Image-based post-SIRT Y-90 microspheres' activity quantification with a LAFOV total-body PET/CT

Ziel/Aim: Since its introduction, LAFOV PET/CT scanners with their inherent high sensitivity became favorable for post-SIRT Y-90 imaging for dose validation. For the dosimetry itself, it is of vital importance to know exactly how much of the prescribed Y-90 activity has been actually injected. Residual activity measurements are performed routinely but they can be inaccurate. Image-based quantification can be an easy way to assess the injected activity. In this study, we compare the measured against the image-based injected activity for a large patient cohort.

Methodik/Methods: This retrospective analysis included 43 patients with liver cancers. Post-SIRT they received a 20-min PET/CT scan (Biograph Quadra) which was calibrated for Y-90. The data were decay-corrected to the time of injection and all relevant corrections were applied. Syngo.via was used for the image analysis. The manually drawn 3D spherical VOI did not have any activity threshold, and the activity was calculated by the product of the mean activity concentration (Bq/ml) with the volume (ml)

Ergebnisse/Results: We were able to identify 3 cases with a discrepancy of 30 to 35% between measured and image-based injected activity and after linear fitting of all the data (without the prior 3 aforementioned which they were regarded as outliers) the image-based activity was 90% (slope: 0.907) of the measured with a strong Pearson correlation coefficient (r=0.99). Possible causes for the difference is the PET accuracy margin itself, operator dependent drawing of the VOI and impact of Y-90 scatter profile that has not been accurately accounted for with the existing scatter correction algorithm. Still the 10% error might be acceptable for the clinical needs in order to confirm if most of the prescribed activity is indeed infused inside the liver.

Schlussfolgerungen/Conclusions: LAFOV total body PET with its significantly higher sensitivity compared to SAFOV scanners, apart for post-treatment SIRT dosimetry validation, can be used for image-based quantification of the injected activity in the FOV. Future study could involve assessment of the quantified activity at decreased simulated scan times e.g. at 5 minutes and analysis of the impact of the scatter correction algorithm used for the quantification.

Publication History

Article published online:
12 March 2025