Author: Laura I. Cervino, Sharif Elguindi, Yu-Chi Hu, Licheng Kuo, Xiaoning Liu, Jennifer Ma, Pengpeng Zhang 👨🔬
Affiliation: Department of Medical Physics, Memorial Sloan Kettering Cancer Center, Radiation Oncology, Memorial Sloan Kettering Cancer Center 🌍
Purpose:
Quad-shot radiotherapy (QSRT) is an emerging paradigm in palliative cancer treatment. This study aimed to streamline the QSRT workflow using automation to improve efficiency while maintaining high-quality treatment for palliative lung cancer.
Methods:
In the conventional QSRT workflow, re-simulation CTs were acquired before each new cycle, with replanning performed if the physician observed significant anatomical changes. This manual process often introduced a 5–7-day between resimulation and treatment initiation. To address these inefficiencies, an automated workflow modeled on offline adaptive radiotherapy was developed. The workflow integrates AI-based auto-contouring of OARs, an in-house auto-optimization program for treatment planning, and a script for dose accumulation over cycles to evaluate treatment. Fourteen patients who underwent three cycles of thoracic QSRT were included in this retrospective study. For new cycles, plans were generated using the workflow and compared to clinical delivered plan. Plan dose over three cycles was accumulated for each patient and analyzed to assess dosimetric benefits. Additionally, the time required for each automated step was recorded to evaluate efficiency.
Results:
The automated steps were completed in an average of under one hour, enabling same-visit treatment scheduling after simulation. Across 28 cycles, the workflow-generated plans improved PTV D95% by 2% (p=0.001) and reduced esophagus Dmax and Dmean by 5% (p=0.041) and 11% (p=0.022), respectively, while heart Dmax decreased by 2% (p=0.027). Greater improvements were observed in cycles treated with the original plan, where auto-replans increased PTV D95% by 4% (p=0.005), reduced esophagus Dmax by 10% (p<0.001), and decreased heart Dmax by 4% (p=0.001). Accumulated doses showed a 3% improvement in PTV D95% and reductions in OAR doses ranging from 0% to 8%.
Conclusion:
The automated QSRT workflow demonstrated significant dosimetric improvements and efficiency gains over the traditional approach, enabling same-day treatment scheduling and practical clinical implementation for palliative lung cancer radiotherapy.