Patient-Specific QA Using Multidimensional Detectors in Single Isocenter Multitarget Stereotactic Radiotherapy for Multiple Brain Metastasis

Purpose: This study used ArcCHECK (AC) to investigate the connection between target- and treatment plan-dependent parameters and the gamma passing rate (GPR) to determine the characteristics of patient-specific quality assurance (PSQA) in single isocenter multitarget (SIMT) stereotactic radiotherapy (SRT). Methods: Twenty-four SIMT SRT treatment plans (92 targets) were evaluated. All treatment plans were developed using dynamic conformal arc therapy (DCAT), and the equivalent multi-leaf collimator (MLC) field size, distance from the isocenter (IC), and monitor/Gy were determined. The absolute dose error from the treatment planning system (TPS) was calculated using 92 targets and an ionization chamber detector. Gamma analysis was performed with AC, a multidimensional detector, and SNC patient software. Threshold values of 10% and 20%, absolute and relative dosimetry, global mode, and dose difference (DD)/distance-to-agreement (DTA) criteria of 3%/2 mm, 2%/2 mm, 2%/1 mm, and 1%/1 mm were employed. Differences in GPR were assessed for each condition. The correlation of GPR with treatment planning parameters and ionization chamber detector errors was investigated. Results: With stricter DD (2%/2 - 2%/1 mm), the GPR fell by an average of 18.6% (86.4% - 67.8%). The average equivalent MLC field size was moderately correlated with GPR and weakly correlated with the maximum error of the ionization chamber detector and MU/Gy, but not with the max target-IC distance. In the case of small targets, the dose in the center was relatively acceptable, but the lower dose range outside of the target was less responsive, frequently resulting in failures. Conclusions: In SIMT SRT, we investigated in detail the differences in GPR based on the AC’s gamma analysis conditions, as well as the relationship between the GPR and treatment planning parameters. The GPR was found to be significantly reduced when the average equivalent MLC field size was small, i.e., there were many small target sizes.