2323 - A Quantitative Comparison of CT- and PET/SPECT-Based Dosimetry Following Yttrium-90 Radioembolization with Radiopaque Glass Microspheres

Materials/Methods: Following ⁹⁰Y RE with glass microspheres, three patients underwent CT imaging and either PET/CT (n=2) or SPECT/CT (n=1) imaging. Patient 1 had two hepatic tumors and patients 2 and 3 had a single tumor. For PET/CT and SPECT/CT, post-treatment dosimetry was performed using the local deposition method with known activity. For CT, a ⁹⁰Y activity distribution was calculated using a calibration curve to translate Hounsfield Units [HU] to microsphere concentration [mg/mL], followed by scaling with the net administered activity [Bq/mg]. The CT-based dose distribution was calculated through convolution of the ⁹⁰Y activity distribution and a dose voxel kernel. Dose-volume histograms were computed for perfused liver volumes (PV), tumors (T), and the perfused normal liver (NL). Extracted dose metrics included the mean dose (D_mean) as well as D_70% for T and V_100Gy for NL. Dose metrics were compared between CT- and PET- or SPECT-based dosimetry.

Results: For patient 1, T₁ and T₂ had a percent difference between CT- and SPECT-based D_mean of -2.4% and 251.3%, respectively. For patients 2 and 3, T₁ had a percent difference between CT- and PET-based D_mean of 1.2% and -59.1%, respectively. The percent difference in D_mean to PV was -12.4%, -31.1%, and -24.6% for patients 1, 2, and 3, respectively. For NL, percent differences in D_mean were -54.8%, -61.8%, and -19.3% for patients 1, 2, and 3, respectively. V_100Gy for NL was consistently smaller in CT-based dosimetry with percent differences of -75.9%, -11.8%, and -32.0% for patients 1, 2, and 3, respectively, while D_70% percent differences varied between -82.2% and 22.4%. Absolute values for D_mean, D_70%, and V_100Gy are provided in the table.

Conclusion: ⁹⁰Y RE with radiopaque glass microspheres presents a novel opportunity to evaluate treatment success through CT-based dosimetry. Due to improved spatial resolution and mitigation of respiratory motion effects, CT-based dosimetry in ⁹⁰Y RE has the potential to provide more accurate dose metrics relative to either PET- or SPECT-based dosimetry. This claim required validation in a prospective study with a large patient cohort.

Abstract 2323 - Table 1