While the biological effective dose (BED) has been used to estimate the damage to tumor cells in radiotherapy, BED does not consider intrafractional interruption (IFI) occurring during irradiation. We aim to develop a framework to evaluate the decrease in BED [ΔBED] and to create a plan compensating for the decrease by IFI. Approach. ΔBED was calculated using a model based on the microdosimetric kinetic model (MKM) for four brain tumor cases treated using a volumetric-modulated arc therapy. Four biologically compensated plans (BCPs) were created in the treatment planning system by a single-time optimization using a base plan considering ΔBED created in in-house software and optimization objectives for the original clinically delivered plan to achieve a homogeneous BED distribution within the planning target volume (PTV). The BED-volume histogram was evaluated for non-compensated plan and BCP with different timepoint of interruption, a percentage of gantry rotation angle (GRA) before interruption in planned GRA,ηand duration of interruptionτ. Characteristics of the dose accumulation were analyzed for different collimator angle sets, Plan A (10°, 85°) and Plan B (45° and 315°), for the first case. Main Results. Hot spots in the ΔBED distribution forη= 25%, 50%, and 75% were observed at superior-and-interior ends, central region, and peripheral region in PTV, respectively. These behaviors could be understood by characteristics of the MKM-based model producing maximum ΔBED at 50% of the dose accumulation. ΔBED50%ranged 4.5-6.6%, 5.0-7.3%, and 5.3-7.7% forτ= 60, 90, and 120 mins, respectively. Plan A showed fast dose accumulation at superior and inferior edges while slow on peripheries in the lateral dose profile. Plan B showed more homogeneous PD distributions than Plan A during irradiation. Significance. The developed framework successfully evaluated and compensated for the decreased BED distribution. .
Keywords: biological adaptive radiotherapy; biological compensation; treatment interruption.
© 2024 IOP Publishing Ltd. All rights, including for text and data mining, AI training, and similar technologies, are reserved.