Author: Danielle Renee Daiger, Nicholas N. Sperling 👨🔬
Affiliation: University of Toledo, University of Toledo Medical Center 🌍
Purpose:
To retrospectively evaluate the impact of planned delivery time on plan quality metrics for breath-hold VMAT SBRT thoracic cancer patients.
Methods:
Patients treated for SBRT thoracic cancer using breath-hold techniques and treated using VMAT at the UTMC Dana Cancer Center between January 1, 2018, and January 1, 2024 were evaluated on the Edge treatment machine (Varian). The delivered treatment plan for each patient was used as the reference, and 10 new plans for each patient were created by adjusting the maximum delivery time in RayStation 11B (RaySearch) for each beam. Each plan was scaled to maintain equivalent target D95% coverage. For each plan the reference prescribed dose was used, with all selected patients having received 50Gy in either 4 or 5 fractions. Each plan was optimized for a total of 240 iterations broken into 6 rounds of 40. To evaluate plan quality conformity indices, critical dose to organs-at-risk (heart, spine, and lungs), and modulation factor were calculated as a function of reported planned delivery time.
Results:
Evaluation of the metrics indicates that an increase in delivery time yields improvements in the Gradient Index, Paddick Conformity Index, and RTOG Conformity Index, as well as lower critical dose to organs including the spine and lungs; while resulting in increased modulation factor. The improvement in plan quality metrics does not follow a linear trend but instead yields diminishing returns with increases in delivery time, with minimal improvement observed for delivery times greater than 50 seconds.
Conclusion:
This work shows that, while extending delivery time can improve plan quality, the improvements may not justify the prolonged duration beyond a certain threshold, emphasizing the need for optimal treatment time selection. Delivery time should therefore be considered as an additional factor in the development of clinical treatment planning protocols.