Poster - Thurs Eve-07: The dosimetric consequences of MLC position inaccuracy in IMRT

The Multileaf Collimator (MLC), the most widely used means of intensity modulating beams for IMRT, is subject to random and systematic errors in leaf positions that may compromise the treatment quality. This work is a simulation study of the effect of random and systematic errors in leaf position on delivered dose distributions. The dosimetric effects of random errors of up to 2 mm and systematic errors (±1mm in 2 banks, ±0.5mm in 2 banks and 2mm in 1 bank of leaves) were analysed for a typical head and neck IMRT plan and a typical prostate IMRT plan. Dose Volume Histograms and Equivalent Uniform Doses (EUD) of the target volumes, bladder and rectum for the prostate plan and brainstem, larynx, parotids and spinal cord for the head and neck plan were calculated with and without MLC positioning errors and compared. The results show that if we adopt a 2% change in EUD of the target and 2 Gy for the OARs as acceptable levels of uncertainty in dose due to MLC effects only, then random errors of up to 2mm may be tolerated but systematic errors in leaf position will need to be limited to 0.5mm. Our study provides guidance, based on a surrogate of clinical outcome, for the development of quality control standards for multileaf collimators.