Variations in dosimetric distribution and plan complexity with collimator angles in hypofractionated volumetric arc radiotherapy for treating prostate cancer

Abstract Purpose Hypofractionated radiotherapy can reduce treatment durations and produce effects identical to those of conventionally fractionated radiotherapy for treating prostate cancer. Volumetric arc radiotherapy (VMAT) can decrease the treatment machine monitor units (MUs). Previous studies have shown that VMAT with multileaf collimator (MLC) rotation exhibits better target dose distribution. Thus, VMAT with MLC rotation warrants further investigation. Methods and materials Ten patients with prostate cancer were included in this study. The prostate gland and seminal vesicle received 68.75 and 55 Gy, respectively, in 25 fractions. A dual‐arc VMAT plan with a collimator angle of 0° was generated and the same constraints were used to reoptimize VMAT plans with different collimator angles. The conformity index (CI), homogeneity index (HI), gradient index (GI), normalized dose contrast (NDC), MU, and modulation complexity score (MCSV) of the target were analyzed. The dose–volume histogram of the adjacent organs was analyzed. A Wilcoxon signed‐rank test was used to compare different collimator angles. Results Optimum values of CI, HI, and MCSV were obtained with a collimator angle of 45°. The optimum values of GI, and NDC were observed with a collimator angle of 0°. In the rectum, the highest values of maximum dose and volume receiving 60 Gy (V60 Gy) were obtained with a collimator angle of 0°, and the lowest value of mean dose (Dmean) was obtained with a collimator angle of 45°. In the bladder, high values of Dmean were obtained with collimator angles of 75° and 90°. In the rectum and bladder, the values of V60 Gy obtained with the other tested angles were not significantly higher than those obtained with an angle of 0°. Conclusion This study found that MLC rotation affects VMAT plan complexity and dosimetric distribution. A collimator angle of 45° exhibited the optimal values of CI, HI, and MCSv among all the tested collimator angles. Late side effects of the rectum and bladder are associated with high‐dose volumes by previous studies. MLC rotation did not have statistically significantly higher values of V60 Gy in the rectum and bladder than did the 0° angle. We thought a collimator angle of 45° was an optimal angle for the prostate VMAT treatment plan. The findings can serve as a guide for collimator angle selection in prostate hypofractionated VMAT planning.

angle for the prostate VMAT treatment plan. The findings can serve as a guide for collimator angle selection in prostate hypofractionated VMAT planning. and-shoot intensity modulated radiotherapy (IMRT). In addition, some studies have revealed that prostate VMAT and IMRT exhibit comparable target coverage and normal tissue (bladder, rectum, and femoral heads) sparing. [1][2][3] In radiotherapy, the sensitivity of tumors to changes in fractionation can be quantified in terms of the a/b ratio. The a/b values for most human tumors are high (typically 10 Gy). Recent studies have suggested that adenocarcinomas of the prostate gland, with a low average a/b ratio of <2 Gy, differ from most other malignancies. 4,5 The treatment of tumors with low a/b ratios through hypofractionated IMRT with a high dose per fraction requires a short duration and exhibits efficacy and toxicity levels similar to those of conventionally fractionated IMRT. [6][7][8] However, increasing the dose per fraction requires a higher number of MUs and a longer treatment duration per fraction.
For the same treatment plan, hypofractionated VMAT requires fewer MUs and a shorter treatment time per fraction than does hypofractionated IMRT and therefore is a viable, safe, and comfortable treatment technique for prostate cancer. 9 VMAT technology simultaneously combines gantry rotation speed, multileaf collimator (MLC) motion, and dose rate modulation.
In general, the complex target shapes, volumes of targets, and multiple prescribed dose levels require the use of two or more VMAT arcs to improve dosimetric distribution. 10,11 MLC is the most suitable tool for beam shaping and is designed to have a tongue-and-groove shape on the side of each leaf for reducing interleaf radiation leakage. However, transmission through the leaves remains nonuniform; thus, MLC rotation in the VMAT can minimize interleaf radiation leakage. In addition, several studies have reported that MLC rotation improves spatial resolution and target dose distribution. [12][13][14][15][16][17] In the present study, the dosimetric distribution and plan complexity obtained using various collimator angles (0°, 15°, 30°, 45°, 60°, 75°, and 90°) for dual-arc hypofractionated regimens of VMAT with simultaneous integrated boost VMAT (SIB-VMAT) in patients with prostate cancer. This study identified the optimum collimator angles for optimizing dosimetric distribution for planning target volume (PTV), sparing of organs at risk (OARs), and plan complexity.
The findings of this study could help planners to select appropriate collimator angles to obtain optimum results.

2.A | Patient selection and planning criteria
This study was designed to compare plan complexity and dose distribution among several collimator angles (0°, 15°, 30°, 45°, 60°, 75°, and 90°). Ten patients with prostate cancer without pelvic lymph node enlargement were recruited for this study. The clinical target volume (CTV) of the prostate gland (CTV P ) consisted of the entire prostate gland and that of the seminal vesicle (CTV S ) consisted of the entire seminal vesicle. The planning target volume (PTV) of the prostate gland (PTV P ) consisted of the CTV P and a 5-mm margin (except at the CTV-rectum interface, where a 3-mm margin was used).
Identical criteria were applied to create the planning target volume of the seminal vesicle (PTV S ).
In order to reduce the complexity of radiation treatment planning, we used single-phase hypofractionated SIB-IMRT regimen published by Maurizio, with the exception that prophylactic irradiation of the pelvic lymph area was not performed in this study. 18 According to the plan, the PTV P and PTV S received 68.75 Gy (2.75 Gy per fraction) and 55 Gy (2.2 Gy per fraction), respectively, in 25 fractions. Acceptable plans were defined as the prescribed dose covering at least 95% of the PTV. The OARs dose-volume constraints were as follows: rectum, V 52 Gy <35% and V 61 Gy <25%; bladder, V 45 Gy <50%; and femoral heads V 50 Gy <10%.

2.B | VMAT plan and treatment delivery
SIB-VMAT plans were generated using the Pinnacle treatment planning system (Philips, Version 9.8.0, Fitchburg, WI, USA) for 10-MV beams from an Elekta Precise Linear Accelerator (LINAC; Elekta Ltd., Crawley, UK) and optimized using the direct machine parameter optimization algorithm. Integrated MLC consists of 40 opposed pairs of leaves, with a projected width of 1 cm at the isocenter. The total leaf travel distance is 32.5 cm and jaws cover a full 40 9 40 cm 2 field. No leaf interdigitation is allowed and the minimum gap between the opposed leaves and opposed adjacent leaves is 0.5 cm.
All calculations were performed using an adaptive convolve with a calculation grid spacing of 0.3 cm. Each plan resulted from dual-arc with a gantry and rotation of 181°-180°-181°. We optimized the dual-arc VMAT treatment plan with the collimator angle set to 0°a nd then used the same constraints to reoptimize VMAT plans using different values for collimator angles (15°, 30°, 45°, 60°, 75°, and 90°) for each patient. A maximum delivery time of 300 s/arc and a final gantry spacing of 4°were used during the optimization. The constraint leaf motion was set to 0.33 cm/deg. The maximum leaf velocity was 2 cm/s, the maximum gantry velocity was 6 deg/s, and the maximum variable dose rate was 600 MU/min. Based on the definition in the International Commission on Radiation Units and Measurements report 62, CI refers to the volume of the target receiving the prescribed dose divided by the volume of the PTV P , and has an optimal value of 1.

2.C | Dosimetric evaluation
The HI is defined as the dose received by 2% of the PTV minus the dose received by 98% of the PTVp divided by the prescribed dose (its optimal value is 0). The HI is calculated as follows: The GI is defined as the ratio of the volume covered by 50% of the prescribed dose to the treated volume of the PTV P . The GI is calculated as follows: The delivery of a high dose to a high-dose target volume unavoidably increases the dose to the surrounding low-dose target volume. To assess the quality of an SIB plan, the NDC is used to compare the dose gradient. Dose contrast (DC) is defined as the mean dose of PTV P divided by the mean dose of PTV S . 19 The ideal DC of an SIB plan is the ratio of the prescribed dose of PTV P to the prescribed dose of PTV S . Therefore, the ratio of the actual DC to the ideal DC is defined as NDC (its optimal value is 1).

2.D | Modulation complexity score for VMAT
For each of the 70 SIB-VMAT plans, the modulation complexity score of the SIB-VMAT plans (MCS V ) was calculated from the DICOM RT files. The modulation complexity score was originally proposed by McNiven for fixed-beam IMRT as a normalized sum over all the segments of the aperture area variability and leaf sequence variability. 20 Masi modified the score to suit VMAT plans by substituting the control points of the arc with segments. 21

2.E | Statistical analysis
The Wilcoxon signed-rank test was used for multiple comparison of the target parameters and critical organs at different collimator angles. P ≤ 0.05 was defined as statistically significant.

| RESULTS
The patient characteristics are presented in Table 1. The target and OARs volumes are summarized in Table 2. The dosimetric results of PTV for all studied collimator angles are summarized in Table 3 and the P values are summarized in Table 4. Comparisons of the PTV dosimetric results are summarized in Table 5. Average accumulated dose-volume histogram (DVH) of the CTV P and the PTV P is shown in Fig. 1. Figure 2 shows the dose distribution of prostate plans at the collimator angles of 0°and 45°.
H, A higher value than a collimator angle of 0°; L, A lower value than a collimator angle of 0°; S, closer to optimal value than a collimator angle of 0°; I, more away from optimal value than a collimator angle of 0°. *P≦ 0.05; **P < 0.01; ÀP > 0.05.
collimator angle of 0°exhibited the most inferior value of HI and the highest value of V 107% (3.29 cm 3 ). However, the lowest value of V 50% and the optimal value of GI were obtained with the 0°angle.
The highest value of V 50% was obtained with a collimator angle of  OARs, organs at risk; sd, standard deviation.
The P-value list of OARs dosimetric analysis (n = 10).

| DISCUSSION
All dual-arc VMAT plans for prostate cancer with seven different collimator angles fulfilled the PTV dose requirements (V 100% ≥95%).
In this study, the collimator angle of 0°had the highest value of   have reported that bladder "hotspots" are related to late bladder toxicity after external beam radiotherapy for prostate cancer. 25,26 In this study, because none of the tested collimator angles produced higher values of V 60 Gy for the bladder than did the 0°angle, MLC rotation was not considered to increase the risk of late bladder toxicity. For the left and right femoral heads, no statistically significant differences were observed for any of the tested angles.

| CONCLUSION
Collimator angle selection could play a vital role in improving the plan quality of SIB-VMAT for treating patients with prostate cancer. This study found that MLC rotation affects VMAT plan complexity and PTV dosimetric distribution. A collimator angle of 45°e xhibited the optimal values of CI, HI, and MCSv among all the tested collimator angles. Some studies have reported that late side effects of the rectum and bladder are associated with high-dose volumes. In this investigation, MLC rotation did not have statistically significantly higher values of V 60 Gy in the rectum and bladder than did the 0°angle. Based on the dose distribution of target and OARs, we thought a collimator angle of 45°was an optimal angle for the prostate VMAT treatment plan. The results of our study could serve as a guide for collimator angle selection with regard to PTV dosimetric distribution, plan complexity, and the sparing of OARs in prostate hypofractionated SIB-VMAT planning.