ESTRO 2022 - Abstract Book

S204

Abstract book

ESTRO 2022

C. Skinnerup Byskov 1 , C. Rønn Hansen 2,3,4 , R. Hedegaard Dahlrot 2,3,5 , L. Dysager 2 , T. Lignell Guldberg 6 , L. Haldbo-Classen 7 , S. Hansen 2 , C. Aaquist Haslund 6 , M. Høyer 3 , S. Lukacova 7 , A. Muhic 8 , P. Witt Nyström 9 , Y. Ramshad-Lassen 3 , A. Kirsten Trip 3,10 , B. Weber 7,3 , J. Folsted Kallehauge 3,10 1 Aarhus University Hospital, Department of Oncology, Aarhus N, Denmark; 2 Odense University Hospital, Department of Oncology, Odense, Denmark; 3 Aarhus University Hospital, Danish Centre for Particle Therapy, Aarhus, Denmark; 4 Odense University Hospital, Laboratory of Radiation Physics, Odense, Denmark; 5 University of Southern Denmark, Institute of Clinical Research, Odense, Denmark; 6 Aalborg University Hospital, Department of Oncology, Aalborg, Denmark; 7 Aarhus University Hospital, Department of Oncology, Aarhus, Denmark; 8 Rigshospitalet, Department of Oncology, Copenhagen, Denmark; 9 Uppsala University Hospital, Department of Oncology, Uppsala, Sweden; 10 Aarhus University, Department of Clinical Medicine, Aarhus, Denmark Purpose or Objective Proton therapy (PT) has the potential to reduce radiation-induced brain injuries in patients with grade 2 and 3 gliomas. The current decision for PT selection is based on sparse evidence and may thus become very subjective. In this situation, expert opinion can guide clinical decision making. This study aimed to determine which parameters are the main drivers for the experts' selection process for PT. Materials and Methods Newly optimised photon and proton plans were made for 42 previously treated glioma patients (Table 1). Each patient, including plans and relevant clinical parameters, was presented at two different workshops to clinical oncologists, who individually assigned each patient to either photon therapy (XT) or PT. Twenty-three patients were evaluated by 11 clinical oncologists at the first workshop and 19 patients were evaluated by 10 clinical oncologists at the second, with two years in between. Multivariable logistic regression was performed to test whether patient age and reduced mean dose (D mean ) to the residual brain (Brain – CTV – brainstem) could predict the probability of the oncologists’ choosing PT. P-values < 0.05 were considered significant.

Results Total agreement was reached by all oncologists for seven patients, where XT was selected for two of these patients and PT was selected for five patients. For three patients, only one or two oncologists disagreed with the choice of XT and for 17 patients, only one or two disagreed with the PT decision. For nine of the patients, no consensus was reached: half of the clinical oncologists chose XT and the other half chose PT. Age at RT and difference in D mean to the residual brain were both significant predictors of the outcome, p = 0.01 and p = 0.04 respectively (Figure 1). Model coefficients were: β 0 = 0.9219 (95% confidence interval [CI]; -1.1512 - 2.9949), β age = -0.0422 per year (95% CI; -0.0746 - -0.0099) and β Δ dose = 0.1591 per Gy (95% CI; 0.0036 - 0.3146).