Advertisement

Factors Affecting Isocenter Displacement and Planning Target Volume Margin in Rectal Cancer Patients Receiving Radiotherapy

  • Reham Mohamed
    Correspondence
    Corresponding Author: Dr. Reham Mohamed, MD, Kaser Alaini street, Department of Radiotherapy and Nuclear Medicine, National Cancer Institute, Cairo University, Cairo, Egypt
    Affiliations
    Department of Radiotherapy and Nuclear Medicine, National Cancer Institute, Cairo University, Cairo, Egypt

    Department of Radiation Oncology, Comprehensive Cancer Center, King Fahad Medical City, Riyadh, Saudi Arabia
    Search for articles by this author
  • Abousaleh Abousaleh Elawadi
    Correspondence
    Autor Responsible for Statistical Analysis: A.A. Elawadi, Medical Physics Department, Comprehensive Cancer Center, King Fahad Medical City, Riyadh, Saudi Arabia
    Affiliations
    Medical Physics Department, Comprehensive Cancer Center, King Fahad Medical City, Riyadh, Saudi Arabia

    Clinical Oncology and Nuclear Medicine Department, Faculty of Medicine, Mansoura University, Mansoura, Egypt
    Search for articles by this author
  • Nwaf Alkhanein
    Affiliations
    Department of Radiation Oncology, Comprehensive Cancer Center, King Fahad Medical City, Riyadh, Saudi Arabia
    Search for articles by this author
  • Muslihah Alharth
    Affiliations
    Radiation Therapy Section, Radiation Oncology Department, Comprehensive Cancer Center, King Fahad Medical City, Riyadh, Saudi Arabia
    Search for articles by this author
  • Mushabbab Asiri
    Affiliations
    Department of Radiation Oncology, Comprehensive Cancer Center, King Fahad Medical City, Riyadh, Saudi Arabia
    Search for articles by this author
Open AccessPublished:September 01, 2022DOI:https://doi.org/10.1016/j.adro.2022.101060

      Abstract

      Background

      : Setup errors are inherent in the process of daily radiation therapy (RT) delivery. Pelvic RT for rectal cancer is one of the body sites associated with the largest shift among other body sites. This study aimed to evaluate inter-fraction random and systematic errors and hence propose the optimum planning target volume (PTV) in rectal cancer patients.

      Patients and methods

      : Translational and angular isocenter displacements were retrospectively collected for 189 patients. Random and systematic errors were determined, then the PTV margin was computed. Effect of positioning, body mass index (BMI), and type of immobilization were studied. Portal images before and after online correction were used to define PTV for daily image-guided radiotherapy (IGRT) and no-daily IGRT respectively.

      Results

      : Before the online correction, the systematic errors were 2.5 mm, 2.8 mm, and 3.0 mm for superior-inferior (SI), right-left (RL), and anterior-posterior (AP) directions respectively, compared to 2.1 mm, 1.7 mm, and 1.8 mm after online correction. The random errors were 6.2 mm, 7.4 mm, and 8.2 mm in SI, RL, and AP respectively before online correction, compared to 4 mm, 4.2 mm, and 4.5 mm after online correction. The recommended PTV margin was 0.7 cm and 1.0 cm for daily IGRT and no-daily IGRT respectively. The prone position and BMI > 30 kg/m2 warrant higher margins in no-daily IGRT cases; 1.2 cm and 1.4 cm respectively.

      Conclusion

      : The prone position, BMI > 30 kg/m2, and belly board device are associated with larger daily setup errors warranting higher PTV margins for no-daily IGRT; however, that can be avoided by using daily IGRT.

      Keywords

      ABBREVIATIONS:

      RT (Radiotherapy), BMI (Body mass index), PTV (Planning target volume), EPID (Electronic portal imaging device), SI (Superior-inferior), RL (Right-left), AP (Anterior-posterior), σ (Random error), Σ (Systematic errors), IGRT (Image guided radiotherapy), IMRT (Intensity-modulated radiotherapy), VMI (Volumetric modulated arc therapy), V45 (The volume receiving 45 Gy), ASTRO (American society of radiation oncology), IRB (Institutional review board), Kv (Kilovoltage), OBI (Onboard imaging), DRRs (Digitally reconstructed radiographs), CBCT (Cone-beam computed tomography), TVE (Total vector error), i (m, Patient, mean), SD (Standard deviation), 3D-CRT (3-dimensional conformal radiotherapy), CTV-PTV (Clinical target volume - Planning target volume), SBRT (Stereotactic body radiotherapy)

      Background

      Precise and reproducible daily placement of treatment isocenter has been the main target for radiotherapy (RT) since its announcement as a medical discipline by Henry Coutard during the International Congress of Oncology in Paris in 1922 (1-2). Fixation aids, planning target volume (PTV), and electronic portal imaging devices (EPIDs) are the commonly used strategies to deal with the uncertainty of daily setup.
      The overall lifetime risk of developing colorectal cancer is 4%, and it is considered the third most diagnosed cancer worldwide (3). Radiotherapy is recommended for stage II-III rectal cancer patients as a neoadjuvant concurrent with chemotherapy. The treatment is delivered either as a short course with a dose of 25Gy/5fx/1w (4-5), or a long course with a dose of 50-50.4 Gy/25-28fx/5-5.5w (6-10). Pelvic irradiation for rectal cancer is associated with large setup errors that are highly affected by body mass index (BMI) and treatment position.
      Prone and supine positions are commonly practiced for rectal cancer RT, with no final agreement on the superiority of one over another (11-15). As the rectum is a posterior pelvic structure, the prone position is preferred by many centers to decrease bowel volume inside the RT field. On the other hand, others believe in the―more comfortable―supine position, especially with the use of advanced RT techniques that allow sparing of the small bowel. Intensity-modulated radiotherapy (IMRT) or volumetric modulated arc therapy (VMAT) can achieve the constraint of small bowel volume receiving 45Gy to be less than 195 cm3 (V45 ≤ 195 cm3).
      The belly board and Vac-Lok are commonly used with the prone position. The idea of using a belly board is to allow placement of the small bowel away from target volume and radiation beams in addition to providing a more comfortable positioning for obese patients.
      Patients with high BMI are difficult candidates for accurate daily positioning, as skin tattoos are mobile and the weekly EPIDs are not enough to correct positioning errors (16).
      The American Society of Radiation Oncology (ASTRO) recommends prone positioning with a belly board device for pelvic RT and expects emerging evidence shortly (17). Also, 90% of the panel―with a lack of clear literature evidence―recommends daily image-guided radiotherapy (IGRT) in addition to IMRT/VMAT technique.
      The frequency of treatment verification using EPID differs from department to department based on the workload and even from time to time in the same department. The logical assumption of having a different PTV margin for daily IGRT compared to no-daily IGRT is valid for our targeted cohort of patients. The studies that have investigated this issue are still not enough to standardize the practice. Accordingly, analysis of setup errors during RT for rectal cancer patients is of major concern

      Aim of the study

      This study evaluated translational and rotational displacements before and after online correction for cancer rectum patients receiving RT. Different factors like treatment position, body mass index, and fixation aids were studied and correlated to the setup variations. Systematic errors, random errors, and the recommended PTV margin were computed.

      Patients and methods

      Upon institutional review board (IRB) approval, the daily portal images―before and after online correction―for rectal cancer patients who received long-course RT were retrospectively reviewed. The isocenter displacement in the superior-inferior (SI), right-left (RL), and anterior-posterior (AP) directions were collected. The rotational displacement was also collected.
      Per departmental policy, rectal cancer patients were advised to have an empty rectum and full bladder before CT simulation and daily treatment. The treatment positions practiced at our department included prone and supine as per physician preference and patient capability. The use of a belly board device, Vac-Lok cushion, or no fixation was discussed between the radiation oncologist and the radiation therapist and decided upon before CT simulation for each patient.

      Departmental Radiation Therapy Verification Protocol for Rectal Cancer Patients

      Kilo-voltage (Kv) portal images using the onboard imaging (OBI) system were acquired. Online matching with digitally reconstructed radiographs (DRRs) ―created from CT simulation images― was followed. Isocenter shift correction was applied as per departmental policy. A displacement of 3 mm shift is accepted without correction. The shift of 3 to 7 mm mandates applying shift correction before treatment. Re-setup is required for > 7 mm translational shift and > 3 rotational displacement. Another portal image was acquired after correction for documentation.

      Frequency of Portal Images

      The portal images were acquired on the first three consecutive days of treatment and then twice weekly for each patient. Cone-beam CT (CBCT) was used for treatment verification at least once weekly. The trend of isocenter shifts in the first three days for each patient was routinely evaluated by a senior radiation technologist to decide on either daily IGRT or no-daily IGRT (keep the same portal image frequency).

      Estimation of Reproducibility

      The isocenter shift was computed using the Offline Review of ARIA®Radiation Therapy Management software system (ARIA RTM version: 16.0120). Auto-match of DRR and the portal images before and after the online correction was done to calculate the setup errors (Figure 1).
      Figure 1:
      Figure 1Matching of DRR with a portal image showing the isocenter shift for a patient with rectal cancer.
      The upper images represent the anterior and lateral Digitally Reconstructed Radiographs (DRRs), while the lower images represent the overlay view of DRRs and portal images. Bony structures (iliac bone, pelvic brim, sacral hollow, and pubic bone) were delineated by green color on DRRs and matched with KV portal images. Green coordinates represent the plan isocenter. The ocean coordinates represent the treatment isocenter for that day. The auto/manual matching generated table showed the isocenter shifts in Vrt (Anterior-Posterior), Lng (Superior-Inferior), Lat (Right-Left), and Rtn (Rotational) directions for that day.
      The translational setup errors, including the three directions, SI, RL, and AP, were calculated, and the total vector error (TVE) was computed for each patient. TVE reflects the overall treatment isocenter shift from the planned isocenter. It is a mathematical function that takes the three directions’ displacements into account simultaneously and computed for each patient (i) from the mean displacements (m):
      TVE(i)=mSI2+mRL2+mAP2


      The rotational displacement was calculated separately and measured in angular degrees. Systematic (Σ) and random (σ) errors were calculated according to Stroom (18) and Van Herk (19). Systematic error is the average setup variations of the target volume for all fractions for a certain patient. The random errors are unpredictable and unavoidable inter-fractional variations.
      A certain direction systematic error for each patient (i) is the standard deviation (SD) of the mean (m) displacement in that direction (SI, RL, or AP):
      ΣSI=SD(mi,SI)


      While a certain direction random error (σ) for each patient (i) is the square root of this direction displacement squared standard deviation:
      σSI=SDi,SI2


      We calculated PTV margins based on the suggested formula by Stroom (18). This formula ensures that 99% of the CTV volume receives 95% of the prescribed dose (V99% covered by ≥ 95% dose).
      Margin=2Σ+0.7σ


      Statistical Analysis

      The translational and rotational mean displacements for all patients were collected. The data were compared for treatment positions (supine vs. prone), BMI groups (≤ 30 kg/m2 vs. > 30 kg/m2), and fixation aids (Vac-Lok™ vs. belly board). Shapiro-Wilk's and Levene's tests were used to calculate the normal distribution of data and equal variances respectively. Two-tailed independent sample t-test was performed to compare the data between different groups. P-value ≤ 0.05 was considered significant.
      The systematic error and random errors were calculated, and then PTV margins were computed using Stroom's formula.
      Matching images before online correction was used to assess the recommended PTV margin for no-daily IGRT treatment. On the other hand, matching images after online correction was used to assess the PTV margin for daily IGRT treatment. The PTV margins for the different subset of patients as per treatment position, BMI, and fixation aids were computed and studied.

      Results

      One hundred eighty-nine rectal cancer patients―treated between 2011 and 2021―were included in the final analysis. We excluded patients treated by short-course RT or due to unavailability of images of both before and after online corrections.

      Patients’ characteristics

      The mean age was 54±15 years (range: 19-93 years). Sixty-five patients (34%) were females, and 124 (66%) were males. The mean BMI was 27±5 kg/m2 (range 13-44.5 kg/m2). Underweight patients having BMI under 18.5 kg/m2 were 10% of the studied group, and normal-weight patients having BMI between 18.5 to 24.9 kg/m2 were 26%. Overweight patients with BMI between 25 to 29.9 kg/m2 were 36% of the studied group of patients. The obese group having BMI > 30 kg/m2 was 53 patients, representing 28% of the studied group, as shown in table 1.
      Table 1Patients’ characteristics
      VariableTotal Number n=189Percent %
      Sex
      • Male
      • Female


      124

      65


      66

      34
      Position
      • Supine
      • Prone


      110

      79


      58

      42
      Immobilization
      • Belly board
      • Vac-Loc
      • None


      50

      20

      119


      26

      11

      63
      BMI grouping
      • ≤ 20
      • 20-25
      • 26-30
      • 31-35
      • > 35


      19

      49

      68

      42

      11


      10

      26

      36

      22

      6
      Treatment technique
      • 3D-CRT
      • VMAT


      63

      126


      33

      67

      Number of verified RT sessions2345 Average (12.4 session/patient) Range (5-23 session)

      Patient age (mean ±SD)54±15 (Range 19-93)
      BMI (mean ±SD)27±5 (Range 13-44.5)
      BMI= body mass index, VMAT= volumetric modulated arc therapy, RT= radiotherapy
      One hundred and ten patients (58%) were treated in the supine position, while 42% were treated in the prone position. The belly board device was used for 50 patients (26%), and Vac-Lok was used for 20 patients (11%). All patients treated using the belly board and Vac-Lok were lying in the prone position.
      The IMRT/VMAT technique was used for treating 126 patients, representing 67% of the whole studied group, while 33% were treated with 3-dimensional conformal radiotherapy (3D-CRT).
      The total number of verified sessions was 2,345 (52%) out of 4,515 treated sessions. The average verified sessions per patient was 12.4 (range 5-25 sessions). The verification images included megavoltage images (Mv), Kv, and CBCT images.

      Setup errors before and after online correction

      The mean translational, angular, and TVE displacements for the whole group of patients and different subgroups before and after online correction are shown in table 2. The mean SI displacement was 0.11±0.25 cm before online correction, compared to 0.06±0.21 cm after online correction.
      Table 2Mean isocenter shift before and after online correction and the effect of different variables.
      Displacement DirectionVariableNDisplacement before online correctionDisplacement after online correction
      Mean ± SD in cmP valueMean ± SD in cmP value
      SI

      Fixation AidBB500.12±0.140.10.15±0.160.8
      Vac Loc200.07±0.120.14±0.11
      PositionSupine1100.12±0.30.9-0.01±0.230.06
      Prone790.11±0.120.14±0.15
      BMI≤ 30 kg/m21360.11±0.260.830.06±0.210.9
      > 30 kg/m2530.11±0.240.06±0.22
      All patients1890.11±0.250.06±0.21
      RLFixation AidBB500.24±0.250.0050.11±0.140.8
      Vac Loc200.07±0.20.13±0.16
      PositionSupine1100.06±0.30.001-0.06±0.170.01
      Prone790.22±0.260.12±0.15
      BMI≤ 30 kg/m21360.10±0.280.020.01±0.180.5
      > 30 kg/m2530.21±0.270.03±0.19
      All patients1890.13±0.280.02±0.18
      APFixation AidBB500.12±0.200.60.11±0.120.1
      Vac Loc200.09±0.20.06±0.14
      PositionSupine110-0.14±0.320.0010.01±0.210.001
      Prone790.12±0.200.11±0.13
      BMI≤ 30 kg/m2136-0.02±0.280.050.06±0.170.3
      > 30 kg/m253-0.06±0.380.03±0.21
      All patients189-0.03±0.310.05±0.18
      RotationFixation AidBB501.5±0.140.071.5±1.00.06
      Vac Loc202.0±0.92.0±0.8
      PositionSupine1100.01±0.30.0010.01±0.420.001
      Prone791.6±0.91.63±0.95
      BMI≤ 30 kg/m21360.60±1.000.080.60±1.000.08
      > 30 kg/m2530.89±1.130.91±1.11
      All patients1890.7±1.00.6±1.0
      TVEFixation AidBB500.42±0.20.030.30±0.130.6
      Vac Loc200.32±0.150.29±0.12
      PositionSupine1100.48±0.30.050.33±0.150.21
      Prone790.40±0.20.30±0.12
      BMI≤ 30 kg/m21360.43±0.250.050.31±0.130.4
      > 30 kg/m2530.51±0.260.33±0.16
      All patients1890.45±0.260.32±0.14
      SI=superior inferior, RL=right left, AP= anterior posterior, BB= belly board, BMI= body mass index, TVE=total vector error.
      The mean RL displacement was 0.13±0.28 cm before online correction, compared to 0.02±0.18 cm after online correction. The mean AP displacement was -0.03±0.31 cm, compared to 0.05±0.18 cm before and after online correction respectively. Supplements 1 and 2 show the clear difference between SI, RL, and AP displacements before and after online correction.
      The mean rotational displacement was 0.7±1.0 before online correction compared to 0.6±1.0 after online correction, as shown in table 2.
      The mean TVE was higher (0.45±0.26 cm) before online correction compared to (0.32±0.14 cm) after online correction, as shown in figure 2.
      Figure 2:
      Figure 2Patients’ isocenter total vector error before and after online correction
      The figure shows total vector error (TVE) difference of treatment isocenter for all patients before (Fig 2A) and after (Fig 2B) online correction.
      The patients treated in the supine position showed significantly lower mean RL, AP, rotational, and TVE displacement before online correction compared to those treated in the prone position, with a p-value of 0.001, 0.001, 0.001, and 0.05 respectively. The same is valid after online correction, except for TVE, which did not show a significant difference between the supine and prone positions, with a p-value of 0.21, as shown in table 2.
      The patients with BMI > 30 kg/m2 showed significantly higher mean RL, AP, and TVE displacement before online correction compared to patients with BMI ≤ 30 kg/m2, with a p-value of 0.02, 0.05, and 0.05 respectively. However, there was no significant difference after online correction between the same groups, as shown in table 2.
      The belly board device showed higher mean RL and TVE displacement before online correction compared to Vac-Lok, with a p-value of 0.005 and 0.03 respectively. However, these differences were lost after online correction, as shown in table 2.

      Systematic, random errors and recommended PTV margins

      The systematic and random errors were computed as per Stroom et al. (18). The SI, RL, and AP random errors before online correction were 6.2 mm, 7.4 mm, and 8.2 mm respectively, compared to 4 mm, 4.2 mm, and 4.5 mm after online correction. The SI, RL, and AP systematic errors before online correction were 2.5 mm, 2.8 mm, and 3.0 mm respectively, compared to 2.1 mm, 1.7 mm, and 1.8 mm after online correction.
      The clinical target volume to planning target volume (CTV-PTV) margin was computed as per Stroom et al. (18). The recommended PTV margin for the patients to be treated with no-daily IGRT was 0.9 cm for the SI direction, 1.0 cm for the RL direction, and 1.1 cm for the AP direction. On the other hand, the PTV margin for patients to be treated with daily IGRT was 0.7 cm for the SI direction, 0.66 cm for the RL direction, and 0.68 cm for the AP direction (table 3).
      Table 3Recommended PTV margins for daily IGRT and no-daily IGRT with the effect of different variables
      VariableNPTV margin in mm for no-daily IGRTPTV margin in mm for IGRT
      SIRLAPSIRLAP
      PositionSupine1109.18.99.07.15.76.4
      Prone798.612.212.46.26.76.5
      Fixation AidBelly Board508.711.912.26.56.66.4
      Vac Loc208.211.613.55.36.87.0
      Body Mass Index≤ 30All1369.310.410.87.06.46.3
      Supine878.98.47.86.95.55.8
      Prone498.612.212.46.26.76.5
      > 30All539.711.414.37.47.17.8
      Supine239.710.211.47.76.47.9
      Prone308.811.912.56.56.86.7
      All rectal cases1899.010.011.07.06.66.8
      SI=superior inferior, RL=right left, AP= anterior posterior, IGRT=image guided radiotherapy.
      For treatment without daily IGRT, the required PTV margin for patients treated in the supine position was 0.9 cm, compared to 1.2 cm for those treated in the prone position. Also, patients with BMI > 30 required a PTV margin of 1.4 cm, compared to 1.0 cm for patients with BMI ≤ 30. For treatment with daily IGRT, the previous differences were lost, and all patients could be treated with a PTV margin of 0.7 cm.
      The fixation aid did not affect the PTV margin for daily IGRT and no-daily IGRT treatments (table 3).
      The random and systematic errors were higher before online correction compared to after online correction irrespective of the treatment position. Interestingly, the systematic errors were higher for the supine position compared to the prone position in contrast to random errors (table 4).
      Table 4Comparison of our data with other published studies
      Study (Total number of patients)Treatment Site (No of Pts per site)PositionFixation AidSystematic errors Σ (mm)Random errors σ (mm)
      APLRSIAPLRSI
      Kasabasic (23) 11 PatientsRectum (1)ProneBB9122.4172218
      Uterus (4)ProneBB
      Cervix (6)ProneBB
      Tamponi (24) 100 Pts

      Rectum (8)ProneBB1.80.41.62.51.52.8
      Uterus (9)Supine-3.22.21.922.12.3
      Prostate (16)Supine-1.71.31.51.81.91.8
      Thasanthan (25)Rectum (50)--2.73.32.62.31.61.6
      Rajeev et al (26)

      Rectum (20)ProneBB1.30.61.22.01.23.1
      Supine-0.90.71.61.81.71.8
      Bouchra et al (27) 44 patientsCervixSupineFF/ UK1.42.01.21.72.91.3
      Rectum1.31.92.91.32.31.0
      Bansal et al (28)Rectum (7)proneTTM4.71.22.19.62.32.0
      Before correction (Present study)Rectum (189)SupineFF/UK3.22.83.13.64.84.0
      ProneBB/VL2.02.61.412.09.98.3
      AllFF,UK,BB,VL32.82.58.27.46.2
      After correction (Present study)Rectum (189)SupineFF/UK2.11.72.33.23.23.5
      ProneBB/ VL/no1.31.51.55.75.44.6
      AllFF,UK,BB,VL1.81.72.14.54.24
      BB = belly board, FF = foot fix, UK = under knees, TTM=thermoplastic mask, VL = Vac-Loc

      Discussion

      In this study, we retrospectively reported the setup error for 189 rectal cancer patients who underwent RT. Most pelvic RT setup error studies include a diversity of diseases; however, we believe that each one warrants a dedicated study due to different setups, types of patients, and disease factors. For rectal cancer patients, immobilization devices widely used for the sake of reproducibility and decreasing irradiated bowel volume failed to reduce setup errors and affected the needed PTV margin (20-22). Also, the increasing trend of using VMAT for those cases mandates examining the isocenter displacement and the recommended PTV margin precisely. The logical assumption of having a lower PTV margin for daily IGRT cases compared to no-daily IGRT cases is valid but not standardized yet due to less evidence in the literature.
      Studying portal images before online correction (first taken images) and after online correction for the same patients made the comparison of the two data sets homogenous enough to assess the difference properly. The high number of patients in our study compared to other studies lends more validity to this study's results.
      As most of the studies are concerned with gynecological and prostate cancer patients, we tried our best to highlight only rectal cancer studies to compare to our study, as shown in table 4.
      Kasabasic et al. reported 11 patients with pelvic malignancy, including rectum cases, and showed higher systematic and random errors compared to our data before online correction. They reported systematic errors ranging from 2.4 mm to 12 mm, while random errors reached up to 18 mm. They recommended margins of 11 mm, 13 mm, and 14 mm in the RL, SI, and AP directions respectively, which is nearly the same as our recommended margins for the prone setup position using a belly board device without IGRT (23).
      Tamponi et al. showed systematic and random errors of 2 to 3 mm in the prostate, rectum, and gynecological cancer patients receiving RT. The calculated CTV–PTV margins were 10 mm in the RL direction and 20 mm in the SI direction. In comparison, our setup errors were lower than these numbers, and the recommended PTV margins for non-IGRT were within 10 mm, while for IGRT they were within 6 to 7 mm (24).
      Thasanthan et al. studied 50 cancer rectum patients with portal images taken on the first two days of treatment only. They used the corrected images for the assessment of isocenter shifts and showed mean AP, SI, and RL displacements of 1.0 mm, -1.8 mm, and 0.8 mm respectively. The systematic errors reported in this study were higher compared to ours, in contrast to the random errors, which were lower. The systematic errors were nearly 3 mm in all directions, while the random errors were 2.3 mm, 1.6 mm, and 1.6 mm for the AP, SI, and RL directions respectively. They recommended PTV margins of 0.84 cm for AP, 0.9 cm for SI, and 0.76 cm for RL directions. They concluded that the routinely used PTV margin of 0.5 cm was not enough for approximately 22% of treatment sessions (25).
      Rajeev et al. studied 20 rectal cancer patients and compared the supine and prone positions. Systematic errors for the supine position were 0.87 mm for the AP direction, 0.66 mm for the RL direction, and 1.6 mm for the SI direction. These values were lower than our current study. The prone position's systematic errors were 1.3 mm, 0.59 mm, and 1.17 mm for the AP, RL, and SI directions respectively. They reported random errors of 1.81 mm (AP), 1.73 mm (RL), and 1.83 mm (SI) for the supine position, compared to 2.02 mm (AP), 1.21 mm (RL), and 3.05 mm (SI) for the prone position. Consequently, the recommended PTV margins were 3.45 mm, 2.87 mm, and 5.31 mm for the AP, RL, and SI directions respectively in the case of the supine position. The prone position's recommended margins were 4.9 mm, 2.3 mm, and 5.1 mm for the AP, RL, and SI directions respectively. The non-practical recommended PTV margins of 2 mm in some directions was a result of far low random error values in this study. We think that this study followed a strict protocol for online correction and that they used the confirmation image after this correction for measuring the setup errors; however, it is not clearly stated (26).
      Bouchra et al. studied 44 patients, including 12 rectal cancer patients treated in the supine position. The reported mean displacements were nearly 3.9 mm in all directions. The systematic errors proved to be 1.3 mm, 1.9 mm, and 2.9 mm for the AP, RL, and SI directions respectively, while the random errors were 1.25 mm, 2.33 mm and 1.04 mm for the AP, RL, and SI directions respectively. Bouchra et al. recommended PTV margins of 8 mm for the RL direction and 5 mm for the SI and AP directions based on their clinical routine work (27).
      Bansal et al. studied seven rectal cancer patients treated in the prone position using a thermoplastic mask. The prone position's systematic and random errors were nearly comparable to our data, apart from the higher value of the SI direction, which could be related to the use of the thermoplastic mask. PTV margins calculated were found to be 0.5 cm, 1.8 cm, and 0.7 cm in the lateral, longitudinal, and vertical directions respectively (28).
      In summary, our data showed a significant statistical difference between the supine and prone positions, with superiority of the supine position regarding reproducibility and higher margins recommended for the prone position. Our results matched Adli et al., who studied gynecological tumors. They concluded that the valuable effect of the prone position on reducing the irradiated bowel volume outweighed its setup uncertainties (29).
      The inter-fractional rectal and bladder filling variation was not studied in our patients due to the unavailability of CBCT for all of them. However, the previous point is well-studied by others for prostate stereotactic body radiation therapy (SBRT), with contradictory results. Byun et al. studied that in 85 patients with prostate carcinoma who received SBRT with 510 acquired CBCT. Patients were instructed to maintain a full bladder and empty rectum before simulation and daily RT. They reported a difference in the bladder and rectal volumes at the time of treatment compared to the planning scans. They doubt the need for excessively strict bladder filling and rectal emptying protocols in the context of IGRT-SBRT (30).
      Studies endorsing rectal RT with daily CBCT-IGRT are needed to determine if the daily variation of bladder and rectal filling will affect the PTV margins. However, we can extrapolate cautiously from Byun et al.’s study that the effect of this factor will be small with the current practice of bladder filling and rectal emptying.

      Conclusion

      The computed PTV margins for daily IGRT are lower compared to no-daily IGRT. Prone position and BMI > 30 kg/m2 require a higher PTV margin in the case of no-daily IGRT; however, this is not the same with the use of daily IGRT. The use of immobilization devices like belly boards should be studied by each department to find the optimal PTV margins for their use.
      We recommend using daily IGRT for rectal cancer radiation treatment and applying a PTV margin of 7 mm. However, in the case of no-daily IGRT, the margins should be 1.0 cm to 1.4 cm based on the BMI, treatment position, and fixation tool. We encourage further studies examining the effect of bladder and rectal filling on the recommended PTV margins.

      Funding

      None.

      Availability of Data and Material

      Research data are stored in our institutional repository and will be shared upon request with the corresponding author.

      Ethics Approval and Consent to Participate

      This study was approved Institutional Review Board at King Fahad Medical City, Riyadh Saudi Arabia (IRB 22-029 ) and registered with OHRP/NIH , USA: IRB00010471 and approval number Federal Wide Assurance NIH, USA : FWA00018774 . I confirm that all steps of scientific research were performed by relevant guidelines and regulations.

      Consent for Publication

      Consent for the publication of personal or clinical details of participants: Not Applicable.

      Authors' Contributions

      RM designed the study, reviewed the data, reviewed statistics, evaluated the results, edited the manuscript, and prepared the manuscript for publishing. AAE reviewed data collection and statistical analysis. Performed the computation of the margins and contributed to the drafting of the final manuscript. NAl, MAl performed the data collection. MA reviewed the results and final manuscript.

      Type of the study

      Retrospective study

      REFERENCES

      • 1
        Halperin EC, Perez CA, Brady LW. The discipline of radiation oncology. In: Halperin EC, Perez CA, Brady LW, editors. Perez and Brady's Principles and Practice of Radiation Oncology. Philadelphia: Wolters Kluwer; 2018. p. 2–6.
        Newhauser WD, Berrington de Gonzalez A, Schulte R, Lee C. A review of radiotherapy-induced late effects research after advanced technology treatments.  Front Oncol. 2016;6:13. doi:10.3389/fonc.2016.00013.
      • 2
      • 3
        Ngan SY, Burmeister B, Fisher RJ, Solomon M, Goldstein D, Mackay J, et al. Randomized trial of short-course radiotherapy versus long-course chemoradiation comparing rates of local recurrence in patients with T3 rectal cancer: Trans-Tasman Radiation Oncology Group trial 01.04. J Clin Oncol. 2012; 30 (31):3827-3833
      • 4
        Bujko K, Nowacki MP, Nasierowska-Guttmejer A, Michalski W, Bebenek M, Kryj M. Long-term results of a randomized trial comparing preoperative short-course radiotherapy with preoperative conventionally fractionated chemoradiation for rectal cancer. Br J Surg. 2006; 93:1215-1223.
      • 5
        Sauer R, Liersch T, Merkel S, Fietkau R, Hohenberger W, Rodel C, et al. Preoperative versus postoperative chemoradiotherapy for locally advanced rectal cancer: results of the German CAO/ARO/AIO-94 randomized phase III trial after a median follow-up of 11 years. J Clin Oncol. 2012;30(16): 1926-1933.
      • 6
        Park JH, Yoon SM, Yu CS, Kim JH, Kim TW, Kim JC. Randomized phase 3 trial comparing preoperative and postoperative chemoradiotherapy with capecitabine for locally advanced rectal cancer. Cancer. 2011; 117:3703-3712.
      • 7
        De Caluwe L, Van Nieuwenhove Y, Ceelen WP. Preoperative chemoradiation versus radiation alone for stage II and III resectable rectal cancer. Cochrane Database Syst Rev. 2013;2:CD006041. doi: 10.1002/14651858.CD006041
      • 8
        Fiorica F, Cartei F, Licata A, et al. Can chemotherapy concomitantly delivered with radiotherapy improve survival of patients with resectable rectal cancer? A meta-analysis of literature data. Cancer Treat Rev. 2010; 36:539-549.
      • 9
        Lefevre JH, Mineur L, Kotti S, et al. Effect of interval (7 or 11 weeks) between neoadjuvant radiochemotherapy and surgery on complete pathologic response in rectal cancer: a multicenter, randomized, controlled trial (GRECCAR-6). J Clin Oncol. 2016;34: 3773-3780.
      • 10
        Wee CW, Kang HC, Wu HG, Chie EK, Choi N, Goodman KA, et al. Intensity-modulated radiotherapy versus three-dimensional conformal radiotherapy in rectal cancer treated with neoadjuvant concurrent chemoradiation: a metaanalysis and pooled-analysis of acute toxicity. Jpn J Clin Oncol. 2018; 48 (5):458-466.
      • 11
        Droge LH, Weber HE, Guhlich M, Leu M, Conradi LC, Woff HA, et al. Reduced toxicity in the treatment of locally advanced rectal cancer: a comparison of volumetric modulated arc therapy and 3-D conformal radiotherapy. BMC Cancer. 2015; 15:750.
      • 12
        Kim JY, Kim DY, Kim TH, Park SY, Lee SB Cho KH, et al. Intensity-modulated radiotherapy with a belly board for rectal cancer. Int J Colorectal Dis. 2007;22(4): 373-379.
      • 13
        Scobioala S, Kittel C, Niermann P, Wolters H, Susek KH, Eich HT, et al. A treatment planning study of prone vs. supine positions for locally advanced rectal carcinoma: comparison of 3-dimensional conformal radiotherapy, tomotherapy, volumetric modulated arc therapy, and intensity-modulated radiotherapy. Strahlenther Onkol. 2018;194: 975-984.
      • 14
        Vajiheh V, Shahrookh N, Kazem A, Maryam N, Mehdi M. Evaluation of set-up errors and determination of set-up margin in pelvic radiotherapy by electronic portal imaging device (EPID). Radiotherapy in Practice.2020; 19 (2):150-156. DOI: https://doi.org/10.1017/S1460396919000566.
      • 15
        Laura EM, Michele A, Jean P, Katsuto S, Mack R. Daily electronic portal imaging for morbidly obese men undergoing radiotherapy for localized prostate cancer. Int J Radiat Oncol Biol Phys. 2004; 59 (1):6–10. doi: 10.1016/j.ijrob.2003.12.027.
      • 16
        Jennifer YW, Anker CJ, Ashman JB, Bhadkamkar NA, Bradfield L, Das P, et al. Radiation Therapy for Rectal Cancer: Executive Summary of an ASTRO Clinical Practice Guideline. Pract Radiat Oncol. 2021; 11(1): 13-25. DOI: https://doi.org/10.1016/j.prro.2020.08.004
      • 17
        Stroom JC, de Boer JC, Huizenga H, Visser AG. Inclusion of geometrical uncertainties in radiotherapy treatment planning by means of coverage probability. Int J Radiat Oncol Biol Phys 1999; 43(4):905–919.
      • 18
        Van Herk M, Remeijer P, Rasch C, Lebesque JV. The probability of correct target dosage-dose population histograms for deriving treatment margins in radiotherapy. Int J Radiat Oncol Biol Phys. 2000; 47(4):1121–1135.
      • 19
        Verhey L. Immobilizing and Positioning Patients for Radiotherapy. Semin Radiat Oncol. 1995; 5(2):100-114. doi: 10.1054/SRAO00500100.
      • 20
        Avinash HU, Shibina N, Nirmala S, Ravindra KB, Sandeep M. Immobilization versus no immobilization for pelvic external beam radiotherapy. Rep of Pract Oncol Radiother. 2018;23(4):233-241. https://doi.org/10.1016/j.rpor.2018.04.007
      • 21
        Kasabasic M, Faj D, Belaj N, Faj Z and Tomas I. Implementing of the offline setup correction protocol in pelvic radiotherapy: safety margins and number of images. Radiol Oncol. 2007;41(1):48-55. https://doi.org/10.2478/v10019-007-0008-7.
      • 22
        Kasabasic M, Faj D, Radojcic DS, Svabic M, Ivkovic A, Jurkovic S. Verification of the Patient Positioning in the Bellyboard Pelvic Radiotherapy. Coll Antropol. 2008; 32 (2): 211–215.
      • 23
        Tamponi M, Poggiu A, Dedola M, Madeddu R, Carnevale A, Meloni G, et al. Random and systematic set-up errors in three-dimensional conformal radiotherapy – impact on planning target volume margins: The experience of the Radiation Oncology Centre of Sassari. Journal of Radiotherapy in Practice. 2014; 13(2): 166-179. doi:10.1017/S1460396913000204
      • 24
        Thasanthan L, Piyasena WRO, Croos AMC, Dhanushia R, Narayanan PP. Assessment of Three-Dimensional Set-up Errors in Pelvic Radiation Therapy. J Cancer Prev Curr Res. 2014; 1(2): 00009. DOI:10.15406/jcpcr.2014.01.00009
      • 25
        Rajeev KR, Menon SS, Beena K, Holla R, Kumar RR, Dinesh M. A comparative study of set up variations and bowel volumes in supine versus prone positions of patients treated with external beam radiation for carcinoma rectum. J Cancer Res Ther. 2014;10(4):937-941. Doi:10.4103/0973-1482.138021. PMID: 25579532.
      • 26
        Bouchra A, Abdellah H, Dounia M, Slimane S. Evaluation of setup errors in conformal radiotherapy for pelvic tumors: Case of the Regional Center of Oncology, Agadir. Radiation Medicine and Protection. 2020; 1: 99–102.
      • 27
        Bansal S, Bhattacharyya M, Kalita A K, Singh M N, Sunku R, Bora G, et al. Determination of Optimal Clinical Target Volume to Planning Target Volume Margins for Conformal Radiotherapy Planning using Image Guidance System in Rectal Cancer in Prone Position. J Med Phys; 2019; 44(1): 65-67.
      • 28
        Adli M, Mayr NA, Kaiser HS, Skwarchuk MW, Meeks SL, Mardirossian G, et al. Does prone positioning reduce small bowel dose in pelvic radiation with intensity‑modulated radiotherapy for gynecologic cancer? Int J Radiat Oncol Biol Phys. 2003; 57:230‑238.
      • 29
        David JB, Daniel JG, Lauren MJ, Laura H, Zhang P, Zelefsky MJ, et al. Strict bladder filling, and rectal emptying during prostate SBRT: Does it make a dosimetric or clinical difference? Radiat Oncol. 2020; 15:239. https://doi.org/10.1186/s13014-020-01681-6.

      Conflict of interest Statement for All Authors

      None (All authors have no conflicts of interest to disclose).

      Acknowledgments

      I would like to acknowledge the radiation therapy and medical physics teams at King Fahad Medical City for their valuable effort in supporting this study.

      Appendix. Supplementary materials