DNA strand breaks based on Monte Carlo simulation in and around the Lipiodol with flattening filter and flattening filter-free photon beams.

Background: The current study aims to investigate the DNA strand breaks based on the Monte Carlo simulation within and around the Lipiodol with flattening filter (FF) and flattening filter-free (FFF) photon beams. Materials and methods: The dose-mean lineal energy (yD) and DNA single- and double strand breaks (DSB/SSB) based on spatial patterns of inelastic interactions were calculated using the Monte Carlo code: particle and heavy ion transport system (PHITS). The ratios of dose using standard radiation (200 kVX) to the dose of test radiation (FF and FFF of 6 MV X-ray (6MVX) and 10 MVX beams) to produce the same biological effects was defined as RBEDSB. The RBEDSB within the Lipiodol and in the build-up and build-down regions was evaluated.
Results: The RBEDSB values with the Lipiodol was larger than that without the Lipiodol at the depth of 4.9 cm by 4.2% and 2.5% for 6 MVX FFF and FF beams, and 3.3% and 2.5% for 10 MVX FFF and FF beams. The RBEDSB values with the Lipiodol was larger than that without the Lipiodol at the depth of 6.5 cm by 2.9% and 2.4% for 6 MVX FFF and FF beams, and 1.9% and 1.4% for 10 MVX FFF and FF beams. In the build-down region at the depth of 8.1 cm, the RBEDSB values with the Lipiodol was smaller than that without the Lipiodol by 4.2% and 2.9% for 6 MVX FFF and FF beams, and 1.4% and 0.1% for 10 MVX FFF and FF beams. Conclusions: The current study simulated the DNA strand break except for the physical dose difference. The lower and FFF beam occurred the higher biological effect.

Introduction

DNA damage by irradiation may lead to cell death. Theoretical calculation approaches complement experimental research on the induction of DNA damage under various conditions. It can simulate the DNA damage by various irradiation conditions quantitatively [1]. Irradiation with photon beam on tissue causes the generation of secondary electrons due to various interactions such as photoelectric absorption and Compton scattering [2]. DNA damage leading to cell death and chromosomal aberrations is caused by the energies of secondary electrons deposited into cells [3]. The biological effectiveness (RBE) can be evaluated by analyzing the local energy deposition of secondary electrons on micro- or nano-meter scale [4].

Previously, Monte Carlo (MC) simulation codes such as PARTRAC, Geant4-DNA, and KURBUC have been used to evaluate the DNA damage by irradiation [5—7]. PARTRAC and KURBUC are based on electron scattering cross sections in water vapor, and Geant-DNA is based on a combination of analytical and interpolated cross sections for liquid water. However, these remain uncertain of the quality of radiation transport at the lower energy below sub-kiloelectron volt. On the other hand, Particle and Heavy Ion Transport code System (PHITS) can simulate the track structure of electrons in liquid water in the low energy range below sub-kilo electron volt [8-10]. The biological effect has been generally investigated with the RBE based on the linear–quadratic (LQ) model [11, 12]. However, the β that is a parameter of the survival fraction calculation based on the LQ model depends on the dose rate and cell lines as well as the profiles of radiation in a complicated manner [13]. Another study represented that the β is increasing with increasing LET [14]. Howkins et al. proposed a new model, Microdosimetric-kinetic (MK) model, which is able to predict the β value with increasing LET [15]. Our previous study showed the RBE enhancement within the Lipiodol with MK model [16]. The RBE according to the MK mode is calculated from dose and lineal energy (yd(y)). The dose mean lineal energy d(y) is able to be obtained from PHITS simulation [17]. RBE was defined as the ratios of dose using reference radiation to the dose of test radiation to produce the same biological effects. However, the RBE defined in the previous study was affected by the physical dose in addition to the biological effect.

Recently, PHITS was updated, so that it can use the electron track structure mode (etsmode) [18]. It enables the evaluation of the impacts of low-energy electrons on the DNA strand break induction such as single strand break (SSB) and double strand break (DSB) with Monte Carlo simulation. Namely, it can evaluate the biological effect directly.

A previous study evaluated the dose enhancement within the Lipiodol (Guerbet, Villepinte, France) which includes a high-atomic number material [19]. Lipiodol has been used as an embolic agent in trans-arterial chemoembolization (TACE). The dose enhancement is larger with FFF beam than that with FF beam [20]. Also, we investigated the dose difference with and without the Lipiodol in the build-up region [21]. The area of the dose enhancement by the backscatter was within 3 mm from the Lipiodol. Other studies reported the dosimetric effects in the build-up and build-down regions for the high-atomic number materials such as the metal material [22, 23]. However, these dose differences were calculated with the physical dose. Thus, the factor for the difference of the DNA damage with and without the Lipiodol was not revealed.

In the current study, the DNA stand breaks and the RBE calculated using the ratio of DNA-DSBs were evaluated in and around the Lipiodol compared for FF and FFF of 6 MVX and 10 MVX beam, except for the physical dose difference.

Materials and methods

Monte Carlo calculations

A TrueBeam linac (Varian Medical Systems, Palo Alto, USA) for FF and FFF of 6 MVX and 10 MVX beams was modelled. Varian provides IAEA-compliant phase-space files located just above the secondary X/Y collimator, which were simulated using the GEANT4 MC code. The IAEA-compliant phase-space was scored above the secondary collimator. The present study modelled the phase-space files below the secondary collimator using BEAMnrc [24]. The phase-space data scored at a source-to-surface distance (SSD) of 90 cm. In the Monte Carlo simulation, a Lipiodol (3 × 3 × 3 cm3) located at a depth of 5.0 cm in a virtual water-equivalent phantom (30 × 30 × 30 cm3) was modelled, as shown in Figure 1. Lipiodol includes an ethiodized oil injection and is a sterile injectable radio-opaque diagnostic agent. The Lipiodol in each millimeter contains 480 mg of iodine organically combined with ethyl esters of fatty acids of poppy seed oil [25]. The physical density and cross-section data were assigned for the Lipiodol and the water.

Figure 1

Geometric scheme of virtual water and Lipiodol located at a depth of 5.0 cm in a water-equivalent phantom (30 × 30 × 30 cm3)

These phase-space files were transferred to the PHITS which is able to deal with the transport of nearly all particles such as protons, neutrons, electrons, and photons over a wide energy range [26]. The photon beams were irradiated with a 5 × 5 cm2 of field size at SSD = 90 cm for the virtual phantom.

The dose-mean lineal energy () is derived by [17]:

where f(y) is the probability density of the lineal energy and d(y) is the dose distribution of the lineal energy. The energy spectrum and the dose-mean lineal energy within the Lipiodol, and in the build-up and build-down region were calculated.

Calculation of relative biological effectiveness

The yields of SSBs and DSBs were calculated with the track structure code which was incorporated in PHITS ver. 3.20. Through the medium, the electrons make much collisional positions. These sequential positions are defined as the track structure. Matsuya et al. reported that DSBs calculated using PHITS is in good agreement with the other simulation codes by Geant4-DNA and experimental data for both electron and photon irradiations [27]. The experimental DSB data was collected from the literature and performed a γ-H2AX focus formation assay [28-30]. Kai et al. showed the track structure is useful for the modeling of DNA damage [31]. In the track structure code in PHITS, the yields of SSBs and DSBs were calculated by stochastically sampling the number of events per track and that of a pair composed of two events within 3.4 nm (10 base pairs) [32]. It was based on the assumption that ionization and electronic excitation are potential causes to induce DNA strand breaks [33, 34]. The yield of SSBs and DSBs were calculated based on assuming that the number of events per keV Nevent/Ein and that of linkage per keV Nevent/Ein are proportional to the induction yield of SSB and DSB.

The coefficients kSSB = 5.66 × 10−12 (keV/Gy/Da) and kDSB = 1.61 × 10−13 (keV/Gy/Da) were obtained by applying the PHITS to the experimental yields of SSB and DSB with 220 kVp X-rays [28].

The SSB and DSB, and the ratio of DSB and SSB (DSB/SSB) were calculated using the energy spectra of electrons after irradiation with FF and FFF of 6 MVX and 10 MVX beams on each grid within the Lipiodol and in the build-up and build-down regions. The build-up region was set at the 4.9 cm and build-down region was set at 8.1 cm.

Moreover, since the absorbed dose is related to the number of DNA-DSBs per nucleus [35], the ratio of DNA-DSBs for standard radiation of 200 kVp X-rays, which was defined as RBEDSB, was calculated by:

Results

The dose-mean lineal energy yD

The dosemean lineal energy yD values with and without the Lipiodol at 4.9 cm (build-up region region), 6.5 cm (within the Lipiodol region), and 8.1 cm (build-down region) for the FF and FFF of 6 MVX and 10 MVX beams are shown in Table 1. The yD values with the Lipiodol were larger than those without the Lipiodol by 0.5% for 6 MVX FF and 0.6% for 6 MVX FFF beams at the depth of 4.9 cm, and 3.9% with the Lipiodol and 4.6% without the Lipiodol at the depth of 6.5 cm for 6 MVX beam. For 10 MVX beam, the yD values with the Lipiodol were larger than thaose without the Lipiodol by 0.3% for 10 MVX FF and 0.4% for 10 MVX FFF beams at the depth of 4.9 cm, and 3.5% for 10 MVX FF and 3.6% for 10 MVX FFF beams at the depth of 6.5 cm. On the other hand, the yD values with the Lipiodol were smaller than those without the Lipiodol beam by −0.3% for 6 MVX FF and −0.4% for 6 MVX FFF beams at the depth pf 4.9 cm in the build-down region. The yD values with the Lipiodol for 10 MVX FFF beam were smaller than 10 MVX FF beam by −0.2% and −0.2% in the build-down region.

Table 1

The dosemean lineal energy in the water and Lipiodol for the flattening filter (FF) and flattening filter-free (FFF) of 6 MVX and 10 MVX beams within the Lipiodol and in the build-up and build-down regions

	6 MVX FF [keV/μm]		6 MVX FFF [keV/μm]		10 MVX FF [keV/μm]		10 MVX FFF [keV/μm]
Region	Water	Lipiodol	Water	Lipiodol	Water	Lipiodol	Water	Lipiodol
Build-up	1.96	1.97	2.06	2.07	1.85	1.86	1.93	1.94
Within Lipiodol	1.97	2.05	2.08	2.17	1.86	1.93	1.93	2.00
Build-down	1.98	1.97	2.08	2.07	1.86	1.85	1.94	1.93

The yield of DSB/SSBs with and without Lipiodol

Figure 2 show the yield ratio of DSB/SSBs at 6.5 cm within the Lipiodol region and at 4.9 cm in the build-up and at 8.1 cm in the build-down regions for FF and FFF of 6 MVX and 10 MVX beams. The DSB/SSBs with the Lipiodol was larger than that without the Lipiodol by 2.5% for 6 MVX FF and 4.2% for 6 MVX FFF beams at the depth of 4.9 cm, and 2.4% with the Lipiodol and 2.9% without the Lipiodol at the depth of 6.5 cm. The DSB/SSBs and RBEDSB with the Lipiodol were smaller than those without the Lipiodol by 2.9% for 6 MVX FF and 4.2% for 6 MVX FFF beams at the depth of 8.1 cm. Moreover, the RBEDSB with the Lipiodol was larger than that without the Lipiodol by 2.5% for 10 MVX FF and 3.2% for 10 MVX FFF beams at the depth of 4.9 cm, and 1.4% with the Lipiodol and 1.9% without the Lipiodol at the depth of 6.5 cm. The RBEDSB with the Lipiodol was smaller than that without the Lipiodol by 0.1% for 10 MVX FF and 1.2% for 10 MVX FFF beams at the depth of 8.1 cm.

Figure 2

The yield ratio of single- and double strand breaks (DSB/SSB) with and without Lipiodol for flattening filter (FF) and flattening filter-free (FFF) of 6 MVX and 10 MVX beams at the depth of 4.9 cm in the build-up region (A), 6.5 cm within the Lipiodol (B), and 8.1 cm in the build-down region (C)

The yield ratio of RBEDSB with and without Lipiodol

Figure 3 shows the yield of and RBEDSB at 6.5 cm within the Lipiodol region and at 4.9 cm in the build-up and at 8.1 cm in the build-down regions for FF and FFF of 6 MVX and 10 MVX beams. The yields of DSBs with the Lipiodol was larger than without the Lipiodol by 3.3% for 6 MVX FF beam and 3.7% for 6 MVX FFF beam at the depth of 4.9 cm, and 2.3% for 6 MVX FF beam and 2.7% for 6 MVX FFF beam at the depth of 6.5 cm. The yields of DSBs with the Lipiodol were smaller than those without the Lipiodol by 2.5% for 6 MVX FF beam and 4.0% for 6 MVX FFF beam at the depth of 8.1 cm. The yields of DSBs with the Lipiodol +were larger than those without the Lipiodol by 1.9% for 10 MVX FF beam and 3.2% for 10 MVX FFF beam at the depth of 4.9 cm, and 1.6% for 10 MVX FF beam and 1.7% for 10 MVX FFF beam at the depth of 6.5 cm. The yields of DSBs with the Lipiodol were smaller than those without the Lipiodol by 0.2% for 10 MVX FF beam and 1.4% for 10 MVX FFF beam at the depth of 8.1 cm.

Figure 3

The RBEDSB with and without Lipiodol for flattening filter (FF) and flattening filter-free (FFF) of 6 MVX and 10 MVX beams at 4.9 cm in the build-up region (A), 6.5 cm within the Lipiodol (B), and 8.1 cm in the build-down region (C)

For the comparison of the RBEDSB with and without the Lipiodol, the DSB/SSBs and RBEDSB with the Lipiodol were larger than those without the Lipiodol by 2.5% for 6 MVX FF and 4.2% for 6 MVX FFF beams at the depth of 4.9 cm, and 2.4% with the Lipiodol and 2.9% without the Lipiodol at the depth of 6.5 cm. The DSB/SSBs and RBEDSB with the Lipiodol were smaller than those without the Lipiodol by 2.9% for 6 MVX FF and 4.2% for 6 MVX FFF beams at the depth of 8.1 cm. Moreover, the RBEDSB with the Lipiodol was larger than that without the Lipiodol by 2.5% for 10 MVX FF and 3.3% for 10 MVX FFF beams at the depth of 4.9 cm, and 1.4% with the Lipiodol and 1.9% without the Lipiodol at the depth of 6.5 cm. The RBEDSB with the Lipiodol was smaller than that without the Lipiodol by 0.1% for 10 MVX FF and 1.4% for 10 MVX FFF beams at the depth of 8.1 cm.

Discussion

In this decades, high atomic number materials have attracted increasing attention as radiosensitizers. The physical dose enhancement has been investigated with the gadolinium, gold nano particle (GNP), and iodine [36-44]. The data of chemical and biological effect with the GNP was shown by Butterworth et al [45-47]. These experimental data represent that degree of radiosensitization is greater than the predicted increase in physical dose. A previous study investigated the biological dose enhancement with and without the Lipiodol [16]. We indicated a possibility of the enhancement of the biological effect by the result that the biological dose enhancement was higher than the physical dose enhancement. However, the biological dose included the physical dose and the biological effect. The current study investigated the DNA strand breaks except for the physical dose difference with and without the Lipiodol. From the result, the DSB/SSB was higher in the build-up and within the Lipiodol than without the Lipiodol. The DSB yield is increased, which indicates the enhancement of the biological effect. This result was agreement with the previous studies in the point of that the radiosensitization considered with the biological effect was greater than the physical dose enhancement by the simulation [45-48]. In the build-up region and within the Lipiodol region, the scattered electrons with lower energy from the Lipiodol were increased. The lower energies are absorbed in the Lipiodol; thus the average energy is increased in the build-down region. The higher average energy decreases the DSB yield and the RBE. The average energy of the incident and scattered photon and electrons affects the DSB yield and the RBE.

In the comparison of the FF and FFF beams, the FFF beam has a larger biological effect than FF beam for both Lipiodol and water. The FFF beam contains more lower photon and electrons than FF beam [21]. Suneil et al. examined in vitro radiosentization by GNP in MDA-MB-231 cells for 6 MV and 15 MV photon energies [47]. The sensitizer enhancement ratio with and without the GNP was 1.29 for 6 MV X-rays and 1.16 for 15 MV X-rays. Rahman et al. reported on the dose enhancement factors at 90% cell survival, which was 2.9 for 6MV X-rays and 3.7 for 12 MV at 0.5 mmol of GNPs [49]. These results indicate that the lower photon energy causes more radiosensitization. Matsuya et al. reported that the maximum of the yield of DSB occurred at 300 eV of the electron energy [27]. In the previous study, we showed that FFF beam contained more photons and electrons with energies mostly at 300 eV [20]. The lower photon beam increases the DSB yield and RBE. Therefore, the RBEDSB with FFF beam was higher in the build-up region and within the Lipiodol than that with FF beam.

In the clinical liver SBRT, Lipiodol has been used as an embolic agent and for tumor seeking in trans-arterial chemoembolization (TACE). Lipiodol remains in the tumor during radiotherapy treatment. The Lipiodol has a benefit to enhance radiobiological effect in the tumor intensively. Moreover, the radiobiological effect can be increased more by the irradiation in the tumor with the Lipiodol with FFF beam. The average energy of the incident and scattered photon and electrons affects the DSB yield and the RBE.

There is a limitation in the current study. The current study was performed with the simulation. The maximum enhancement of the RBEDSB was 4.2% for 6MVX FFF beam. Further study to evaluate the effects of radiobiology by comparing the result in vivo or vitro and simulation is needed.

Conclusions

The current study simulated the DNA strand break except for the physical dose difference. The result suggest that DNA damage increased within the Lipiodol and the build-up region. Additionally, the lower and FFF beam occurred the higher biological effect. It might be benefit for the clinical treatment in the point that the Lipiodol enhance the DNA damage for the tumor locally.