Author: Michael Buckstein, Yang Lei, Tian Liu, Charlotte Elizabeth Read, Jing Wang, Kaida Yang, Jiahan Zhang 👨🔬
Affiliation: Icahn School of Medicine at Mount Sinai 🌍
Purpose: Inter-fraction anatomic variations in liver SBRT can cause significant discrepancies between planned and delivered doses. We developed a CBCT-based accumulative algorithm to directly compare planned and delivered doses in the planning target volume (PTV) and organs at risk (OARs).
Methods: A cycle diffusion network was used to deformably register planning CT images to daily pre-treatment CBCTs. Electron densities from the planning CT were propagated to the daily anatomy, generating deformed CTs for dose calculation. Daily doses were mapped back to the planning CT through a reverse registration process for direct comparison with the planned dose. The accumulated delivered dose was calculated as the average of deformed daily doses across all fractions. In this proof-of-concept study, 15 liver SBRT patients (10 Gy × 5 fractions) treated using abdominal compression for motion management were analyzed. Patients were simulated with an Alpha Cradle and SBRT board with abdominal compression, using intravenous contrast for enhanced imaging. An internal target volume (ITV) was generated with 4D CT and OARs were contoured on the initial scan. Three-fold cross validation was employed. Dose verification metrics included PTV coverage and OAR dose-volume parameters.
Results: The mean accumulated delivered dose showed a lower PTV V100% coverage of 88.25% (±11.81) compared to the mean planned dose of 90.96% (±10.93), with a maximum PTV coverage drop of 10.69%. For OARs, mean doses delivered showed minimal variations compared to plans, with a maximum increase of 2.44 Gy in lung D0.03cc and 0.51 Gy in large bowel D0.03 cc. Overall, dose variations were well controlled with abdominal compression.
Conclusion: The proposed CBCT-based approach offers an accumulative solution for a direct dose comparison between the plan and daily treatment in liver SBRT. This precise dose verification method paves the way to dose escalation in future ablative radiotherapy.