Session Item

Our center has been using a library-based plan of the day (PotD) strategy for EBRT for locally advanced cervical cancer patients since 2011. The goal of this strategy is to reduce dose to normal tissue. Since 2011, various adjustments have been made to the PotD and the treatment protocol. However, the performance of the PotD has not been assessed longitudinally. This study analyzes PotD performance over nine years and compares it to a maximum attainable normal tissue sparing based on daily replanning.

All patients clinically treated with the PotD protocol were included. As part of the protocol, patients were categorized as movers or non-movers based on tip-of-uterus displacement between an empty and full bladder CT scan. Non-movers and movers received one and two personalized library plans respectively, based on a motion model constructed from the full and empty bladder CT scan. An additional motion-robust plan (PTV_BU) was available for all patients. Starting date of radiotherapy, selected plans, and frequency and reason for full replanning were registered. Full replanning (new planning CT scans and plan library) was performed when PTV_BU was selected three or more times during treatment.

Planning structures were collected. Weighted PTV volumes (PTV_w) were calculated for all patients using the weighted sum of the PTV volumes from the library plans actually used for the fractions. This was considered representative for clinical practice. A fully-online adaptive strategy was simulated for each patient (PTV_{fully adapt.}), which comprised the full-bladder CTV and elective CTV with a 5-mm margin.

The performance of the protocol was assessed by the percentage of patients requiring full replanning, the percentage of used motion-robust plans, and the comparison between the PTV_w, PTV_BU and PTV _fully-adapt. volumes. 

A total of 376 patients was treated between January 2011 and May 2020, of which 18% required full replanning. 52% of these cases was attributable to shortcomings of the motion model. The total use of back-up plans decreased from 20% from 2011 to 5-10% beyond 2014. Figure 1 shows that this and protocol adjustments have improved the PotD strategy over time, resulting in a mean decrease of 500 cc in PTV_w volume. A stabilization is visible after 2015. Figure 2 displays the boxplots of the PTV_BU, PTV_w, and PTV_{fully adapt} of patients treated in the second half of 2016 and later. This figure illustrates that the use of the plan library decreased PTV_w with 17% compared to PTV_BU and that a fully online-adaptive strategy could increase this reduction to 37%. 

This long-term analysis shows that plan-library-based PotD can consistently reduce PTV volume by 17%. Further reduction is possible, but at the expense of a higher workload.