The microdosimetry of low-energy photons in radiotherapy

THE APPLICATION OF NEURAL NETWORK TECHNOLOGY BASED ON MEA-BP ALGORITHM IN THE PREDICTION OF MICRODOSIMETRIC QUALITIES

The most abundant products of the interaction between radiation and matter are low-energy electrons, and the collisions between these electrons and biomolecules are the main initial source of radiation-based biological damage. To facilitate the rapid and accurate quantification of low-energy electrons (0.1-10 keV) in liquid water at different site diameters (1-2000 nm), this study obtained ${\overline{y}}_{\mathrm{F}}$ and ${\overline{y}}_{\mathrm{D}}$data for low-energy electrons under these conditions. This paper proposes a back-propagation (BP) neural network optimized by the mind evolutionary algorithm (MEA) to construct a prediction model and evaluate the corresponding prediction effect. The results show that the ${\overline{y}}_{\mathrm{F}}$ and ${\overline{y}}_{\mathrm{D}}$ values predicted by the MEA-BP neural network algorithm reach a training precision on the order of ${10}^{-8}$. The relative error range between the prediction results of the validated model and the Monte Carlo calculation results is 0.03-5.98% (the error range for single-energy electrons is 0.1-5.98%, and that for spectral distribution electrons is 0.03-4.4%).

A Monte Carlo investigation of some important radiation parameters and tissue equivalency for photons below 1 keV in human tissues

The attenuation coefficients are important input values in estimating not only the dose and exposure in radiotherapy and medical imaging, but also in the proper design of photon shields. While studies are widely available above 1 keV, the attenuation coefficients of human tissues for photon energies less than 1 keV have not been studied yet. In this study, the attenuation coefficients of water and some human tissues were estimated for low energy photons using the MCNP6.1 code in the energy region 0.1 keV-1 keV. Mass attenuation coefficients were estimated at photon energies of 100, 150, 200, 250, 300, 350, 400, 450, 500, 550, 600, 650, 700, 750, 800, 850, 900, 950 and 1000 eV for water and ten human tissues (Soft, Breast, Lung, Bone, Brain, Eye lens, Ovary, Skin, Thyroid and Prostate). Results were compared with those available in literature and a fairly good agreement has been obtained. These data were then used to calculate the mean free path, half value layer, tenth value layer, effective atomic number and specific gamma-ray constant (useful for calculation of dose rate) as well. Moreover, for comparison the effective atomic number of the water has been obtained using the results of this work and using the data available in NIST database from 0.1 to 1 keV. In addition, the human tissues were compared with some tissue equivalent materials in terms of effective atomic number and specific gamma-ray constant to study the tissue equivalency from the results, the muscle-equivalent liquid with sucrose has been found to be the best tissue equivalent material for soft tissue, eye lens and brain with relative difference below 4.1%.

Microdosimetric investigation of the radiation quality of low-medium energy electrons using Geant4-DNA

The increasing clinical use of low-energy photon and electron sources (below few tens of keV) has raised concerns on the adequacy of the existing approximation of an energy-independent radiobiological effectiveness. In this work, the variation of the quality factor (Q) and relative biological effectiveness (RBE) of electrons over the low-medium energy range (0.1 keV-1 MeV) is examined using several microdosimetry-based Monte Carlo methodologies with input data obtained from Geant4-DNA track-structure simulations. The sensitivity of the results to the different methodologies, Geant4-DNA physics models, and target sizes is examined. Calculations of Q and RBE are based on the ICRU Report 40 recommendations, the Kellerer-Hahn approximation, the site version of the theory of dual radiation action (TDRA), the microdosimetric kinetic model (MKM) of cell survival, and the calculated yield of DNA double strand breaks (DSB). The stochastic energy deposition spectra needed as input in the above approaches have been calculated for nanometer spherical volumes using the different electron physics models of Geant4-DNA. Results are normalized at 100 keV electrons which is here considered the reference radiation. It is shown that in the energy range ~50 keV-1 MeV, the calculated Q and RBE are approximately unity (to within 1-2%) irrespective of the methodology, Geant4-DNA physics model, and target size. At lower energies, Q and RBE become energy-dependent reaching a maximum value of ~1.5-2.5 between ~200 and 700 eV. The detailed variation of Q and RBE at low energies depends mostly upon the adopted methodology and target size, and less so upon the Geant4-DNA physics model. Overall, the DSB yield predicts the highest RBE values (with RBE_max≈2.5) whereas the MKM the lowest RBE values (with RBE_max≈1.5). The ICRU Report 40, Kellerer-Hahn, and TDRA methods are in excellent agreement (to within 1-2%) over the whole energy range predicting a Q_max≈2. In conclusion, the approximation Q=RBE=1 was found to be valid only above ~50 keV whereas at lower energies both Q and RBE become strongly energy-dependent. It is envisioned that the present work will contribute towards establishing robust methodologies to determine theoretically the energy-dependence of radiation quality of individual electrons which may then be used in subsequent calculations involving practical electron and photon radiation sources.

MONTE CARLO-BASED INVESTIGATION OF MICRODOSIMETRIC DISTRIBUTION OF HIGH ENERGY BRACHYTHERAPY SOURCES

FLUKA-based Monte Carlo calculations were carried out to study microdosimetric distributions in air and in water for encapsulated high energy brachytherapy sources (60Co, 137Cs, 192Ir and 169Yb) by simulating a Tissue Equivalent Proportional Counter (Model LET1/2) having sensitive diameter of 1. 27 cm for a site size of 1 μm. The study also included microdosimetric distributions of bare sources. When the sources are in air, for a given source, the source geometry does not affect the y¯F and y¯D values significantly. When the encapsulated 192Ir, 137Cs and 60Co sources are in water, y¯F and y¯D values increase with distance in water which is due to degradation in the energy of photons. Using the calculated values of y¯D, relative biological effectiveness (RBE) was obtained for the investigated sources. When 60Co, 137Cs and 192Ir sources are in water, RBE increases from 1.03 ± 0.01 to 1.17 ± 0.01, 1.24 ± 0.01 to 1.46 ± 0.02 and 1.50 ± 0.01 to 1.75 ± 0.03, respectively, when the distance was increased from 3-15 cm, whereas for 169Yb, RBE is about 2, independent of distance in water.

Biological effects of low-dose-rate irradiation of pancreatic carcinoma cells in vitro using 125I seeds

AIM: To determine the mechanism of the radiation-induced biological effects of ¹²⁵I seeds on pancreatic carcinoma cells in vitro.

METHODS: SW1990 and PANC-1 pancreatic cancer cell lines were cultured in DMEM in a suitable environment. Gray’s model of iodine-125 (¹²⁵I) seed irradiation was used. In vitro, exponential phase SW1990, and PANC-1 cells were exposed to 0, 2, 4, 6, and 8 Gy using ¹²⁵I radioactive seeds, with an initial dose rate of 12.13 cGy/h. A clonogenic survival experiment was performed to observe the ability of the cells to maintain their clonogenic capacity and to form colonies. Cell-cycle and apoptosis analyses were conducted to detect the apoptosis percentage in the SW1990 and PANC-1 cells. DNA synthesis was measured via a tritiated thymidine (³H-TdR) incorporation experiment. After continuous low-dose-rate irradiation with ¹²⁵I radioactive seeds, the survival fractions at 2 Gy (SF2), percentage apoptosis, and cell cycle phases of the SW1990 and PANC-1 pancreatic cancer cell lines were calculated and compared.

RESULTS: The survival fractions of the PANC-1 and SW1990 cells irradiated with ¹²⁵I seeds decreased exponentially as the dose increased. No significant difference in SF2 was observed between SW1990 and PANC-1 cells (0.766 ± 0.063 vs 0.729 ± 0.045, P < 0.05). The ¹²⁵I seeds induced a higher percentage of apoptosis than that observed in the control in both the SW1990 and PANC-1 cells. The rate of apoptosis increased with increasing radiation dosage. The percentage of apoptosis was slightly higher in the SW1990 cells than in the PANC-1 cells. Dose-dependent G2/M cell-cycle arrest was observed after ¹²⁵I seed irradiation, with a peak value at 6 Gy. As the dose increased, the percentage of G2/M cell cycle arrest increased in both cell lines, whereas the rate of DNA incorporation decreased. In the ³H-TdR incorporation experiment, the dosimetry results of both the SW1990 and PANC-1 cells decreased as the radiation dose increased, with a minimum at 6 Gy. There were no significant differences in the dosimetry results of the two cell lines when they were exposed to the same dose of radiation.

CONCLUSION: The pancreatic cancer cell-killing effects induced by ¹²⁵I radioactive seeds mainly occurred via apoptosis and G2/M cell cycle arrest.

Calculation of relative biological effectiveness of a low-energy electronic brachytherapy source

Low-energy x-rays are known to have a higher relative biological effectiveness (RBE) than higher energy photons such as the gamma rays from 192Ir and 60Co. In this work the initial yield of single- and double-strand DNA breaks (SSB and DSB) and the RBE was estimated for a novel electronic brachytherapy source (EBS), emitting 40-50 kVp photons. An EGSnrc Monte Carlo model of the source was used in combination with the 'Monte Carlo damage simulation' program (Semenenko and Stewart 2004 Radiat. Res. 161 451-57; 2006 Phys. Med. Biol. 51 1693-706). The results indicate a substantially reduced SSB yield and increased DSB yield for the EBS compared to 60Co or 192Ir, leading to an enhanced RBE by 40-50%. The RBE estimate for the low-energy x-ray EBS was found to be very similar to the low-energy gamma ray brachytherapy isotope 125I. Biological damage was estimated in several human tissues: muscle, breast, calcified breast and cortical bone. SSB and DSB yields were similar in all media, except in bone. These findings should be taken into account if the EBS is intended to replace brachytherapy with the commonly used 192Ir isotope.

Spectroscopic characterization of a novel electronic brachytherapy system

The Axxent developed by Xoft Inc. is a novel electronic brachytherapy system capable of generating x-rays up to 50 keV. These low energy photon-emitting sources merit attention not only because of their ability to vary the dosimetric properties of the radiation, but also because of the radiobiological effects of low energy x-rays. The objective of this study is to characterize the x-ray source and to model it using the Geant4 Monte Carlo code. Spectral and attenuation curve measurements are performed at various peak voltages and angles and the source is characterized in terms of spectrum and half-value layers (HVLs). Also, the effects of source variation and source aging are quantified. Bremsstrahlung splitting, phase-space scoring and particle-tagging features are implemented in the Geant4 code, which is bench-marked against BEAMnrc simulations. HVLs from spectral measurements, attenuation curve measurements and Geant4 simulations mostly agree within uncertainty. However, there are discrepancies between measurements and simulations for photons emitted on the source transverse plane (90 degrees).