Estimating Cell Survival In Ac-225-PSMA Radionuclide Therapy Using PBPK Modeling, Monte Carlo Simulation, and the Two-Lesion Kinetic Model Considering Time-Dependent Dose Rate
Abstract
Purpose
This study aims to estimate prostate cancer cell survival following Ac-225-PSMA targeted radionuclide therapy by integrating physiologically based pharmacokinetic (PBPK) modeling, Monte Carlo simulations, and the two-lesion kinetic (TLK) radiobiological model, while explicitly accounting for temporal variations in dose rate.
Methods
The spatiotemporal distribution and time–activity curve of Ac-225-PSMA in prostate cancer tumors were first simulated using a PBPK model. Tumor absorbed dose was subsequently calculated using Monte Carlo simulations in GATE based on the dose point kernel method. DNA damage induction, quantified as double-strand breaks (DSBs), was then simulated at the cellular level for PC3 cells using Geant4-DNA. DNA damage repair kinetics were modeled using the Belov repair model to derive the Lea-Catcheside (LC) factor, which accounts for continuous dose-rate reduction effect on survival. Finally, cell survival following Ac-225-PSMA irradiation was estimated using the TLK model incorporating the calculated LC values.
Results
PBPK modeling yielded the tumor time-integrated activity curve for Ac-225-PSMA, and subsequent Monte Carlo simulations showed that the absorbed dose delivered to a 2-cm tumor was 15.82 Gy per 1 μmol·m⁻³ of injected activity. Geant4-DNA simulations provided DSB yields and repair dynamics, from which LC values were obtained. Using the TLK model, the survival fraction (SF) of PC3 cells following 2 Gy irradiation was estimated to be 0.011 ± 0.005, in close agreement with previously reported experimental values (SF = 0.015 ± 0.005).
Conclusion
These results demonstrate that combining PBPK modeling, Monte Carlo simulations, and the TLK cell survival model—while accounting for time-dependent dose-rate effects—provides a robust framework for estimating cellular survival in Ac-225-PSMA radionuclide therapy across varying spatial and temporal activity distributions within tumors.