Modeling tumour control for prostate treatments can be challenging. Models used to predict tumour control are typically based on the well known linear-quadratic (LQ) model for cell survival. The LQ model accurately predicts cell killing for a given radiation dose in most cases, however in some cases the LQ model falls short and LQ model predictions differ from what has been observed experimentally. Cell survival following low radiation doses <1Gy and radiation induced bystander effects are two examples where the LQ model potentially falls short at modelling cell survival following an absorbed radiation dose.
Using mathematical models that account for radiation induced bystander and low dose hypersensitivity effects, we look at how these effects potentially change predicted outcomes of tumour control when compared to standard LQ predictions. We start by investigating how the basic LQ model predicts tumour control under large geometric miss errors under intensity modulated radiation therapy (IMRT) and volume modulated arc therapy (VMAT) modalities. Next, we look at the geometric miss scenario again, but rather than compare two different delivery techniques we investigate how radiation induced bystander effects potentially change predicted tumour control under geometric miss cases. We then expand on this idea and study how radiation induced bystander effects potentially influence predicted outcome under heterogeneous dose distributions. Finally, we compare the relative biological effectiveness (RBE) of bystander effects with low dose hypersensitivity effects and spectral effects for out-of-field radiation doses.
The majority of this work focused on how radiation induced bystander effect changed biological modeling of tumour control in external beam radiotherapy for prostate cancer. This work ultimately demonstrates that bystander effects can modify predicted outcomes of tumour control. We have shown that bystander effects potentially improve TCP under geometric miss and that bystander effects may relax the common planning constraint of aiming for dose homogeneity within a target volume.