ESTRO 2024 - Abstract Book

S3510

Physics - Dose prediction, optimisation and applications of photon and electron planning

ESTRO 2024

In Fig.2, we increased SD(α) from 3-50% and SD(d p ) from 1-20% and determined the accuracy of both recipes for two hypo and one regular fractionation scheme: a) n f =1 SD(β)=0, b) n f =1 SD(β)=SD(α), and c) d f =2Gy SD(β)=SD(α). For hypo without SD(β), both recipes predicted the simulated outcomes within 3%. Adding SD(β), yielded deviations up to 18% for Eq.1 for those cases where SD(α ) >> SD(d p ), whereas Eq.4 remained within 3%. The same trend could be seen for the fractionated treatments with fixed d f =2Gy.

Conclusion:

This study attempts to shed new lights on the effect of inter-patient radio-sensitivity variation on dose-effect studies and its resulting dose constraints. Quantitative data on SD(α) are scarce but our results indicate that SD(α) is likely <40%. Otherwise SD(d p ) becomes irrelevant and there is no dose-effect relation anymore. If SD(α)>20%, the simple d true -recipe containing only SD(D) (Eq.1) can underestimate the OAR safety margins by >10% and it is advisable to use the more complex recipe also including SD(α) (Eq.4) in these cases.