Assessment of Tumor Dose Distribution in the Presence of Tissue Heterogeneities for the task of Larynx Cancer Radiation Therapy: Phantom Study and Monte Carlo Simulation

Introduction: Precise measurement/calculation of the tumor absorbed dose in the presence of the tissue/background heterogeneities has been a significant challenge in external radiotherapy. This work set out to compare/investigate the absorbed dose distribution within the tumor by Monte Carlo (EGSnrc) simulation and Anisotropic Analytical Algorithm (AAA) (commercial software) in the presence of the tissue heterogeneities in larynx cancer therapy considering the phantom study and experimental dosimetry as reference.  To this end, a dedicated phantom was designed and built to conduct experimental dosimetry for larynx cancer radiation therapy.

Methods: A dedicated head and neck phantom was designed for larynx cancer dosimetry at Arvin Oncology Center.  This phantom contains a tumor-like component located adjacent to a pyramidal-shaped air duct (Figure 1).  This configuration enables us to conduct tumor dosimetry at different tumor-to-air volume ratios.  Moreover, this phantom allows for measurement of the tumor absorbed dose at different locations using pin-point ion chambers.  The commercial Analytical Anisotropic Algorithm (AAA) treatment planning system (implemented in Eclipse version 13.1, Varian Medical Systems) was employed to estimate the absorbed dose (dose distribution) within the tumor in the phantom.  Furthermore, a Monte Carlo simulation of the phantom and a 6 MeV photon beam of Varian IX Linac was carried out (based on the EGSnrc, BEAMnrc, and DOSXYZnrc functions).  The outcome of experimental tumor dosimetry via the pin-point ion chambers, the Monte Carlo simulation and the commercial software were investigated/compared.

Results: The maximum and minimum tumor volumes in this phantom were 17.59 cm³ and 9.48 cm^3, respectively, residing next to air ducts with volumes of 13.45 cm³ and 26.16 cm³.  The maximum discrepancy between the experimental dosimetry and the commercial software outcomes was 9% and occurred when tumors with smaller volumes were located next to large air ducts.  This indicates that reduced tumor volume (for instance, as a result of radiation therapy) would adversely affect the accuracy of the commercial dose calculation platform.  On the other hand, the lowest difference between the experimental measurement and commercial software of 7.5% was obtained when the tumor was located next to the smallest air duct.  The Monte Carlo simulation confirmed the results of experimental dosimetry with an average dose difference of 0.5%.  Moreover, there were no significant differences between the results of IMRT and 3D CRT radiation therapy frameworks.

Conclusions: This work introduces a novel phantom design for dosimetry in Larynx cancer radiation therapy in the presence of the tissue heterogeneities, wherein the commercial treatment planning system (AAA algorithm implemented in the Eclipse platform) exhibited significant overestimation of the tumor absorbed dose.  This issue warrants particular attention when Larynx radiation therapy is conducted under heterogeneous conditions.

• Download figure
• Open in new tab
• Download powerpoint