Analysis of the influence of In on Y-bremsstrahlung SPECT based on Monte Carlo simulation.

Objective: Y-ibritumomab tiuxetan (Zevalin) which is used for the treatment of malignant lymphomas can be used for SPECT imaging based on bremsstrahlung from Y beta particles. However, gamma rays emitted by In, which is administered to evaluate the indication for the treatment, contaminate the Y bremsstrahlung images. Our objective is to investigate the influence of In on the Y SPECT images using Monte Carlo simulation.

Methods: We used an in-house developed simulation code for the Monte Carlo simulation of electrons and photons (MCEP). Two hot spheres with diameters of 40 mm were put in an elliptical phantom. Both spheres ("sphere 1" and "sphere 2") were filled with Y and In mixed solutions. The activities of Y in sphere 1 and sphere 2 were 241 and 394 kBq/mL, respectively, and the ones of In were 8.14 and 13.3 kBq/mL, respectively. The background activity of Y was 38.6 kBq/mL, whereas that of In was 1.30 kBq/mL; moreover, the acquisition time was 30 min. Two energy windows were used: one is 90-190 keV included the In photopeak; the other is 90-160 keV. To evaluate the quality of the SPECT images, the contrast recovery coefficient (CRC) and the constant noise ratio (CNR) of the SPECT images were derived.

Results: For the energy window between 90 and 160 keV, the In count was 74 % of the total. In that case, the CRC values were 30.1 and 30.7 % for "sphere 1" and "sphere 2", respectively, whereas the CNR values were 6.8 and 12.1, respectively. For the energy window between 90 and 190 keV, the In count reached 85 % of the total count. The CRC and CNR values were 38.6 and 40.0 % and 10.6 and 19.4, respectively.

Conclusions: Our simulation study revealed that the cross talk between In and Y in SPECT imaging is rather serious. Even for the energy window excluding the In photopeak, the count ratio of Y was less than 30 % of the total. However, the influence of In on Y-SPECT imaging cannot be ignored, and the count ratio because of In is important to estimate the density of Y.