Existing radiotherapy treatment planning protocols treat a cancer as a fixed entity within a "target volume". Efforts have been directed towards faster or more accurate planning processes and more accurate or reliable delivery methods. We propose a treatment planning system for a radiotherapy apparatus comprising a computational processor applying a set of instructions to an input data set, the input data set comprising a three-dimensional dose distribution which is to be achieved by the radiotherapy apparatus within a volume to be irradiated, a three-dimensional volume image characterising the tissue types within the volume to be irradiated, and a set of apparatus parameters which characterise the radiotherapy apparatus, with the set of instructions including a computational process arranged to output a treatment plan apt to deliver the three-dimensional dose distribution when implemented by the radiotherapy apparatus, by optimising a function that includes at least one term representing a time-dependent response of the tissue types to an applied radiation. The set of instructions can then include a fractional distribution step arranged to assign fractional parts of the dose distribution to successive treatment fractions, taking into account time- and dose-dependent characteristics of the tissue types within the volume image, and the iterative computational process can then be arranged, for each non-identical fraction, to take into account the fractional part of the dose distribution and the set of apparatus parameters, to derive a set of treatment instructions apt to deliver the fractional part when carried out by the radiotherapy apparatus. The three-dimensional volume image will ideally characterise the tissue types according to at least their local chemical composition, to assist in modelling the dose response. The optimisation function can include terms representing time-dependent responses such as a cell repair model and a tumour diffusion model.La présente i
