Wouter Crijns1, Hans Van Herck2, Gilles Defraene3, Laura Van den Bergh3, Pieter Slagmolen4, Karin Haustermans3, Frederik Maes5, Frank Van den Heuvel6. 1. Department of Oncology, Laboratory of Experimental Radiotherapy, KU Leuven, Herestraat 49, 3000 Leuven, Belgium and Medical Imaging Research Center, KU Leuven, Herestraat 49, 3000 Leuven, Belgium. 2. Medical Imaging Research Center, KU Leuven, Herestraat 49, 3000 Leuven, Belgium and Department of Electrical Engineering (ESAT) - PSI, Center for the Processing of Speech and Images, KU Leuven, 3000 Leuven, Belgium. 3. Department of Oncology, Laboratory of Experimental Radiotherapy, KU Leuven, Herestraat 49, 3000 Leuven, Belgium. 4. Medical Imaging Research Center, KU Leuven, Herestraat 49, 3000 Leuven, Belgium; Department of Electrical Engineering (ESAT) - PSI, Center for the Processing of Speech and Images, KU Leuven, 3000 Leuven, Belgium; and iMinds-KU Leuven Medical IT Department, KU Leuven, 3000 Leuven, Belgium. 5. Medical Imaging Research Center, KU Leuven, Herestraat 49, 3000 Leuven, Belgium; and Department of Electrical Engineering (ESAT) - PSI, Center for the Processing of Speech and Images, KU Leuven & iMinds, 3000 Leuven, Belgium. 6. Department of Oncology, Laboratory of Experimental Radiotherapy, KU Leuven, Herestraat 49, 3000 Leuven, Belgium and Department of Oncology, MRC-CR-UK Gray Institute of Radiation Oncology and Biology, University of Oxford, Oxford OX1 2JD, United Kingdom.
Abstract
PURPOSE: Intensity modulated radiotherapy (IMRT) and volumetric modulated arc therapy have become standard treatments but are more sensitive to anatomical variations than 3D conformal techniques. To correct for inter- and intrafraction anatomical variations, fast and easy to implement methods are needed. Here, the authors propose a full dosimetric IMRT correction that finds a compromise in-between basic repositioning (the current clinical practice) and full replanning. It simplifies replanning by avoiding a recontouring step and a full dose calculation. It surpasses repositioning by updating the preoptimized fluence and monitor units (MU) using a limited number of fiducial points and a pretreatment (CB)CT. To adapt the fluence the fiducial points were projected in the beam's eye view (BEV). To adapt the MUs, point dose calculation towards the same fiducial points were performed. The proposed method is intrinsically fast and robust, and simple to understand for operators, because of the use of only four fiducial points and the beam data based point dose calculations. METHODS: To perform our dosimetric adaptation, two fluence corrections in the BEV are combined with two MU correction steps along the beam's path. (1) A transformation of the fluence map such that it is realigned with the current target geometry. (2) A correction for an unintended scaling of the penumbra margin when the treatment beams scale to the current target size. (3) A correction for the target depth relative to the body contour and (4) a correction for the target distance to the source. The impact of the correction strategy and its individual components was evaluated by simulations on a virtual prostate phantom. This heterogeneous reference phantom was systematically subjected to population based prostate transformations to simulate interfraction variations. Additionally, a patient example illustrated the clinical practice. The correction strategy was evaluated using both dosimetric (CTV mean dose, conformity index) and clinical (tumor control probability, and normal tissue complication probability) measures. RESULTS: Based on the current experiments, the intended target dose and tumor control probability could be assured by the proposed method (TCP ≥ TCP(intended)). Additionally, the conformity index error was more than halved compared to the current clinical practice (ΔCI(95%) from 40% to 16%) resulting in improved organ at risk protection. All the individual correction steps had an added value to the full correction. CONCLUSIONS: A limited number of fiducial points (no organ contours required) and an in-room (CB)CT are sufficient to perform a full dosimetric correction for IMRT plans. In the presence of interfraction variation, the corrected plans show superior dose distributions compared to our current clinical practice.
PURPOSE: Intensity modulated radiotherapy (IMRT) and volumetric modulated arc therapy have become standard treatments but are more sensitive to anatomical variations than 3D conformal techniques. To correct for inter- and intrafraction anatomical variations, fast and easy to implement methods are needed. Here, the authors propose a full dosimetric IMRT correction that finds a compromise in-between basic repositioning (the current clinical practice) and full replanning. It simplifies replanning by avoiding a recontouring step and a full dose calculation. It surpasses repositioning by updating the preoptimized fluence and monitor units (MU) using a limited number of fiducial points and a pretreatment (CB)CT. To adapt the fluence the fiducial points were projected in the beam's eye view (BEV). To adapt the MUs, point dose calculation towards the same fiducial points were performed. The proposed method is intrinsically fast and robust, and simple to understand for operators, because of the use of only four fiducial points and the beam data based point dose calculations. METHODS: To perform our dosimetric adaptation, two fluence corrections in the BEV are combined with two MU correction steps along the beam's path. (1) A transformation of the fluence map such that it is realigned with the current target geometry. (2) A correction for an unintended scaling of the penumbra margin when the treatment beams scale to the current target size. (3) A correction for the target depth relative to the body contour and (4) a correction for the target distance to the source. The impact of the correction strategy and its individual components was evaluated by simulations on a virtual prostate phantom. This heterogeneous reference phantom was systematically subjected to population based prostate transformations to simulate interfraction variations. Additionally, a patient example illustrated the clinical practice. The correction strategy was evaluated using both dosimetric (CTV mean dose, conformity index) and clinical (tumor control probability, and normal tissue complication probability) measures. RESULTS: Based on the current experiments, the intended target dose and tumor control probability could be assured by the proposed method (TCP ≥ TCP(intended)). Additionally, the conformity index error was more than halved compared to the current clinical practice (ΔCI(95%) from 40% to 16%) resulting in improved organ at risk protection. All the individual correction steps had an added value to the full correction. CONCLUSIONS: A limited number of fiducial points (no organ contours required) and an in-room (CB)CT are sufficient to perform a full dosimetric correction for IMRT plans. In the presence of interfraction variation, the corrected plans show superior dose distributions compared to our current clinical practice.
Authors: Ilana Feain; Chun-Chien Shieh; Paul White; Ricky O'Brien; Sandra Fisher; William Counter; Peter Lazarakis; David Stewart; Simon Downes; Michael Jackson; Siddhartha Baxi; Brendan Whelan; Kuldeep Makhija; Chen-Yu Huang; Michael Barton; Paul Keall Journal: Adv Radiat Oncol Date: 2016-11-08