First Experience of Standard Linac, Rapid Arc and Dose-drop Scheme Based Total Body Irradiation

Background and purpose: To introduce a standard linac with true attachment free, rapid arc and dose-drop scheme for positioning related dose deviation control based total body irradiation(TBI). Materials/Methods: One eight years old girl diagnosed with acute lymphocytic leukemia underwent TBI in 2020. Target volumes and organs at risk were contoured after CT simulation. Total sixteen ARC and four AP-PA from ve isocenters were designed. A dose-drop scheme on both sides of adjacent region were performed to reduce positioning-related dose deviation. A series of quality assurance before radiotherapy and real-time dose monitoring during radiotherapy were carried out. Results: The average on board imaging (OBI) time of per fraction was 40.3 min, the average beam on time of per fraction was 37.2 min, the average time to change from head rst to feet rst position was 18.4 min. The average mean lung dose was 9.89 Gy, the maximum lens dose was 7.60 Gy, the mean PTV_total dose was 12.17Gy, 98.23% PTV_total volume was covered by 90% of the prescription dose. The maximum dose (Dmax) of PTV_total was 13.65Gy. Dmean and V10.8 of PTV_total are only slightly different(0.49% - 1.89% and 0.26% - 1.04% respectively) even with an error of 5 - 20mm longitudinal misalignments. Gamma passing rate(3mm/ 3% Gamma criteria) are between 93.5% and 100%. Real-time dose monitoring showed an overall deviation of -3.9%±5.51%. Conclusions: Standard linac, rapid arc and dose-drop for positioning-related dose deviation control based total body irradiation is feasible, accurate, and reliable. It is worthy of clinical application.


Introduction
1-3 TBI is frequently used as a component of hematopoietic stem cell transplant (HSCT). 4 The techniques of TBI vary widely from institution to institution. Conventional TBI often use two or three dimensional radiotherapy techniques, for example opposing anterior and posterior elds or lateral elds.
Disadvantages of conventional TBI include: 1. attachment devices and long source skin distance needed.
2. shielding speci c organs are not precise. 3. poor dose uniformity. The inverse-planned intensitymodulated radiation therapy(IMRT) based TBI technique was developed to improve these shortcomings. But many IMRT TBI studies also need some special attachment and rare positioning related dose deviation control techniques. Now we report a standard linac with true attachment free, rapid arc and dose-drop positioning related dose deviation control based total body irradiation.

Materials And Methods
One eight years old girl diagnosed with acute lymphocytic leukemia underwent allogenic stem-cell transplantation following TBI and VP16 chemotherapy treatment at Hongkong university Shenzhen hospital in 2020. Patient was informed about the treatment and it`s possible adverse events and about necessary diagnostics prior to treatment. Written consent by the patient herself and parents was obtained.
Positioning, immobilization and simulation Simulation occurs on a computed tomography scanner (Philip Brilliance CT Bigbore) with a setting of 120kVp, 3 mm slice. The scan length is limited to 1.15meters. Both upper limbs are placed at the bilateral sides of the body. Head & neck and shoulders thermoplastic mask, head rest and moldable head and neck support are used to x the head and neck and shoulders. Whole body thermoplastic cast and whole body evacuated vacuum bags are used to x thorax, abdomen, and limbs. The patient was positioned on a total body base board, in order to connect all above positioning devices as a whole unit as shown in Fig. 1. Prior to CT acquisition, radiopaque markers are placed near the umbilicus to serve as origin (Fig. 2) and a merge point of the two scans. The overlap area of two scanned images should be longer than 7cm(5 cm for minimized dose gradient at radiation elds junctions, 2 cm for scattered dose calculation at radiation elds edge). If the patient is taller than 1.15 meters, two scans will be performed, the rst scan (upper body) goes from the top of the head to pelvis with head rst position, the second scan (lower body) goes from the bottom of the feet to the pelvis with feet rst position. Both scans include the above origin markers. To allow for quality assurance (QA) measurements during radiotherapy, total eight 0.5 cm bolus are also scanned along with the patient, eight metal oxide semi-conductor eld effect transistor (MOSFET) will be placed under bolus for timely treatment dose monitoring as shown in Fig. 1.

Contouring
Body was contoured using search body automatic tool, then manually modi ed to include all the outer body contour.
Even with a well-positioned devices, given the movement of the chest wall and ribs, The planning target volume (PTV) was contoured using the outer body contour minus bilateral lungs (except 3 mm margin of lung tissue adjacent to the ribs and chest wall to ensure full dose coverage of the ribs and chest wall).
PTV_upper and PTV_lower were contoured on upper body and lower body scans respectively. PTV_total is equal to PTV_upper plus PTV_lower. PTV_crop was equal to PTV_total shrinks by 3 mm in the threedimensional direction. The right, left and bilateral lungs were contoured using the pulmonary windows separately according to RTOG atlases for Organs at Risk (OARs) in Thoracic Radiation Therapy.
Due to protection of lung tissues and dose coverage of chest wall, helping structures PTV_chestwall_1 cm was de ned as the extrapulmonary three-dimensional area within 1 cm. In order to ensure adequate skin dose coverage, the area with PTV_upper and PTV_lower expansion of 3 mm were named PTV_upper_3 mm and PTV_lower_3 mm respectively. At the same time, in order to reduce the actual dose coverage error caused by the positioning error, ve sections of dose-drop transition zone were drawn continuously at the junction, the length of each section was 1 cm, ve sections were named as Step_12Gy-10Gy, Step_10Gy-8Gy, Step_8Gy-6Gy, Step_6Gy-4Gy and Step_4Gy-2Gy respectively as shown in Fig. 2. The parts of PTV_total in the dose-drop region was named PTV_drop. However, for dose statistics and dose volume histograms(DVH), the anatomical lungs are the relevant structures. The right, left and bilateral kidneys were contoured if pre-existing renal insu ciency. Additional helping structures within the overlapping regions were contoured and used for steering the optimizer leading to an improved dose distribution in those areas. Other OARs were not routinely contoured and involved in dose optimization.

Treatment planning and irradiation
The full-body CT scan was imported into Eclipse treatment planning system, version 15.0. Isocenter was created for treatment planning of each scan. The total prescription dose was 12 Gy, 2 Gy per fraction, two fractions per day, the minimum interval was 6 h. Irradiation was delivered at linac (Triology, Varian) based photon energy of 6 MV over three consecutive days. The elevator and linear accelerator room were routinely disinfected before each fraction radiotherapy. The dose constraint is as follows: The minimum dose (Dmin) of PTV is more than or equal to 90% of prescription dose; The Dmax of PTV is less than or equal to 130% of prescription dose; The volume of PTV covered by 120% of the prescription dose should be less than 10%. The mean dose of both lungs is less than or equal to 10 Gy; The mean dose of both kidneys is less than or equal to 10 Gy if renal insu ciency, no special dose limitation for both kidneys if normal renal function.
As shown in Fig. 2, total sixteen ARC and four AP-PA from ve isocenters were designed. For upper body treatment planning, total 12 ARCs were designed with three iso-centers because of the collimator eld limitation is 40 cm X 40 cm. For lower body treatment planning, four AP-PA elds in elds were designed for bilateral shanks with a isocenter rstly. Then, another four ARCs with another isocenter were used to cover other parts of lower body. For the convenience of positioning, the coordinate values of each isocenter point are only different in the longitudinal direction. The connection between each rapid arc plan at each isocentric point was administrated by the function module of Base Dose Plan Compensation (BDPC).
Regarding the adjacent region dose distribution between upper and lower body radiotherapy, slight errors may cause signi cant hot and cold spots if conventional radiotherapy plans. Therefore, we designed a dose-drop scheme on both sides of adjacent region, which decreased from the prescribed dose (12 Gy) to 2 Gy within a length of 5 cm.
We can treat upper body rst with head rst position, but re-position was needed with feet rst position due to the limitation of the length of the linear accelerator treatment bed. For image guidance, kilo-voltage on broad imaging was used to collect images at anterior and right lateral directions for head and neck, abdomen, pelvis and lower limbs, and right lateral oblique directions for thorax to avoid obstruction by the arms. Online matching of the images with digital reconstruction radiograph (DRR) from the planning CTs were performed. Radiotherapy was permitted only if the senior physician and senior therapist con rm that the position error in each direction was less than 2 mm. Treatment team will monitor the whole process with audio and video.
Quality assurance Before radiotherapy, dose veri cation was performed on each isocentric rapid arc or AP-PA plan and each radiation elds using ScandiDos Delta4PT 3D QA phantom. PTW Octavius 4D QA phantom was used for Gamma passing rate (3 mm/ 3% Gamma criteria) at intersection of each two isocentric rapid arc plans of upper body. Point dose deviation at bilateral lower limbs and the junction of upper body and lower body were veri ed by a 60 cm(length) X 30cm(width) X 10 cm(height) solid water phantom, PinPoint 0.015 cc and UNDOSE electrometer. During the process of each fraction radiotherapy, total eight interest point sites dose monitoring were needed using MOSFET. Points of interest include forehead, bilateral chest, navel, perineum, bilateral knee and unilateral foot. When plans were recalculated with longitudinal misalignments, DVH parameters of PTV_total and PTV_crop such as Dmax and Dmax difference(%), Dmean and Dmean difference(%), V10.8 and V10.8 difference(%) were obtained respectively. Table 1 showed that Dmean and V10.8 are only slightly different even with an error of 5-20 mm longitudinal misalignments. Considering that the positioning error of OBI veri cation can be controlled smaller than 2-5 mm, the change of Dmax is also acceptable.  Table 2 showed dose veri cation of each isocentric rapid arc or AP-PA plan and each radiation elds were in compliance with the requirements perfectly. Gamma passing rates (3 mm/ 3% Gamma criteria) at intersection of each two isocentric plans of upper body were 98.4% and 95.9% respectively, the standard requirements have been well passed. Point dose deviation at bilateral lower limbs and the junction of upper body and lower body were from 1.5-3.38%, the values were also meet the standard requirements. During the process of each fraction radiotherapy, total eight interest point sites dose monitoring were done, then all the isocentric accumulative doses were added and analyzed, the results showed an overall deviation of -3.9%±5.51%. Discussion 5,6 TBI is mainly used for leukemia, followed by malignant lymphoma, myelodysplastic syndrome, multiple myeloma and other malignant diseases. It is also used for benign diseases such as aplastic anemia. [7][8][9][10][11] Outcomes after fractionated TBI were superior as compared with chemotherapy based conditioning with regard to overall survival, leukemia-free survival, relapse incidence, and non-relapse mortality. TBI in conjunction with chemotherapeutic agents has proven useful for eradicating residual malignant cells and for immunosuppression before HSCT. Unique features of TBI that make it a valuable component of transplant preparative regimens include: 1. No sparing of "sanctuary" sites such as testes and the central nervous system. 2. Dose homogeneity to the whole body regardless of blood supply. 3. Less chance of cross-resistance with other antineoplastic agents (chemotherapy). 4. No problems with excretion or detoxi cation. 5. Ability to tailor the dose distribution by shielding speci c organs or by "boosting" sites. It is essential that the complicated treatment and care of the patient receiving TBI be well coordinated among the various subspecialties (medical oncology, radiation oncology, etc.) and caregivers (physicians, nurses, physicists, therapist psychologists, dieticians, etc). 12 Most centers use opposing anterior and posterior elds with the patient standing upright several meters from the source and the beam pointed horizontally, thickness variations have less effect on dose homogeneity for This approach; however, the typical standing positioning can be strenuous with poor tolerance. 13 Patients also can be irradiated with lateral elds in a sitting or partly reclining position with better tolerance, but variations in patient thickness can cause large dose heterogeneities, especially if compensators are not utilized during treatment. Most importantly, these above conventional large-eld techniques require an extended source-to skin distance (SSD) that may not be available in standard-sized Linear accelerator room. 14  planned IMRT based TBI technique was developed to improve upon these shortcomings, 15,16 a bene t could be demonstrated with regard to dose distribution and homogeneity and the selective dose-reduction to organs at risk. 17 Patients taller than 120 cm cannot be treated in one position due to the limited cranial-caudal couch shift capacities of the linac. Therefore, patients are usually turned from a head-rst supine position (HFS) to a feet-rst supine position (FFS) to overcome this limitation. Losert C et al showed a newly developed rotatable tabletop consists completely of carbon ber, including the ball bearing within the base plate of the rotation unit. The patient can be turned 180° from a HFS to a FFS position within a few seconds, without the need of repositioning. 18 Treatment plans with an indexed rotational immobilization system had multi-isocentric volumetric modulated arc therapy (VMAT) beams to the upper body and parallel opposed elds to the lower body, with a 12 Gy prescription dose to > 90% of the body and mean lung dose ∼8 Gy. In the end-to-end test, point dose measurements had < 10% error. Compared to conventional TBI, the VMAT-based TBI technique increased the mean dose to the body by ∼1.0-1.5 Gy and decreased the mean dose to the lung by ∼1.0-1.5 Gy. The main problem of this approach is additional use of the special rotatable tabletop at the linear accelerator, it is not available in most hospital around the world. 16 Bora Tas et al introduced total-body irradiation using linac-based volumetric modulated arc therapy, high-dose junction regions were eliminated after the registration of two CT sets via bias-dose properties of Monaco 5.11 TPS, Plan adaptation delivery while ensuring OAR tolerances never exceeded due to biasdose planning because TPS considers dose distribution from previous plans while optimizing the cumulative dose distribution, but when plans were recalculated with each 3.0-mm, 6.0-mm, 9.0-mm, and 12.0-mm longitudinal misalignments, an average of 2.0% ± 0.7, 3.7% ± 1.2, 6.5% ± 1.7 and 7.2% ± 1.7 higher mean lung doses; 2.6% ± 1.

Declarations
Funding and Competing interests