3146 - Biology-Guided Radiotherapy with Second Generation Tracked Dose Delivery Algorithms: Initial Plan Quality Evaluation

Purpose/Objective(s): Biology-guided radiotherapy (BgRT) is a new radiation modality that utilizes real-time PET data to deliver a dynamically tracked dose distribution to a tumor. This study evaluates BgRT treatment plan quality achievable with new tracked dose delivery algorithms under development* against currently available BgRT treatment plan quality.

Materials/Methods: This retrospective comparative evaluation of BgRT plan quality included 20 de-identified clinical cases covering a broad mix of tumor sizes (median GTV of 8.35 cc, range 0.5 – 45.6 cc) and anatomical locations (neck, thorax, abdomen) suitable for stereotactic radiotherapy. A simulation tool was used to convert diagnostic FDG PET images from these cases into synthetic BgRT planning PET data for the existing platform and a new wide PET FOV future platform accounting for the system sensitivity, acquisition time, reconstruction method and geometry of each platform. These synthetic PET planning data were subsequently used to generate and compare simulated BgRT treatment plans for the current platform with existing algorithms and the future platform using the new tracked dose delivery algorithms. Plan quality was evaluated using target coverage, homogeneity index (HI), conformity index (CI), CI at 80% prescription dose (CI80), CI at 50% prescription dose (CI50), and maximum dose at 2 cm from the PTV (D2cm). The mean and standard deviation for each dosimetry metric were calculated for both sets of plans and paired t-tests were used to compare each set of metrics with a significance level of p < 0.05. To support fair comparisons, PTV coverage was kept as close to 95% as possible and well matched between pairs of plans being compared. Additionally, to avoid tradeoffs between dose gradients within the PTV and other plan quality metrics, a maximum PTV dose constraint of 125% prescription dose was applied to all cases to control the HI.

Results: The measured mean ± standard deviation for each dosimetry metric is reported in the table below for both sets of plans along with the p values from the t-test comparisons. By design, no significant difference was observed in target coverage between the two platforms. For all other metrics evaluated, a statistically significant improvement was observed for the new platform incorporating the second generation tracked dose delivery algorithms.

Conclusion: This investigation provides initial evidence that the next generation tracked dose delivery algorithms being developed on the new platform may significantly improve BgRT plan quality. These improvements in plan quality have the potential to expand the number of patients that could be treated with BgRT on the new platform.