Comparison of different methods for post-therapeutic dosimetry in [177Lu]Lu-PSMA-617 radioligand therapy

BACKGROUND: Dosimetry is of high importance for optimization of patient-individual PSMA-targeted radioligand therapy (PSMA-RLT). The aim of our study was to evaluate and compare the feasibility of different approaches of image-based absorbed dose estimation in terms of accuracy and effort in clinical routine. METHODS: Lu]Lu-PSMA-617 (range 3-10.9 GBq). Dosimetry was performed by use of the following approaches: 2D planar-based dosimetry, 3D SPECT/CT-based dosimetry, and hybrid dosimetry combining 2D and 3D data. Absorbed doses were calculated according to IDAC 2.1 for the kidneys, the liver, the salivary glands, and bone metastases. RESULTS: Mean absorbed doses estimated by 3D dosimetry (the reference method) were 0.54 ± 0.28 Gy/GBq for the kidneys, 0.10 ± 0.05 Gy/GBq for the liver, 0.81 ± 0.34 Gy/GBq for the parotid gland, 0.72 ± 0.39 Gy/GBq for the submandibular gland, and 1.68 ± 1.32 Gy/GBq for bone metastases. Absorbed doses of normal organs estimated by hybrid dosimetry showed small, non-significant differences (median up to 4.0%) to the results of 3D dosimetry. Using 2D dosimetry, in contrast, significant differences (median up to 10.9%) were observed. Regarding bone metastases, small, but significant differences (median up to 7.0%) of absorbed dose were found for both, 2D dosimetry and hybrid dosimetry. Bland-Altman analysis revealed high agreement between hybrid dosimetry and 3D dosimetry for normal organs and bone metastases, but substantial differences between 2D dosimetry and 3D dosimetry. CONCLUSION: Hybrid dosimetry provides high accuracy in estimation of absorbed dose in comparison to 3D dosimetry for all important organs and is therefore feasible for use in individualized PSMA-RLT.