Feasibility of Collapsed Cone Convolution Dose Calculation Algorithm for Nanoparticle-Enhanced Radiotherapy: A Comparison with GEANT4 Monte Carlo Simulations
Abstract
Purpose
This study investigates the limitations of Collapsed Cone Convolution (CCC) dose calculation algorithm in accurately modeling dose perturbations in nanoparticle-enhanced radiotherapy for lung tumors. The feasibility of CCC is compared with a Monte Carlo (MC) simulation platform using GEANT4, which provides a more accurate representation of radiation transport and particle interactions in heterogeneous tissues.
Methods
The RayStation™-CCC algorithm was initially evaluated at the dose point kernel (DPK) level for lung tumors treated with HfO₂ nanoparticles. This analysis assessed CCC’s ability to account for complex tissue heterogeneities and high-Z nanoparticle–radiation interactions that influence local dose deposition. An in-house MC simulation framework using GEANT4 was then used to model photon, electron, and secondary particle transport in HfO₂-loaded tissue geometries. Dose distributions derived from GEANT4 simulations were compared with CCC calculations to quantify discrepancies and assess the clinical feasibility of CCC for nanoparticle-enhanced radiotherapy.
Results
Density-scaled DPKs in homogeneous water and HfO₂-loaded tissues showed discrepancies up to 20% at high nanoparticle concentrations (6 wt%). Simulations of layered tissue–HfO₂ (1–6 wt%) phantoms revealed pronounced dose buildup at material interfaces. CCC algorithm failed to reproduce this interface dose enhancement, arising from increased photoelectron production in high-Z regions, with discrepancies exceeding 50% in the first 2 cm of the interface. On the other hand, GEANT4-MC simulations, which track individual photons, electrons, and secondary particle interactions, provided more accurate characterization of dose enhancement in high-Z regions.
Conclusion
This study demonstrates the dose calculation limitations of DPK-based algorithms for nanoparticle-enhanced radiotherapy. Preliminary results indicate that GEANT4-MC simulations offer a more precise representation of dose enhancement, especially at high nanoparticle concentrations. Evaluating CCC against GEANT4 highlights the feasibility and limitations of CCC for clinical treatment planning in high-Z nanoparticle therapies.