Investigation of Halcyon multi‐leaf collimator model in Eclipse treatment planning system: A focus on the VMAT dose calculation with the Acuros XB algorithm

Abstract

Purpose The dual‐layer multi‐leaf collimator (MLC) in Halcyon involves further complexities in the dose calculation process, because the leaf‐tip transmission varies according to the leaf trailing pattern. For the volumetric modulated arc therapy (VMAT) treatment, the prescribed dose for the target volume can be sensitive to the leaf‐tip transmission change. This report evaluates the dosimetric consequence due to the uncertainty of the dual‐layer MLC model in Eclipse through the dose verifications for clinical VMAT. Additionally, the Halcyon leaf‐tip model is empirically adjusted for the VMAT dose calculation with the Acuros XB. Materials and methods For this evaluation, an in‐house program that analyzes the leaf position in each layer was developed. Thirty‐two clinical VMAT plans were edited into three leaf sequences: dual layer (original), proximal single layer, or distal single layer. All leaf sequences were verified using Delta4 according to the dose difference (DD) and the global gamma index (GI). To improve the VMAT dose calculation accuracy, the dosimetric leaf gap (DLG) was adjusted to minimize the DD in single‐layer leaf sequences. Results The mean of DD were −1.35%, −1.20%, and −1.34% in the dual‐layer, proximal single‐layer, and distal single‐layer leaf sequences, respectively. The changes in the mean of DD between leaf sequences were within 0.2%. However, the calculated doses differed from the measured doses by approximately 1% in all leaf sequences. The tuned DLG was increased by 0.8 mm from the original DLG in Eclipse. When the tuned DLG was used in the dose calculation, the mean of DD neared 0% and GI with a criterion of 2%/2 mm yielded a pass rate of more than 98%. Conclusion No significant change was confirmed in the dose calculation accuracy between the leaf sequences. Therefore, it is suggested that the dosimetric consequence due to the leaf trailing was negligibly small in clinical VMAT plans. The DLG tuning for Halcyon can be useful for reducing the dose calculation uncertainties in Eclipse VMAT and required in the commissioning for Acuros XB.

Investigation on the impact of the leaf trailing effect using the Halcyon integrated platform system

“…We did not find any evident discrepancies in VMAT treatments, probably because in VMAT the leaves move faster and perform multiple sweepings across the beam aperture. This is in agreement with recent findings 17 and can be explained by the fact that in VMAT treatments dose discrepancies in the patient tend to smear out during gantry rotation. In dMLC plans, on the contrary, there is no gantry rotation while the beam is on and the leaves move slower and in only one direction, which increases the risk of dose discrepancies accumulating in the same region.…”

“…This produces dose deviations as great as 10% for low trailing distances and sweeping gaps of 5 mm calculated with Eclipse (see Figure 1 for further details). Miyasaka et al (2022) 17 recently evaluated sequences of clinical VMAT plans and found no dosimetric consequences associated to the trailing effect. However, to the best of our knowledge, the impact of this effect on clinical dynamic multileaf collimator (dMLC) plans has not yet been investigated.…”

Dosimetric Evaluation and Comparison of Collapsed Cone Algorithm with AAA and AcurosXB Algorithms for Lung Cancer Treatment Planning on the Varian Halcyon Accelerator

“…A study creating 1-5 arc VMAT plans for 15 prostate and 15 head and neck patients [31] found that using three or more arcs improved uniformity and conformity indices and OAR doses without affecting treatment time, with similar conclusions reported in [32]. A study by [33] suggests that adjusting the dosimetric leaf gap (DLG) of the Halcyon model can reduce the dose calculation uncertainties of AcurosXB in EclipseVMAT plans. Previous studies predominantly utilized the Halcyon model within the Varian Eclipse treatment planning system.…”

Testing of an enhanced leaf model for improved dose calculation in a commercial treatment planning system

“…As dose calculation, treatment delivery, and treatment QA are being pushed to their limits, discrepancies between measured and calculated doses have been reported. A number of publications advocate the empirical tuning of MLC parameters according to the used treatment technique in order to optimize agreement between measured and calculated patient QA [5][6][7][8][9]12 . Some create a separate beam dataset with MLC parameters tuned to be used only for stereotactic deliveries.…”