Author: Grant Gibbard, Chunhui Han, An Liu 👨🔬
Affiliation: RefleXion Medical, Department of Radiation Oncology, City of Hope National Medical Center 🌍
Purpose: To evaluate the efficacy of a new approach to streamline treatment planning for biology-guided radiotherapy (BgRT) of fluorodeoxyglucose (FDG)-avid bony lesions on a BgRT machine with real-time positron emission tomography (PET) guidance.
Methods: Treatment Planning data were retrieved for ten patients who received BgRT to bony lesions. Based on pre-treatment PET scans on the BgRT machine, the average activity concentration was 15.0±10.7 kBq/ml for the planning target volume (PTV). Using a new BgRT plan optimization approach, shell structures were created at fixed distances from the PTV to control dose conformity and dose drop-off. A standard optimization parameter template was used as the starting point in plan optimization. Optimization parameters were subsequently adjusted systematically based on objective cost output after each round of plan optimization. Target dose conformity was evaluated with the RTOG conformity index (CI), while dose drop-off was evaluated with CI80 and CI50 (ratio of the 80% and 50% prescription isodose volume to the PTV volume, respectively).
Results: With an average PTV volume of 41.3±31.6 cm3 and prescribed dose ranging from 10 to 30 Gy in one to three fractions, the clinical BgRT treatment plans had average CI, CI80, and CI50 of 1.38±0.26 (range: 1.12 – 1.84), 2.88±0.60 (range: 2.05 – 3.95), and 7.36±1.67 (range: 5.29 – 9.60); the average homogeneity index (HI) was 1.34±0.14. With the new plan optimization approach applied, the average CI, CI80, and CI50 was 1.25±0.11 (range: 1.10 – 1.46), 2.46±0.33 (range: 2.08 – 3.06), and 6.48±1.31 (range: 4.61 – 8.56); the average HI was 1.38±0.12. Statistically significant differences were found for CI, CI80, and CI50 between clinical and new plans.
Conclusion: Improvement in target dose conformity and dose drop-off were found with statistical significance with the new optimization approach, which could allow a more efficient BgRT planning workflow and more consistent plan quality.