Preliminary Results on In-Vivo Dose Measurement of Intensity-Modulated Radiotherapy for Cervical Cancer
X. Tan,H. Luo,S. Li,F. Jin,Y. Wang,Y. Z. Wu
DOI: https://doi.org/10.1016/j.ijrobp.2020.07.671
2020-01-01
Abstract:To investigate the relationship between the results of in-vivo dose (IVD) measurement and setup errors based on Alderson radiotherapy (ART) phantom, and to monitor the IVD changes of intensity-modulated radiotherapy (IMRT) for cervical cancer patients. Electronic portal imaging device (EPID) images and logfile data were collected from ART phantom and 12 patients with cervical cancer, which were used for two-dimensional (2D) IVD verification [evaluation indexes: Gamma analysis (γ) and percent dose difference (DD)] and three-dimensional (3D) IVD verification (evaluation indexes: γ), respectively. The positioning errors with ranges of 0.1∼1 cm and 1∼3cm in the step size of 0.2 cm and 1 cm were artificially introduced in ART phantom. The correlations between IVD results of phantom and its setup errors were analyzed, and the correlations between IVD results of the patients and their treatment fractions were evaluated. The 26 groups of EPID images and corresponding logfiles data from ART phantom and 206 groups from the patients were obtained respectively. The setup errors of ART phantom were not correlated with the corresponding 3D IVD γ1%/1mm (P > 0.05). The 2D IVD averages of ART phantom (γ3%/3mm,mean, γ2%/2mm, mean, γ1%/1mm,mean, DD3%,mean) showed strongly negative correlations with the setup errors (R∈[–0.997, –0.765], P < 0.05). The treatment fractions of all patients were not correlated with their 3D IVD γ1%/1mm (P > 0.05). For 2D IVD verification, DD3%,mean of all patients was negatively correlated with the treatment fractions (R∈[–0.953, –0.499], P < 0.05), among which, γ3%/3mm, mean of 9 patients was more than 89% in all treatment fractions, and γ3%/3mm, mean of the 9 patients was more than 93% for 98.57% treatment fractions. For other 3 patients, γ3%/3mm, mean showed strongly negative correlations with the treatment fractions (R∈[–0.903, –0.824], P < 0.05), with the range of 38% and 100%. Their CBCT images showed that the bladder volume decreased significantly (the relative changes: 82.08%, 84.41% and 73.59%, respectively) and the target retracted significantly (the relative changes: 38.12%, 59.79% and 24.46%, respectively). Conducting IVD verification could monitor the mechanical stability of accelerator and the changes of IVD in patients, and also evaluate the positional deviation of tumors and surrounding organs with IGRT technology, which could improve the safety and quality assurance of radiotherapy for cervical cancer patients and provide guidance for patients with adaptive radiotherapy.