Experimental Radionuclide Therapy Quantitative Pinhole planar and SPECT imaging, aiming for dosimetry based on biokinetic modeling and small animal S-values

Abstract: New radionuclide therapies in humans should be preceded by experimental animal therapies, in order to evaluate the efficacy of the new therapy regime. These experimental studies have to be performed with a reliable and accurate quantification system, with properties that have been evaluated and where possible flaws have been corrected for. The aim of this work was to evaluate the SPECT pinhole collimator high resolution imaging system to investigate the possibilities of using that imaging system as an in vivo activity quantification system in vivo in a small animal model. The pinhole collimated SPECT system used in this study was a Vision SMV DST-XL two head scintillation detector system and a lead pinhole collimator with exchangeable inserts of 3 and 1 mm pinhole diameter. The imaging system was evaluated for sensitivity, contrast, spatial resolution and countrate performance using three different phantoms, a tube phantom, a line phantom and a volume phantom. The different radionuclides used were chosen according to the different calibration measurements and according to the subsequent small animal study. The radionuclides used were 99mTc, 111In and 177Lu. Each of the different calibration measurements were set up with its corresponding phantom mounted on a stand in the center of the rotational axis in the camera geometry. Data acquisition was then performed during 360 detector rotation in discrete steps to obtain a complete data set for tomographic reconstruction. The acquired information was then processed with an iterative reconstruction algorithm in a computer to generate tomographic SPECT images. The reconstruction method used was a cone beam OSEM algorithm.The pinhole SPECT investigation was then followed by an in vivo experiment utilising the knowledge from the earlier pinhole SPECT evaluation. The in vivo studies were based on a tumor model consisting of normal rats inoculated with tumor cells. The rats were injected with about 80 MBq 111In / kg and about 1 GBq 177Lu / kg. The results from the pinhole SPECT studies with the line phantom showed good spatial resolution in imaging with 3 mm pinhole collimator insert with a distance between pinhole aperture and object of 28 mm. The resulting line profile through the image was analyzed, and the resolution was measured to 3.15 mm (FWHM) and the peaks from line sources separated by 4.5 mm were easily distinguished. The measured sensitivity was 3.5 cps/MBq with the 3 mm pinhole insert for 99mTc and 3.1 cps/MBq for 177Lu on the central pinhole axis at a distance of mm from the pinhole aperture. The results from the measurements of the countrate capability of the detector system showed a non-linearity in time which was surprising. Several measurements and analyses suggested this was due to a normalization procedure included in the reconstruction algorithm. This discovery led to the development of a renormalization method to allow the use of the pinhole SPECT technique as a quantification method in future measurements. The activity quantifications in the in vivo measurements showed very good correlation with the activity measurements of the dissected tumors measured in an activity well counter. In conclusion this study has demonstrated the feasibility of using pinhole-SPECT for quantitative in vivo imaging in small animals.