Patient-specific Monte Carlo-based dose-kernel approach for inverse planning in afterloading brachytherapy.

Département de Radio-Oncologie et Centre de Recherche en Cancérologie de l'Université Laval, Hôtel-Dieu de Québec, Québec, QC, Canada.
International journal of radiation oncology, biology, physics (Impact Factor: 4.59). 11/2010; 81(5):1582-9. DOI: 10.1016/j.ijrobp.2010.09.029
Source: PubMed

ABSTRACT Brachytherapy planning software relies on the Task Group report 43 dosimetry formalism. This formalism, based on a water approximation, neglects various heterogeneous materials present during treatment. Various studies have suggested that these heterogeneities should be taken into account to improve the treatment quality. The present study sought to demonstrate the feasibility of incorporating Monte Carlo (MC) dosimetry within an inverse planning algorithm to improve the dose conformity and increase the treatment quality.
The method was based on precalculated dose kernels in full patient geometries, representing the dose distribution of a brachytherapy source at a single dwell position using MC simulations and the Geant4 toolkit. These dose kernels are used by the inverse planning by simulated annealing tool to produce a fast MC-based plan. A test was performed for an interstitial brachytherapy breast treatment using two different high-dose-rate brachytherapy sources: the microSelectron iridium-192 source and the electronic brachytherapy source Axxent operating at 50 kVp.
A research version of the inverse planning by simulated annealing algorithm was combined with MC to provide a method to fully account for the heterogeneities in dose optimization, using the MC method. The effect of the water approximation was found to depend on photon energy, with greater dose attenuation for the lower energies of the Axxent source compared with iridium-192. For the latter, an underdosage of 5.1% for the dose received by 90% of the clinical target volume was found.
A new method to optimize afterloading brachytherapy plans that uses MC dosimetric information was developed. Including computed tomography-based information in MC dosimetry in the inverse planning process was shown to take into account the full range of scatter and heterogeneity conditions. This led to significant dose differences compared with the Task Group report 43 approach for the Axxent source.

1 Bookmark
  • [Show abstract] [Hide abstract]
    ABSTRACT: To establish the accuracy and speed of bGPUMCD, a GPU-oriented Monte Carlo code used for high dose rate brachytherapy dose calculations. The first objective is to evaluate the time required for dose calculation when full Monte Carlo generated dose distribution kernels are used for plan optimization. The second objective is to assess the accuracy and speed when recalculating pre-optimized plans, consisting of many dwell positions. bGPUMCD is tested with three clinical treatment plans : one prostate case, one breast case, and one rectum case with a shielded applicator. Reference distributions, generated with GEANT4, are used as a basis of comparison. Calculations of full dose distributions of pre-optimized treatment plans as well as single dwell dosimetry are performed. Single source dosimetry, based on TG-43 parameters reproduction, is also presented for the microSelectron V2 (Nucletron, Veenendaal, The Netherlands). In timing experiments, the computation of single dwell position dose kernels takes between 0.25 and 0.5 s. bGPUMCD can compute full dose distributions of previously optimized plans in ∼2 s. bGPUMCD is capable of computing pre-optimized brachytherapy plans within 1% for the prostate case and 2% for the breast and shielded applicator cases, when comparing the dosimetric parameters D90 and V100 of the reference (GEANT4) and bGPUMCD distributions. For all voxels within the target, an absolute average difference of approximately 1% is found for the prostate case, less than 2% for the breast case and less than 2% for the rectum case with shielded applicator. Larger point differences (>5%) are found within bony regions in the prostate case, where bGPUMCD underdoses compared to GEANT4. Single source dosimetry results are mostly within 2% for the radial function and within 1%-4% for the anisotropic function. bGPUMCD has the potential to allow for fast MC dose calculation in a clinical setting for all phases of HDR treatment planning, from dose kernel calculations for plan optimization to plan recalculation.
    Medical Physics 07/2012; 39(7):4559-67. · 2.91 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: To assess the dosimetric effects of the presence of the applicator, air pockets in clinical target volume (CTV) and OARs along with tissue heterogeneities using the Monte Carlo (MC) method in high dose rate (HDR) gynecologic interstitial brachytherapy with a Syed-Neblett template. The CT based dosimetry has been achieved with the Geant4 MC toolkit version 9.2. DICOM-RT files of 38 patients were imported into our own platform for MC simulations. The dose distributions were then compared to those obtained with a conventional TG-43 calculation. Taking account of heterogeneities has effects of the order of 1% on the HDR gynecological dose distributions. However, the exclusion of air pockets and applicator from the DVH calculation can lower the CTV D90 and V100 by as much as 8.7% and 5.0% in comparison with TG-43. Rectum dosimetric indices can also be lowered by approximately 3% compared with TG-43 for most cases. Differences for urethra and bladder are for most cases below 1%. Exclusion of non-biological material such as air pockets and applicator volume from the CTV is important for both TG-43 and MC calculations. It could be easily implemented and automated in treatment planning systems without affecting computation times.
    Radiotherapy and Oncology 10/2013; · 4.52 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: To present a new accelerated Monte Carlo code for CT-based dose calculations in high dose rate (HDR) brachytherapy. The new code (HDRMC) accounts for both tissue and nontissue heterogeneities (applicator and contrast medium). HDRMC uses a fast ray-tracing technique and detailed physics algorithms to transport photons through a 3D mesh of voxels representing the patient anatomy with applicator and contrast medium included. A precalculated phase space file for the(192)Ir source is used as source term. HDRM is calibrated to calculated absolute dose for real plans. A postprocessing technique is used to include the exact density and composition of nontissue heterogeneities in the 3D phantom. Dwell positions and angular orientations of the source are reconstructed using data from the treatment planning system (TPS). Structure contours are also imported from the TPS to recalculate dose-volume histograms. HDRMC was first benchmarked against the MCNP5 code for a single source in homogenous water and for a loaded gynecologic applicator in water. The accuracy of the voxel-based applicator model used in HDRMC was also verified by comparing 3D dose distributions and dose-volume parameters obtained using 1-mm(3) versus 2-mm(3) phantom resolutions. HDRMC can calculate the 3D dose distribution for a typical HDR cervix case with 2-mm resolution in 5 min on a single CPU. Examples of heterogeneity effects for two clinical cases (cervix and esophagus) were demonstrated using HDRMC. The neglect of tissue heterogeneity for the esophageal case leads to the overestimate of CTV D90, CTV D100, and spinal cord maximum dose by 3.2%, 3.9%, and 3.6%, respectively. A fast Monte Carlo code for CT-based dose calculations which does not require a prebuilt applicator model is developed for those HDR brachytherapy treatments that use CT-compatible applicators. Tissue and nontissue heterogeneities should be taken into account in modern HDR brachytherapy planning.
    Medical Physics 05/2014; 41(5):051712. · 2.91 Impact Factor