Testing MBDCA performance in 192 Ir HDR brachytherapy for lip carcinoma

Abstract

Introduction Model-based dose calculation algorithms (MBDCAs) for 192 Ir HDR brachytherapy have become clinically available to account for effects disregarded by TG-43 dosimetry. International committees recommend the preparation of clinical test cases for their benchmarking. Purpose To evaluate the dosimetric accuracy of a MBDCA in clinical scatter conditions departing from TG-43 assumptions (homogeneity, isotropic scatter). Materials and methods Treatment planning dosimetry was performed for three clinical cases using the Advanced Collapsed Cone Engine (ACE) of the OncentraBrachy v4.5 treatment planning system. Corresponding Monte Carlo (MC) simulation reference data were obtained using MCNP6 with input files prepared from parsing dicom RT data with the BrachyGuide software tool. ACE and MC were compared in terms of percentage dose differences, Dose-Volume Histograms (DVHs) and indices of clinical interest for the Planning Target Volume (PTV) and Organs-At-Risk. Results ACE and MC-based DVH indices were in excellent agreement for the PTV except for an ACE underestimation of the high dose volume (5% for V150 and V200). While an excellent agreement was also observed in the high dose regions of the mandible, ACE overestimated D30, D50 and Dmean by up to 6.3%. ACE and MC results agreed within 0.6% for the skin and normal tissue surrounding the PTV. Conclusion ACE improves dosimetric accuracy relative to TG-43. The clinical importance of observed discrepancies from MC should be evaluated taking radiobiological analysis into account. A corresponding test case is available online ( www.rdl.gr ). Disclosure VP is supported by IKY fellowship of excellence for postgraduate studies in Greece-Siemens Program.