Standardization of Stereotactic MR-Guided Adaptive Radiation Therapy Planning for Prostate Cancer

Abstract

To assess the impact of a standardized planning class solution (PCS) in stereotactic MR-guided adaptive radiation therapy (SMART) for prostate cancer on efficiency and robustness for daily adaptive radiation therapy treatments. Data collection of 20 previous SMART prostate plans: number of objective structures, number of beam segments, treatment delivery time (TDT) and the plan adaption robustness (PAR). The PAR was scored from 1 to 3 for the robustness of the plan to re-adapt and achieve similar dose metric scores to the baseline plan on all fractions. These scores were, 1: a single re-optimization required, 2: cost-function adjustments required and 3: medical physics advice required. Five medical dosimetrists provided preferences for objective structures and the optimization thresholds used. A PCS was created based on this data and implemented into SMART prostate plans between October 2022 and December 2022. A consecutive sample of 20 patients (PCSpost) were then retrospectively compared with the previous 20 patients prior to the PCS (PCSpre). The PCS showed minimal difference in the TDT and segment number mean ± (standard deviation), 9.5 ± (0.7) mins and 77.8 ± (6.9) segments PCSpost respectively compared to 9.5 ± (0.8) mins and 80.8 ± (5.9) segments PCSpre. However, the PCS resulted in a considerable reduction in PAR, with 11 patients requiring only a single re-optimization for each fraction to achieve dose metric targets compared to 7 PCSpre patients. Also, the median PAR score was 1 in PCSpost compared to a score of 2 with PCSpre. The PCS had good compliance with the criteria met in 18/20 patient baseline plans. Feedback surveys showed positive results for efficiency saving from multiple disciplines; in therapeutic radiographers from reduced PAR, medical physicists increased automation in plan checking and consistent planning guidelines for medical dosimetrists. The PCS is in clinical use, results have shown improved treatment adaption efficiency and robustness. Fewer re-optimizations per treatment were required onset to achieve similar dose metrics than before its implementation.