Development of an optimum photon beam model for head and neck IMRT

Abstract

Introduction: Intensity modulated radiotherapy for complex sites such as head and neck tumours requires a level of accuracy in the dose calculation beyond that currently used for conformal treatment planning. Recent advances in treatment planning systems have aimed to improve the dose calculation accuracy by improved modelling of machine characteristics such as inter-leaf leakage, tongue and groove, and rounded MLC leaf ends. What is uncertain is the extent to which these model parameters improve the agreement between the dose calculation and measurements for IMRT treatments.  Methods: Optimisation of additional photon beam model parameters was carried out using Pinnacle version 7.4f for both an Elekta Precise and a Varian linac. One additional parameter was added to the beam models in turn and a series of models commissioned to investigate the dosimetric impact of each model parameter on five clinical H&N IMRT plans. The magnitude and location of differences between the models was determined from gamma analysis of the calculated planar dose maps. A final model was commissioned that incorporated all of the changes. For the Elekta Precise, the impact of all the changes was determined using a gamma analysis comparison with measured films.  Results: Cumulative differences of >2%/2mm were observed in up to 5% of pixels when comparing dose distributions with and without all of the model changes. Individually, the addition of modelling the rounded MLC leaf ends caused the most dramatic change to the calculation of the dose distribution for both linacs, generating a difference of 1%/1mm in 10-25% of pixels (up to 5% of pixels to 3%/3mm) for the 5 patient plans sampled. The impact of the tongue-and-groove effect was more significant for the Varian (mean of 25% of pixels at 1%/1mm, compared to 5% of pixels with the Elekta). The combined changes to the Elekta model were found to improve agreement with measurement. Conclusions: Current commercially available treatment planning systems offer sufficient accuracy for clinical implementation of H&N IMRT. For this treatment site the ability to accurately model the rounded MLC leaf ends has the greatest impact on the similarity of the calculated dose distribution to measurements. In addition, for the Varian linac the modelling of the tongue & groove effect was also advantageous.