Predictive SIRT dosimetry based on a territorial model.

Area of Science:

• Medical Physics
• Oncology
• Radiology

Background:

• Accurate dosimetry is crucial for planning selective internal radiation therapy (SIRT) in liver cancer treatment.
• Current methods may not fully capture dose variability within the liver.
• Personalized dosimetry is needed to optimize treatment efficacy and minimize toxicity.

Purpose of the Study:

• To propose and evaluate an optimized partition model for SIRT dosimetry based on arterial liver territories.
• To characterize dose variability within the whole liver using a new model.
• To compare the proposed model with classical methods for activity and dose calculation.

Main Methods:

• Developed a generalized absorbed dose calculation for arbitrary partitions based on arterial supply areas.
• Defined normal liver partitions per arterial supply and tumor partitions.
• Excluded an 11 mm margin around tumors from normal liver calculations.
• Performed activity and dose calculations for five patients using body-surface-area (BSA), classical, and territorial partition models.

Main Results:

• The territorial model resulted in lower normal liver doses and higher tumor doses compared to the classical model.
• Excluding a region around tumors significantly impacted mean liver dose.
• Tumor activities calculated with the proposed method were higher when limited by normal liver dose.
• BSA method yielded the lowest activity calculations.

Conclusions:

• The territorial model offers a more localized and patient-specific dose distribution in normal liver by considering arterial supply areas.
• This arterial liver territory-based partition model can support SPECT-independent activity calculation and dose prediction.
• The model requires artery-based simulation for particle distribution for accurate implementation.