2020 Abstract book

S355 ESTRO 2020

induction of damage during the pulse. Using a statistical model the influence of oxygenation on the introduction of complex damage is estimated and can be applied to a dose deposition model. The comparison with MCDS is performed using standard deviation and range of the residual difference between simulated and calculated. Results of FTPS are compared with data obtained from a flash crypt survival experiment in mice treated with a 6MeV electron 3.5μs pulse at our centre. In addition, we reconstructed the experiments by Montay-Gruel (RO 2017) and the historical in-vitro experiment by Town (Nature 1967). Finally, a hypothetical 6 MV photon 3.5 μs pulse machine was constructed in FTPS illustrating the possibilities. Results Table 1 shows the parameters of the distribution of our model compared to the results obtained using MCDS from 0 to 60% pO2. All numbers indicate #CD/Gy/cell/Gbp. SD min max electrons E = [ 1e-07 - 10 MeV] 0.16 -1.18 1.18 protons E = [ 0.01 - 1000MeV] 0.21 -0.52 1.37 alpha-particle E = [ 0.01 - 10000 MeV] 0.08 -0.28 0.56 Carbon ions E = Ibid 0.13 -1.12 0.52 All experimental data could be reproduced through choice of parameters like depletion rates and initial oxygenation levels. Agreement was found at 5mmHg partial oxgyen pressure (pO2) for the crypt experiments. Good agreement was found at the 10 mmHg pO2 for the Montay paper, while the in-vitro paper needed 76mmHg pO2. Figure 1 illustrates the effect of a three beam (6Gy at Dmax) sequential (rotating gantry) treatment on a phantom. The images show differences of dose with effective dose (i.e. dose having generating the same damage in normal mode, red = increased sparing). Four levels of initial oxygenation are illustrated.

Proffered Papers: Proffered papers 31: Quantitative imaging and radiomics

OC-0582 Dosimetric stability of high FDG-uptake volumes during dose escalated RT of NSCLC D. Sloth Møller 1 , C.M. Lutz 1 , A. Haraldsen 2 , A.A. Khalil 3 , M.M. Knap 3 , L. Hoffmann 1 1 Aarhus University Hospital, Department of Medical Physics, Aarhus C, Denmark ; 2 Aarhus University Hospital, Department of Nuclear Medicine and PET-Centre, Aarhus C, Denmark ; 3 Aarhus University Hospital, Department of Oncology, Aarhus C, Denmark Purpose or Objective In a multi-centre dose escalation trial (NARLAL2), locally advanced non-small cell lung cancer (LA-NSCLC) patients are randomized between heterogeneous dose escalation (mean dose up to 95Gy/33fx/5) and standard homogeneous dose (66Gy/33fx/5). The heterogeneous dose escalation is guided by high 18F-fluorodeoxyglucose (18-FDG) uptake volumes. We investigate the geometric and dosimetric stability of these high FDG uptake volumes for successive PET/CT scans during the first weeks of radiotherapy (RT) for escalated treatment plans regardless Sixty-two patients were included in the study. Three successive PET/CT scans were acquired at the same scanner: two weeks before RT (PET/CT0), one week (PET/CT1) and two weeks (PET/CT2) after RT start. The GTV of the primary tumour and a sub-volume defined by the 50% threshold of SUVpeak were delineated on each scan (V0 , V1 and V2). CT1 and CT2 were rigidly registered to CT0 for GTV. The overlap fraction OF 0X =(V0 ∩ VX)/min(V0,VX), X=1,2 was determined for all patients. The impact of changes in FDG uptake on dose coverage was investigated by recalculation of the escalated dose plan on CT2 with V2. A Chi-square test was used to test if low OF 02 (<0.7) correlated with large dose decrease (>3%). Results Eleven patients with GTV<4cm3 and five patients with only PET/CT1 were excluded from this analysis. In total 46 patients were analysed. Median [min;max] values of OF 01 and OF 02 were 0.84 [0.44;1.00] and 0.82 [0.38;1.00], respectively. In 40 (87%) and 36 (78%) of the patients, OF was above 70% at PET/CT1 and PET/CT2, respectively. These patients fell in two categories: In 44% of the patients, tumour shrinkage caused an increase in mean dose to V 02 by 1.5Gy [0.1;3.7], see Fig 1a. In the remaining 56%, dose decreased by -1.1Gy [-6.0;0.0] and, only minor changes were seen in tumour size, see Fig1b. In only one of the patients with OF above 70%, the dose decreased >3% due to a large decrease in SUVpeak at PET/CT2. This was in contrast to the ten patients (22%) with OF 02 below 70%, where the V 02 dose decreased by median -4.3% [-17.4;-2.3] in seven patients while the dose changed by less than 1% in the remaining three patients. The primary cause of OF 02 <70% was decline of SUVpeak to ~2 times background activity, changing the region included in V2, see Fig 1c. Additionally, in two patients necrotic regions partly dissolved and in one patient a large atelectasis at tumour site disappeared. In Fig. 2, OF 01 and OF 02 are shown for all patients. There was a significant correlation between patients with OF 02 <70% and patients with mean dose decrease >3% to V2 (p<0.001). of randomisation result. Material and Methods

Conclusion All experimental data fall within the envelope of the possible parameters used in the FTPS. We find a critical dependency of the size of the protective effect on the initial oxygenation level, which can be overcome by choosing high enough doses, enabling the re-introduction of in-vitro cell experiments.