Simulations of hypoxic HNSCC with clinically trialled and new radiation fractionation using the HYP-RT Monte Carlo Model

Full text: The model uses MC techniques to divide individual cells to grow and treat a virtual tumour, including the effects of tumour hypoxia, accelerated repopulation (AR) and reoxygenation (Rox). Simulations have been performed to compare modelled cell kill with clinical data, and to rank schedules based on biological effective doses (BED) relating to normal tissues. New fractionation schedules have also been simulated. Epithelial cell hierarchy has been modelled to simulate the division of stem, transit and differentiated cells. Oxygenation altered cell survival, with the dose/fraction additionally altering the implemented oxygen enhancement ratio curves. Fifteen altered schedules were simulated, including 10 from clinical trials, I with a 2 week break, and 4 new schedules. The trend of requiring 15% more dose for hypoxic versus oxic tumours for the 5 x 2 Gy/wk schedule continued for other schedules. Averaged results follow clinical data well, but demonstrate the large variation in doses required for specific AR and ROx onset times (Fig. I). The combination of ROx onset at 0 weeks and AR onset at 2 weeks appeared to be the most likely combination. The BED ranking analysis showed that schedules 1--4 were acceptable for early normal tissue effects, however to minimise late effects the hyperfractionated schedules 2, 3, 5 and 9 were optimal. Conclusions An efficient model, simulating hypoxic head and neck cancer has been extended to model alternate radiotherapy schedules. Results outline the large range of doses required to achieve individualised tumour control for tumours with specific oxygenation, AR and ROx properties.