Influence of the complexity metrics of the multi-leaf collimator on the γ-pass rate of volumetric modulated arc therapy plans for nasopharyngeal carcinoma
10.3760/cma.j.cn112271-20240430-00166
- VernacularTitle:多叶准直器复杂度指标对鼻咽癌容积旋转调强放疗计划γ通过率的影响
- Author:
Junwen TAN
1
;
Yusong LONG
1
;
Xiantao HE
1
;
Gang LI
1
;
Yongfu FENG
1
;
Weixue LIANG
1
;
Zhanyu WANG
1
Author Information
1. 柳州市工人医院肿瘤科，柳州 545005
- Publication Type:Journal Article
- Keywords:
VMAT;
γ-passing rate;
Complexity metric;
Nasopharyngeal carcinoma
- From:
Chinese Journal of Radiological Medicine and Protection
2025;45(4):309-316
- CountryChina
- Language:Chinese
-
Abstract:
Objective:To investigate the influence of the complexity metrics of the multi-leaf collimator (MLC) on the γ-pass rate of volumetric modulated arc therapy (VMAT) plans for nasopharyngeal carcinoma (NPC).Methods:A total of 60 VMAT plans for NPC were selected to measure the γ-pass rate. The MLC data across all control points (CPs) in each VMAT plan were analyzed to calculate the mean field area (MFA), average leaf gap (ALG), small aperture score (SAS), and their corresponding weighted metrics including MFAW, ALGW, and SASW, considering dose weight (W). Pearson′s bivariate correlation analysis was conducted to examine the correlations between MLC complexity metrics and the γ-pass rate. Moreover, the receiver operating characteristic (ROC) analysis was employed to assess the predictive efficacy of MLC complexity metrics on dose verification result.Results:Under the 3%/3 mm, 3%/2 mm, and 2%/2 mm criteria, the MFA in the 60 VMAT plans exhibited a positive correlation with the γ-pass rate ( r=0.82, 0.79, 0.72, P<0.05), and the MFAW was also positively correlated with the γ-pass rate ( r=0.83, 0.81, 0.75, P<0.05). The ALG manifested a positive correlation with the γ-pass rate ( r=0.82, 0.79, 0.74, P<0.05), as did the ALGW ( r=0.83, 0.81, 0.77, P<0.05). The SAS(0-1 cm), SAS(1-2 cm), SAS(2-3 cm), and SAS(3-4 cm) displayed negative correlations with the γ-pass rate ( r= -0.86, -0.82, -0.71, -0.84, -0.82, -0.72, -0.79, -0.79, -0.73, -0.30, -0.35, -0.42, P<0.05), whereas the SAS(4-5 cm), SAS(5-6 cm), and SAS(6-40 cm) showed positive correlations with the γ-pass rate ( r=0.49, 0.45, 0.33, 0.73, 0.71, 0.59, 0.79, 0.79, 0.76, P<0.05). The outcomes of SASW reveal slightly stronger correlations than those of SAS. In terms of correlations among complexity metrics, a positive correlation was observed between MFA and ALG ( r=0.98, P<0.05). ALG was negatively correlated with SAS(0-1 cm), SAS(1-2 cm), SAS(2-3 cm), and SAS(3-4 cm) ( r= -0.95, -0.94, -0.89, -0.39, P<0.05), and positively correlated with SAS(4-5 cm), SAS(5-6 cm), and SAS(6-40 cm) ( r=0.51, 0.77, 0.92, P<0.05). The weighted result mirrored these correlations. The ROC-derived analytical result indicate that MFA, MFAW, ALG, ALGW, SAS(0-1 cm), SAS(1-2 cm), SAS(2-3 cm), SAS(6-40 cm), SASW(0-1 cm), SASW(1-2 cm), SASW(2-3 cm), and SASW(6-40 cm) demonstrated exceptional predictive efficacy for dose verification result [Area under the curve (AUC) > 0.9, P<0.05]. Conclusions:The γ-pass rate of VMAT plans for NPC is strongly correlated with MLC complexity metrics, which demonstrate excellent predictive efficacy for dose verification result. The increased uncertainty in dose calculations and measurements caused by narrow fields generated by the MLC is a significant factor contributing to the reduced γ-pass rate of VMAT plans. This finding is associated with discrepancies in the precision of treatment planning system (TPS) modeling and the accuracy of dose verification tools, providing a reference for similar challenges.
