KEY POINTS
- The computational framework modelled unlabelled actinium-225 chloride delivered around a spherical 1.5-cm postoperative cavity, with five equally spaced catheter sites positioned 10 mm beyond the cavity wall.
- Transport incorporated diffusion, pressure-driven convection, and physical decay over 15 hours. TOPAS Monte Carlo simulations included the complete actinium-225 decay chain, and dose was integrated over 50 days using a baseline alpha-particle RBE of 5.
- The modelled therapeutic radius—defined by at least 99% clonogenic-cell killing—was approximately 1.43 cm from each injection site. At 50 days, dose fell from approximately 52.2 Gy(RBE)/MBq at 1.5 cm to 0.37 Gy(RBE)/MBq at 3 cm.
- A five-catheter plan with 10 MBq total activity reduced global surviving fraction from 100% without alpha therapy to approximately 3.4%. Survival fell further to approximately 2.0% with 20 MBq and 1.4% with 30 MBq.
- Residual clonogenic cells were concentrated more than 2 cm from injection sites, demonstrating that activity escalation could not fully compensate for incomplete geometric coverage.
- Catheter position had a larger effect than the assumed RBE. Moving the injections from 10 mm beyond the cavity wall to the wall itself increased surviving fraction from 3.4 ± 0.9% to 14.5 ± 3.6%, while increasing RBE from 3 to 10 reduced it only from 4.9 ± 1.2% to 2.0 ± 0.5%.
- Rapid repopulation substantially weakened predicted control. With a 14-day doubling time, surviving fraction was approximately 30% when repopulation began immediately but below 1% when its onset was delayed to day 50.
CLINICAL TAKEAWAY
The model suggests that catheter geometry, margin coverage, and treatment timing may matter more than simple activity escalation for local actinium-225 therapy. Translation is distant: the simulation assumed homogeneous isotropic tissue, omitted biological clearance, backflow and explicit daughter migration, modelled one treatment cycle, and did not estimate normal-brain toxicity.