ESTRO 2021 Abstract Book

S1566

ESTRO 2021

PO-1837 Biological optimization in proton therapy accounting for hypoxia and variable RBE H. Henjum 1 , T. Johnsen Dahle 1 , C. Stokkevåg 2,1 , C. Grindeland 3 , K. Røe Redalen 4 , H. Minn 5,6 , E. Malinen 7 , S. Pilskog 1,2 , A. Mairani 8,9 , K. Smeland Ytre-Hauge 1 1 University of Bergen, Department of Physics and Technology, Bergen, Norway; 2 Haukeland University Hospital, Department of Oncology and Medical Physics, Bergen, Norway; 3 Haukeland University Hopsital, Department of Oncology and Medical Physics, Bergen, Oman; 4 Norwegian University of Science and Technology, Department of Physics, Trondheim, Norway; 5 Turku University Hospital, Department of Oncology and Radiotherapy, Turku, Finland; 6 University of Turku, Turku PET Centre, Turku, Finland; 7 University of Oslo, Department of Physics, Oslo, Norway; 8 Centro Nazionale di Adroterapia Oncologica , Department of Medical Physics, Pavia, Italy; 9 Universität Heidelberg, Heidelberg Ion Beam Therapy Center, Heidelberg, Germany Purpose or Objective Tumors with hypoxic regions are associated with poor treatment outcome. Their increased radioresistance as compared to well-oxygenated regions is quantified by the oxygen enhancement ratio (OER). Including the OER, as well as the relative biological effectiveness (RBE), for radiobiological and OER weighted dose (ROWD) optimization in proton therapy could therefore increase the local tumor control. The aim of this study was to implement and explore ROWD guided by differences in partial oxygen pressure (pO 2 ). Materials and Methods An ROWD optimization method was implemented in a FLUKA Monte Carlo based treatment planning software. The method is based on the linear quadratic model, with proton α and β parameters as a function of the OER, and therefore a function of the linear energy transfer and pO 2 . The method was applied for optimization of simple spread-out Bragg peak (SOBP) scenarios in water with varying pO 2 as well as used for optimization of a head and neck cancer (HNC) patient plan. Tumor pO 2 levels was estimated from [ 18 F]-EF5 PET images, and with a prescription dose of 70 Gy(RBE) and three different radiosensitivity parameters, (α/β) x ROWD was used and compared for four models: an RBE of 1.1 (RBE 1.1,OER ), two variable RBE models, Rørvik (ROR) and McNamara (MCN) and a reference plan neglecting OER (RBE 1.1 ). Results ROWD for the hypoxic regions in the SOBP resulted in higher LET d and physical dose compared to RBE 1.1 (Figure 1). For the lowest pO 2 levels, the maximum physical dose and LET d was seen for RBE 1.1,OER and corresponded to a factor of 1.44 greater than RBE 1.1 , whereas ROR and MCN resulted in a factor of 1.3 and 1.2, respectively for (α/β) x of 2 Gy, and 1.4 for (α/β) x of 10 Gy. The LET d for the RBE 1.1,OER provided the highest LET d values with a peak of 10 keV/µm . In the HNC plan, the physical dose was increased to the hypoxic regions for all models, as shown by the DVHs in Figure 2, but unlike the phantom case no increase in LET d was obtained. Figure 2 further shows the large variations in dose that was required by the ROWD with the RBE 1.1,OER model to counteract hypoxia in comparison to the RBE 1.1 plan. ROWD for the MCN model resulted in a mean physical dose increase to the PTV of 4.9 Gy for (α/β) x values ranging between 5 – 15 Gy, while the ROR strategies increased with 4.6 Gy. The RBE 1.1 (OER) plan provided a mean physical dose of 71.4 Gy, and in the lower region of the variable RBE models.