Effect of auto flash margin on superficial dose in breast conserving radiotherapy for breast cancer

Abstract Purpose To investigate the dose‐effect of Auto Flash Margin (AFM) on breast cancer's superficial tissues based on the Treatment Planning System (TPS) in the breast‐conserving radiotherapy plan. Methods A total of 16 breast‐conserving patients with early stage breast cancer were selected, using the X‐ray Voxel Monte Carlo (XVMC) algorithm. Then, every included case plan was designed using a 2 cm‐AFM (the value of AFM is 2 cm) and N‐AFM (without AFM). Under the condition of ensuring the same configuration of #MU and collimator, the absorbed dose after a simulated inspiratory motion was calculated again using the new plan center, which moved backward to the linac source. The dose difference between the measurement points between AFM and N‐AFM groups was compared. Results In the dose results, PTVV50Gy of the AFM group was superior to that of the N‐AFM group, PTVD2%, PTVDmean, Lung_IpsiV20Gy, Lung_IpsiDmean, and BodyDmax. Also, the dose results of the N‐AFM group were significantly higher than those of the AFM group. However, there was no significant difference between Lung_ContraV5Gy, HeartDmean, and Breast_ContraV10Gy in the two groups. In the collimator alignments at the same angle between groups, the AFM group formed an apparent air region outside the collimator compared with the N‐AFM group. In the XVMC algorithm feature parameter, the AFM group had less #MU, higher QE, and slightly longer optimization time. The #segments of both groups were close to the 240 control points preset by the plan. The validation results of EBT3 film in both groups were more significant than 95%, meeting the clinical plan's application requirements. The difference in film results between groups was mainly reflected in the dose distribution at the near‐source. 4DCT was used to summarize the maximum and minimum inspiratory motion distances of 7.31 ± 0.45 and 3.42 ± 0.91 mm respectively. Conclusions These results suggest that the AFM function application could significantly reduce the possibility of insufficient tumor target caused by inspiratory motion and ensure sufficient tumor target exposure.

Currently, the management and control technique of inspiratory motion in the superficial area of breast cancer is mostly to make up for the off-target irradiation of the chest wall by expanding PTV or adding a virtual bolus. However, for high-precision radiotherapy techniques like 3DCRT and IMRT, the tumor target location's deviation is likely to result in a higher dose, making the surrounding organs at risk. Besides, the virtual bolus will misjudge the actual absorbed amount in the superficial breast area and affect the radiotherapy dose's accuracy. 4,5 AFM function is a tool that produces reasonable monitor units per fraction and a small standard deviation in a plan. Application of this tool provides a uniform distribution in the buildup region. In the current common IMRT schemes for breast cancer, the radiation field is formed according to the shape of the target area. In this case, off-target irradiation due to inhalation motion cannot be avoided. After applying the AFM function, a cavity within the radiation exposure range is established outside the body surface, so even under the inhalation motion, the superficial tissues can still be guaranteed to be irradiated.

2.A | Data
In this study, we enrolled 16 patients (nine patients with T1N0M0 and seven patients with T 2 N 0 M 0 , aged 23-45 years) whose left breast had cancer and underwent breast-conserving surgery. Here, we first used a 4D computerized tomography (CT) scan ( ie "beamstep -beam," a technique for performing continuous multiphase CT scans, CT rotation time was 0.8 seconds, the axial thickness of 2.5 mm, a voltage of 120kV, a current of 350mAs ) to axial film the thoracic inlet to the base of the lungs under free breathing. The patient's body surface was longitudinally fitted with a 5 cm lead wire from the sternum to monitor the patient's respiratory status expressed with sinusoidal waveform, whereby every 10% was a 1time phase (0~90%). Next, maximum and minimum intensity projected images (MIP, MIN) and average intensity projected images (AIP) were obtained after the reconstruction of 10-time phases using the respiration time-phase fusion control technology. ,6 Lastly, all the acquired images were uploaded to the Monaco TPS, and radiotherapy physicians (with more than five years of work experience) contoured tumor targets following the National Comprehensive Cancer Network-China guidelines 2019 and subsequently reviewed by the radiotherapy chief physicians. They had a work experience of more than 10 years.

2.B | Plan design and requirements
First, cases were prescribed with a VMAT of 6MeV photon energy, a fraction dose of 2 Gy, a fraction of 25, and a total amount of 50 Gy. Next, regarding the International Radiation Oncology Collaborative Group (RTOG) 1304 report, the OARs and tumor targets were dose-constrained. Subsequently, the beam was designed using two arcs counter-clockwise (150~300°) at a: grid spacing of 3 mm, beam margin of 5 mm, max.# control points per arc of 120, and a min.segment width of 6 mm. Consequently, we calculated the dose deposition to medium and adopted it for planning design. In this study, we planned to use the Monaco 5.11 clinical planning system (Swedish Elekta company) with HP Z820 server, 128G memory, and NVIDIA TESLA C2075 GPU mount to design the treatment plan. Lastly, the procedures were carried out using Versa HD™ clinical linear accelerator (Sweden) that was equipped with a collimator of Elekta Agility™ (80 pairs of leaves).

2.C | Experimental design of inspiratory motion
When selecting measurement points: we applied both interest points and markers to choose a range of 11~14 from skin boundary markers at different degrees such as 30°, 60°, 90°, and 120°.
Also, a range of I5~I8 from 3 mm of the subcutaneous skin and a volume at a measurement point of 0.081cc, 7-9 as presented in     3.C | Difference in collimator alignment between the AFM and the N-AFM group at the same angle

3.F | Comparison of validation results between groups
To validate the AFM and the N-AFM plans in the same group, and EBT3 film was placed parallel to the treatment table surface in solid water (5 cm above and below). As illustrated in Figure 7, this film was then scanned using an EPSON 10000XL scanner, and the results were analyzed using doselab software.
In film validation (3, 3%), both the AFM and the N-AFM groups had a pass rate of over 95%. However, the N-AFM group's pass rate was slightly higher than that of the AFM group, hence, meeting the clinical application planning requirements. In the comparative analysis of film validation results between the AFM group and the N-AFM group, the two groups' dose distribution at the far source end was the same. Still, the AFM group's dose distribution at the nearsource end was significantly higher than that of the N-AFM group.

CONF LICT OF I NTERESTS
The authors declare no conflict of interest. Hua: analyzed the experimental data and drafted the paper.

D A T A A V A I L A B I L I T Y S T A T E M E N T
The data that support the findings of this study are openly available in public repositories. All references are cited and the DOIs are attached. Please refer to section REFERENCES for details. The data that support the findings of this study are available from the corresponding author upon reasonable request. The data that support the findings of this study are available in the supplementary material of this article.