Accurate voxel S-value calculation for internal dosimetry of 68Ga-PSMA: comparison of XCAT phantom and personalized methods
摘要
68Ga-PSMA PET/CT is widely used for prostate cancer management, but minimizing patient radiation exposure remains a key concern. This study uses GATE Monte Carlo (MC) simulations on patient data to calculate organ-specific S-values for 68Ga-PSMA. These values are compared to those from the XCAT phantom to evaluate differences between personalized and phantom-based models, and further compared with S-values for 18F-FDG and 68Ga-DOTATATE, common clinical tracers.
MethodsThis study analyzed 68Ga-PSMA PET/CT images from ten patients. Following image registration and multi-organ segmentation, GATE MC simulations were conducted to calculate voxel S-value (VSV) maps based on the Medical Internal Radiation Dose (MIRD) schema. The XCAT anthropomorphic phantom was similarly MC simulated to derive corresponding S-values. S-values were calculated for eight target organs: bladder, spleen, lungs, kidneys, liver, stomach, heart wall, and pancreas. 68Ga-PSMA VSVs were further compared to those of 18F-FDG and 68Ga-DOTATATE.
ResultsFor 68Ga-PSMA, the highest S-value was observed in the spleen (mean: 5.42E-05 mGy/MBq·s), and the lowest in the pancreas (mean: 3.29E-06 mGy/MBq·s). The comparison between personalized and XCAT phantom-based methods showed discrepancies of less than 30%, indicating acceptable accuracy; however, the data demonstrate that personalized methods yield more accurate estimates for internal dosimetry. The S-values associated with 68Ga-PSMA were consistently lower for all evaluated target organs than those observed for 18F-FDG.
ConclusionsWhile XCAT phantom-based methods offer a practical and reasonably accurate approach for estimating internal dosimetry without relying on patient-specific data, personalized dosimetry remains essential for achieving superior accuracy. These findings emphasize the importance of individualized S-value calculations in optimizing radiation dose assessments for clinical applications.