CT-guided intensity-modulated radiotherapy for bladder cancer: Isocentre shifts, margins and their impact on target dose

Background and purpose: Image-guided radiotherapy (IGRT) has great potential to improve the treatment of tumour sites that exhibit large geometrical uncertainties such as the bladder. This study quantifies the size and direction of the daily 3D isocentre shift resulting from the use of IGRT and its impact on reducing the margins required. The changes to target dose coverage, delivered dose and dose homogeneity appearing when used in conjunction with intensity-modulated radiotherapy (IMRT) are also assessed. Materials and methods: The study is performed using a series of 19 bladder cancer patients that underwent weekly repeat conventional CT scanning during their 6-7 weeks treatment course. The isotropic margin required to cover various percentages of the volumes of the repeat scans CTVs is found by growing the planning scan PTV by incrementing the margin in steps of 1 mm until full coverage is obtained. The optimum daily isocentre shift is determined by moving the repeat scan CTV in 3D so that its volume that ties outside the planning CTV is minimised. Reduction in the isotropic margin after applying the optimum isocentre shift is found by repeating the above procedure. Individual optimum sizes of the six margins required in the sup/inf/ant/post/right/left directions are determined using a recently published empirical margin determination method. Finally, five-field IMRT plans are set up (that minimise the dose deviation throughout the target) using different planning CTV margin alternatives, the optimal isocentre shifts are applied and the target dose distributions are assessed after recalculation without changing any of the beam parameters. Results: Without use of IGRT (i.e., without shifting to the optimal isocentre), an isotropic margin of 30 mm is required to cover fully all the CTV volumes on the repeat scans for all patients. The direction of the optimum isocentre shift required is random with an average directional shift less than 1 mm and is unlikely to exceed an absolute value of 15 mm (average of 7 mm). Applying the optimum isocentre shift reduces the isotropic margin to 16 mm for full volume coverage. Determining the optimum margin individually in each of six orthogonal directions reduces the target volume by approximately 30% but requires complex daily planning. Applying the optimum isocentre shift to IMRT plans shows little change to the overall mean target dose of 100% (an average increase of 1%), but produces a spread in the daily mean dose ranging from 96% to 106%. The 3D dose variation over the target is within the 95-107% acceptable range in 72% of cases for a 12 mm uniform margin, which increases to 91% if any deviation from the daily prescribed dose is removed. The minimum and maximum doses within the target can show significant changes. Conclusion: The use of IGRT in the treatment of bladder cancer leads to a marked reduction in the margins required. When used in conjunction with IMRT, the pre-treatment plan is shown to be acceptable for daily treatment (after shifting the isocentre) in terms of the resulting dose distribution provided the correct daily mean dose is delivered. That can be achieved by rescaling the daily monitor units for each treatment beam. (c) 2006 Elsevier Ireland Ltd. All rights reserved.