Partitioning of multiple brain metastases improves dose gradients in single-isocenter radiosurgery

TL;DR

This work tackles the island-blocking challenge in single-isocenter VMAT for multiple brain metastases by introducing a multi-target partitioning algorithm that divides targets into subsets treated by separate arc passes. The method is formulated with a dynamic-programming cost function that minimizes island blocking and can handle multiple arc trajectories, and it is integrated into an automated RayStation planning workflow. In simulations (20 cases with 10 metastases) the partitioning approach substantially improves dose gradient metrics (GI and $G\eta$) and reduces brain V12Gy compared with simultaneous treatment, albeit with higher MUs and longer delivery times. Retrospective clinical testing on six cases shows comparable or slightly improved plan quality relative to hand-crafted clinical plans, supporting generalizability; overall, partitioning enables high-quality multi-target SRS on widely available linacs by decoupling island blocking from beam arrangement decisions.

Abstract

Background: A growing number of cancer patients with brain metastases can benefit from stereotactic radiosurgery (SRS) thanks to recent advances in systemic therapies. With an increasing patient load, single-isocenter treatments on widely available C-arm linear accelerators are an attractive option. However, the planning of such treatments is challenging for multi-target cases due to the island blocking problem, which occurs when the multi-leaf collimator cannot conform to all targets simultaneously. Purpose: We propose a multi-target partitioning algorithm that mitigates excessive exposure of normal tissue caused by the island blocking problem. Methods: The algorithm divides (partitions) the set of targets into subsets to treat with separate arc passes, optimizing both subsets and collimator angles to minimize island blocking. The algorithm was incorporated into a fully automated treatment planning script and evaluated on 20 simulated patient cases, each with 10 brain metastases and 21 Gy prescriptions. It was also retrospectively evaluated on six clinical cases. Results: Partitioning significantly improved the gradient index, global efficiency index, and brain V12Gy compared to simultaneous treatment of all metastases. For example, the average gradient index improved from 5.9 to 3.3, global efficiency index from 0.32 to 0.46, and normal brain V12Gy from 49 cm3 to 26 cm3 between 3 and 9 arcs. The proposed algorithm outperformed baselines in utilizing a limited number of arcs. All target partitioning strategies increased the total number of monitor units (MUs). Conclusions: The dose gradient in single-isocenter VMAT plans can be substantially improved by treating a smaller subset of metastases at a time. This requires more MUs and arcs, implying a trade-off between delivery time and plan quality which can be explored using the algorithm proposed in this paper.

Figures (7)

• Figure 1: Illustration of the target partitioning approach. The island blocking problem occurs when targets line up in the direction of leaf travel, leading to exposure of normal tissue between the targets, as shown in the leftmost aperture. By partitioning the set of targets into subsets, we may avoid the island blocking problem completely, as in the three apertures to the right.
• Figure 2: Summary of the 200 prestudy experiments. The values on the y-axis represent the minimum number of arcs (summed over the three couch angles) required to fully eliminate island blocking using a target partitioning approach.
• Figure 3: Plan quality comparison between the proposed algorithm and the baselines. White triangles indicate mean values. Note that objective value refers to the dose-based objectives used in the VMAT optimization, and that PCI and the PTV DVH metrics are computed per individual PTV (10 per plan) whereas the other metrics are computed per plan.
• Figure 4: Delivery time trade-off. This example illustrates how the dose gradient varied with the number of arcs for the proposed algorithm in one of the simulated cases. PTV contours are shown in black. Note how the dose gradient clearly improves between 3, 6, and 9 arcs, whereas the difference between 9 and 30 arcs is small. Since the delivery time is several times longer for 30 arcs, the 9-arc solution is much preferred in clinical practice.
• Figure 5: Plan comparison for the six clinical cases included in the retrospective planning study. Since the prescription dose levels and number of fractions varied across cases, the PTV DVH metrics are presented relative the prescription dose level, and the OAR near-max doses ($\text{D}_{0.1~\unit{\centi\meter\cubed}}$) are presented relative their fractionation scheme-specific acceptance levels.