The effects of rotational setup errors in total body irradiation using helical tomotherapy

Abstract Purpose Helical tomotherapy (HT) is a form of intensity‐modulated radiation therapy that is employed in total body irradiation (TBI). Because TBI targets the whole body, accurate setup positioning at the edge of the treatment volume is made difficult by the whole‐body rotational posture. The purpose of this study is to clarify the tolerance for rotational setup error (SE) in the vertical direction. In addition, we perform a retrospective analysis of actually irradiated dose distributions using previous patients’ irradiation data. Methods To clarify the effects of rotational SE on the dose distribution, the planned CT images of 10 patients were rotated by 1–5° in the vertical (pitch) direction to create a pseudo‐rotational SE image. Then, the effect of the magnitude of the rotational SE on the dose distribution was simulated. In addition, the irradiated dose to the patients was analyzed by obtaining recalculated dose distributions using megavoltage CT images acquired before treatment. Results The simulation results showed that the average value of the lung volume receiving at least 10 Gy did not exceed the allowable value when the SE value was ≤2°. When the rotational SE was ≤3°, it was possible to maintain the clinical target volume dose heterogeneity within ±10% of the prescribed dose, which is acceptable according to the guidelines. A retrospective analysis of previous patients’ irradiation data showed their daily irradiation dose distribution. The dose to the clinical target volume was reduced by up to 3.4% as a result of the residual rotational SE. Although whole‐course retrospective analyses showed a statistically significant increase in high‐dose areas, the increase was only approximately 1.0%. Conclusions Dose errors induced by rotational SEs of ≤2° were acceptable in this study.


| INTRODUCTION
Total body irradiation (TBI) combined with chemotherapy is widely used as a pre-bone marrow transplant regimen in hematological malignancies, and it has superior treatment results to those of chemotherapy-only regimens. [1][2][3] Helical tomotherapy (HT) is a form of intensity-modulated radiation therapy that improves target dose uniformity and reduces the dose to organs at risk. Thus, HT has been used as a safer method of TBI administration. [4][5][6] Because TBI targets the whole body, the whole-body rotational posture makes accurate setup positioning at the field edge difficult. Additionally, when irradiating a large target, as in TBI, the patient couch sag peculiar to the HT apparatus increases, generating systematic rotational error in the vertical (pitch) direction. 7 However, HT systems are not equipped with a function to correct the rotational setup error (SE) in the pitch direction. 8 Furthermore, surface dose deviations due to SE have major effects in HT radiotherapy. 9, 10 Takenaka et al. recommended that the translational SE in the horizontal direction be within 5 mm in TBI with HT. 10 However, the dosimetric impact of rotational SE in the pitch direction in TBI using HT is unclear, and reports assessing this disadvantage are not yet available. Repeating the image-guided radiotherapy (IGRT) process, which means rotational correction in first MVCT and acquire the MVCT again for confirmation position is a problem because it requires a considerably long treatment time and increases patient distress. Moreover, the HT system allows you to skip the scan step, but it is inadvisable to do so because the external skin marks may not be a reliable indicator of the target position. In addition, it is impossible to achieve exactly the same patient position as that in the planning CT data. Therefore, knowing the tolerance level of rotational SE could contribute to rationalizing the IGRT process in HT systems.
In this study, we aimed to clarify the tolerance for rotational SE in the pitch direction in TBI using HT. Furthermore, we demonstrated the effects of residual rotational SE on the whole-body dose distribution by a retrospective analysis of irradiation data from previous patients.

2.A | Patients and treatment planning
This study's subjects were 10 patients who underwent TBI and were enrolled from January to December 2018, as approved by our hospital's Institutional Review Board (reception number: 18-034). The patients' characteristics are listed in Table 1. A whole-body suction fixture and a thermoplastic head mask were used to ensure fixation accuracy. The planning CT data of patients who underwent TBI were obtained. A 5-mm thick image was obtained using a 16-sensor data acquisition system-type wholebody CT system (Aquilion LB, Canon Medical Systems, Tochigi, Japan). The field of view was 550 mm. Then, separate treatment plans were created for the upper and lower parts of the body. The upper and lower plans were defined as the "head-first plan" (HF) and the "feet-first plan" (FF), respectively, according to reversed head-totail direction with respect to the system. For the safety of the image data acquisition process, the lengths (120 and 100 cm for the upper and lower body, respectively) were controlled to be shorter than the longest irradiation range of the HT system (i.e., 135 cm). 11,12 The anatomy of the pelvis can be used for image registration between the HF and FF images, which can improve the image registration accuracy. Therefore, the pelvis can be included in both images by acquiring images of length 120 cm and 100 cm from the top of the head and the toes, respectively. Moreover, we need to set the patient up twice for the individual plans (HF and FF plans) while maintaining the patient's posture. Therefore, we used a Styrofoam (polystyrene) board under the whole-body suction fixture, and the patient's position was rotationally exchanged to the other direction on the treatment couch along with the base (Styrofoam board and whole body suction fixture) by many medical staff members. The clinical target volume (CTV) was the whole-body contour excluding the lung, which is an organ at risk. In consideration of the setup margin, the planning target volume (PTV) was set to a volume obtained by adding 5 mm to the CTV toward the lung contour. This prevented insufficiency of the dose to the sternum and ribs adjacent to the lungs. According to a previous report, 10 no margin was added to the body contour to prevent an increase in excessively high-dose areas on the body surface. The prescription dose was optimized by a radiation treatment planning system (TomoHDA System Planning Station version 5.1.1.6, Accuray, Madison, WI, USA) using a constraint to cover 95% of the PTV with 12 Gy. In the FF plan, the CTV was the body contour. In the treatment planning for the HT plan, the field width was 5.0 cm, the pitch was 0.287, and the modulation factor was determined by adjusting each patient's value from 2.5 as a reference value. To reduce the region of field junctional overdose, we narrowed the target volume by 2.5 cm (5 slices) at both the upper and lower body irradiation junctions. 13  makes it possible to achieve a gradually attenuated dose distribution at the junction. In addition, Two regions were defined as the prediction area of large dose deviations for the CTV in each plan. The outer boundary of the CTVbs (CTV of the body surface) is the contour of the patient's surface. The inner boundary of the CTVbs is expanded 5 mm inward from the patient's surface. The inner boundary of the CTVls (CTV of the lung surface) is the lung contour. The outer boundary of the CTVls is expanded 5 mm outward from the lung contour. (Fig. 1). Additionally, one region was defined as the prediction area of large dose deviations for the CTV in the overall combined plan. The CTVjt region (CTV junction between the HF and FF plans) was defined as that limited to 5 cm (10 slices) in both the head and feet directions from the junction slice in the CTV.

2.B | Pseudo setup error simulation
To simulate the effects of rotational SE on dose distribution, the planning CT images of each patient were rotated by 1-5°in the pitch direction. In our study, the center in the body axis direction in the CT image was located at the rotational center of the SE. Therefore, the slice number of the CT image at the rotational center was 60 and 50 in the HF and FF plans, respectively. Simultaneously, each organ structure dataset attached to the planning CT image was rotated using commercially available software (MIM Maestro version 6.5.9, MIM Software, Inc., Cleveland, OH, USA to be maintained within ± 10% of the prescription dose: V 110% must be 10% or less and V 90% must be 90% or more of the CTV. We quantitatively compared the change in high-and low-dose regions within the CTVs between the original and simulation doses. The lungs were evaluated for mean dose and V 10Gy (the volume receiving at least 10 Gy). The V 10Gy value was not allowed to exceed 40% of the whole lung volume, with reference to the predictors of radiation pneumonia in patients who underwent radiation therapy. 15 Therefore, changes in the dose distribution within the CTVs and lungs due to increased rotational SE were compared with the original plan dose distribution to establish SE tolerance.
Additionally, because of the couch sag and the differences in the patient's posture, the pitch error exerts its effects in opposite directions in the HF and FF plans. Therefore, we analyzed the impact on the junction area in the overall combined plan. Assuming the pitch offsets in the direction in which the couch is expected to sag situation, we analyzed the effects of rotational SE on the junction area in the head-down and toe-down directions for the HF and FF plans, respectively.
The differences between the means of the original planned dose distribution and simulation dose distribution with rotational SE were considered statistically significant when p < 0.05 (two-tailed t-test).
R version 1.41 (www.r-project.org) was used for the paired t-tests.

2.C | Evaluation of delivered dose distribution
For the HT treatment procedure, we obtained the megavoltage CT (MVCT) data from all treatment periods before each irradiation.
Furthermore, the SE correction made using these data was finalized  Tables 2 and 3 show the dose indexes with various rotational SE values using the HF and FF plans, respectively. In the following text, statistically significant differences are described as significant differences. The mean D 95% values ± standard deviation (SD) of the original HF and FF plans in the CTV were 12.0 Gy and 11.9 Gy, respectively, with SD values of <0.1.

3.A | Tolerance levels of SE in the pitch direction
Conceptual diagram from deformation processing of megavoltage CT (MVCT) to the integration process into the planning CT image. Red and blue pixels on the MVCT image indicate 110% and 90% of the prescribed dose, respectively, and thus denote high-and low-dose points, respectively. Because each pixel on the MVCT image is assigned to the corresponding pixel on the planning CT image by deformable image registration (DIR) processing, the dose associated with the MVCT pixels was also assigned to the planning CT pixels' positions. In the process of integration, high-dose and lowdose points may cancel or add to each other. In this figure, the positions of the canceled doses are shown in white, which means no dose difference.
T A B L E 2 Effects caused by rotational setup error in the pitch direction using the head-first plan.

Plan
Rotational SE (degrees) in pitch direction. (Average ± SD) 12.0 ± 0.0 11.9 ± 0.0* 11.9 ± 0.0* 11.8 ± 0.0* 11.6 ± 0.0* 11.4 ± 0.0* D 2% (Gy) 12.9 ± 0.0 13.    Figure 5 shows the intensity of the dose distribution at the junction of the HF, FF, and combined plans. By narrowing the target volume at both body sides in the treatment planning, the dose intensity in the junction region was attenuated in a gradation pattern to ensure that its effect on the dose distribution at the confluence of the HF and FF plan images was kept within an acceptable range. Table 4 shows the results of the impact on the junction area in the overall combined T A B L E 3 Effects caused by rotational setup error in the pitch direction using the feet-first plan.

3.B | Accumulated dose in the recalculation of delivered dose distribution
The residual of the rotational SE in the pitch direction, which cannot be corrected, was extracted for each patient. Table 5      | 99 V 110% and V 90% of the CTV in the FF plan were 5.3% ± 2.4% and 99.6% ± 0.1%, respectively ( Table 3).
The results of the impact on the junction area in the overall combination plan showed that CTVjt D 2% was a statistically significant differences from the original planned dose at SE values of ≥2°.
When the rotational SE was 2°, the near maximum dose in the CTVjt was 18% higher than that without SE. However, D 98% and D 95% were not significantly different from the original planned dose at SE values of ≤4°. In addition, the average value of V 110% in whole CTV was ≤10%, therefore, because the dose increased only in the local region at the rotational SE value of 2°, the effect on the entire CTV was small.
The simulation results showed that a high-dose region was caused by the SE, with D 2% and V 110% showing significant increases even at the rotational SE value of 1°. Specifically, the maximum dose in the CTVbs increased with increased rotational SE. Figure 6 shows that the effects on the CTVbs value were quite large: the D 2% value was increased by 12.7% at the SE value of only 1°. There is a highbeamlet fluence area around the body surface to maintain a sufficient dose in the build-up region. 10  In this study, rotational SE was found to cause an unexpected expansion of the high-dose region near the contour surface of the patient's body. Therefore, a future task is to improve treatment planning methods to reduce the effects of rotational SE on the dose distribution.

| CONCLUSION
The effect of rotational SE values of ≤3°on dose heterogeneity in the CTV was within ± 10% of the prescribed dose. We conclude that the rotational SE value range of ≤2°is an acceptable limit. This is because the results in the high-dose region of the lung became unacceptable when the rotational SE was ≥3°. By clarifying the permissible value of rotational SE, it is expected that the SE correction procedure in IGRT will be rationalized and that burden on the patient due to the extension of treatment time will be prevented.
However, rotational SE in TBI treatment using HT may increase the radiation dose, particularly on the patient's body surface.

ACKNOWLEDG MENTS
We thank the members of the Department of Radiology of Juntendo University Hospital.

CONF LICT OF I NTEREST
The authors declare that they have no conflict of interest.

ETHICAL APPROVAL
Since this study was a noninterventional and noninvasive study, and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards.

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 available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.