Explore the feasibility of using spot‐scanning proton arc therapy for a synchrotron accelerator‐based proton therapy system – A simulation study

Xiaoda Cong, Gang Liu, Peilin Liu, Lewei Zhao, Shupeng Chen, Xiaoqiang Li, Jiajian Shen, Xuanfeng DingJournal of Applied Clinical Medical Physics 2024 · DOI

TL;DR

This study simulates Spot-Scanning Proton Arc (SPArc) therapy on a synchrotron-based proton therapy system, finding it offers superior dosimetric plan quality compared to conventional Intensity-Modulated Proton Therapy (IMPT). However, the synchrotron's limitations lead to significantly longer treatment delivery times, posing a challenge for clinical implementation.

What Problem Does This Paper Address?

Proton therapy, particularly Intensity-Modulated Proton Therapy (IMPT), is a powerful tool for cancer treatment due to its precise dose delivery and reduced damage to healthy tissues. Spot-Scanning Proton Arc (SPArc) therapy, introduced in 2016, has shown promise for even better dosimetric quality and more efficient delivery. However, most SPArc research has focused on cyclotron-based systems. Synchrotron-based proton therapy systems, while capable of delivering protons, have historically faced challenges with lower dose rates, longer cycling times, and limited maximum charges per spill. This raises a critical question: can SPArc therapy, with its complex beam delivery, be effectively implemented on a synchrotron system, and what are the trade-offs in terms of plan quality and treatment time? This paper aims to answer this by simulating SPArc plans on a synchrotron platform and comparing them to IMPT.

Methods

The researchers used a HITACHI ProBEAT PBS system's physics beam model within the RayStation treatment planning system (TPS) to generate both IMPT and SPArc plans. They selected five representative disease sites (head and neck, lung, liver, brain chordoma, and prostate) for retrospective analysis.

  • IMPT Plan Generation: Standard IMPT plans were generated using the existing clinical plan parameters, serving as a baseline for comparison.
  • SPArc Plan Generation: SPArc plans were created using an in-house developed optimization algorithm within RayStation. This involved a multi-step process starting with a coarsely sampled IMPT plan and iteratively refining it with a 2.5° sampling frequency for the arc. The algorithm also focused on redistributing energy layers and filtering low-weight control points to maintain efficiency.
  • Delivery Time Simulation: Both static and dynamic beam delivery times (BDT) were simulated. The static BDT was based on a previously published synchrotron delivery sequence model, while the dynamic BDT incorporated a proton arc gantry mechanical model integrated with the synchrotron delivery sequence. These simulations excluded gantry rotation time and other logistical factors, focusing purely on the beam delivery process.
  • Dosimetric Evaluation: Plan quality was assessed using dose-volume histograms (DVHs), conformity index (CI), and integral dose (ID). Specific metrics included D99% for target coverage and maximum/mean doses for organs at risk (OARs).
  • Robustness Evaluation: The study also evaluated the robustness of the plans under worst-case scenarios, assessing mean dose, D95, and V95.

Datasets

The study utilized retrospective patient data from five representative disease sites (head and neck, lung, liver, brain chordoma, and prostate) treated at Corewell Health, William Beaumont University Hospital. No new benchmark datasets were introduced; rather, existing patient data was used for simulation.

Key Insights and Ideas

  • SPArc's Dosimetric Advantage Confirmed: Even on a synchrotron system, SPArc consistently demonstrated superior dosimetric plan quality compared to IMPT across all tested disease sites. This included improved target dose conformity and significant reductions in dose to critical OARs.
  • Synchrotron's Delivery Time Bottleneck: The primary challenge identified is the significantly prolonged treatment delivery time for SPArc on a synchrotron system. This is attributed to the inherent limitations of the synchrotron's single energy extraction (SEE) mechanism, leading to numerous cycling periods.
  • Trade-off Between Quality and Speed: The study highlights a clear trade-off: while SPArc offers better dose distribution, its clinical feasibility on a synchrotron is hampered by the extended treatment duration.
  • Potential for Future Improvements: The findings point towards the need for advancements in synchrotron technology (e.g., multi-energy extraction) and TPS optimization algorithms to mitigate the delivery time issue.

Key Results

  • Dosimetric Improvement: SPArc plans showed significantly better plan quality, with improved target dose conformity and reduced OAR doses across all five disease sites. For example, in the head and neck case, the maximum dose to the spinal cord was reduced from 38.59 Gy (IMPT) to 11.76 Gy (SPArc). The conformity index (CI) also improved for SPArc in most cases.
  • Integral Dose Reduction: SPArc plans exhibited a 1.90% to 24.56% lower integral dose compared to IMPT plans.
  • Increased Delivery Time: SPArc plans resulted in a prolonged delivery time. On average, static delivery time increased by 49.22% and dynamic delivery time by 59.10% compared to IMPT. For the skull chordoma case, the delivery time more than doubled.
  • Robustness: The robustness evaluation showed equivalent performance between SPArc and IMPT plans under worst-case scenarios.

Limitations and Future Work

The authors acknowledge several limitations:

  • SEE Mechanism: The delivery time simulations were based on the single energy extraction (SEE) mechanism, which is a bottleneck. The study suggests that multi-energy extraction (MEE) techniques, which are being implemented, could significantly improve delivery efficiency.
  • Full-Circle Gantry Assumption: The simulations assumed a full-circle gantry. Centers using half-circle gantries might require additional couch/chair rotation, impacting delivery time.
  • Idealized IMPT Simulation: The IMPT delivery time simulation assumed an ideal auto step-and-shoot scenario, which might not reflect real-world clinical practice.

Future work should focus on:

  • MEE Integration: Investigating SPArc delivery with MEE synchrotron systems.
  • Advanced Optimization: Developing new cycling techniques and optimization algorithms to reduce energy layer numbers and improve efficiency.
  • Fast Line Scanning: Exploring fast line scanning techniques for large target volumes.
  • Manual Arc Trajectory Planning: Investigating manual arc trajectory planning to avoid critical OARs, as demonstrated in their supplemental material.

Summary updated

Abstract

This study presents the first simulation results of delivering the SPArc plans using a synchrotron-accelerated proton therapy system. The result shows its feasibility and limitations, which could guide future development.
BibTeX
@article{cong2024explore,
  author = {Xiaoda Cong and Gang Liu and Peilin Liu and Lewei Zhao and Shupeng Chen and Xiaoqiang Li and Jiajian Shen and Xuanfeng Ding},
  title = {{Explore the feasibility of using spot‐scanning proton arc therapy for a synchrotron accelerator‐based proton therapy system – A simulation study}},
  journal = {Journal of Applied Clinical Medical Physics},
  year = {2024},
  doi = {10.1002/acm2.14526},
}

Loading related content...

Figure