GSOR7 Presentation Time: 9:30 AM: Shortening the Learning Curve: Pre-Planning with a Digital Library of 3D Printed Custom Templates for Improved Dosimetry on Day One in Combined Tandem and Ovoid + Interstitial Cases

Abstract

<h3>Background</h3> In cervical brachytherapy the addition of interstitial (IS) needles to intra-cavitary (IC) applicators can significantly enhance dosimetry by improving target coverage without increasing dose to normal tissues when needles are optimally placed. It usually takes more than one implant to learn and adapt to the patient specific challenges to provide the best possible needle distribution, resulting in improved dosimetry metrics over the course of multi-fraction treatments. <h3>Material and Methods</h3> We developed a digital library of customized supplemental templates that attach to the tandem applicator and guide needles to optimized positions for different tumor topologies. The library consists of CT scans of tandem and ovoid (T&O) applicators with needles guided through these 3D printed templates that can be used for treatment planning (Figure 1A). Following a pre-procedure MRI, CTV contours were drawn and a grid overlaid. Then, an appropriate template can be selected based on tumor dimensions and extent from the uterine canal and a treatment plan generated to best prepare for the first implant (Figure 1B). An appropriate strategy could thus be defined for the first implant, including the specific custom applicator, number of needles and depth of needles. We started using this preplanning library approach in March 2021. We compared the dosimetry of hybrid implants with library preplanning to those that we were not pre-planned from the preceding year and compared the CTV V100%. <h3>Results</h3> Thirteen patients were pre-planned (PP) using the digital library and compared to the previous year's hybrid cases without preplan (n=16). The mean CTV V100% for the no pre-plan (NPP) group was 86.2%, 90.8%, 91.5% and 93.0% for fractions 1-2-3-4 respectively whereas the pre-planning group achieved a V100% of 90.8%, 94.7%, 94.3% and 92.9%. The difference in the first fraction's V100% is statistically significant (p=0.015). Similar dosimetry was achieved in the second fraction of the NPP cohort as compared to the first fraction of the PP cohort, showing the benefit of detailed preplanning (Figure 1C). By taking into account all the available pre-procedure anatomic information and potential implant geometry options in preplanning, many of the challenges of the first applicator insertion can be overcome and thereby shortening the learning curve to optimized dosimetry right from the first implant. <h3>Conclusions</h3> By developing a digital library of customized hybrid applicators and then utilizing it for planning on pre-procedure MRI, we were able to execute dosimetrically on the first fraction what was historically achieved on the second fraction. The pre-planning steps outlined above significantly improved the CTV coverage of the first fraction, showing that many of the patient specific learning steps can be surmounted before the first implant.