Author: Zhuo Chen, Tinsu Pan, Allan Thomas 👨🔬
Affiliation: Mallinckrodt Institute of Radiology, Washington University School of Medicine, WashU Medicine, UT MD Anderson Cancer Center 🌍
Purpose: Misregistration between data-driven gated (DDG)-PET and CT can limit the benefits of motion correction and improved localization and quantitation. DDG-CT offers a solution to these issues. Here, the effects of misregistration and motion correction on quantitation in radiopharmaceutical therapy (RPT) patient selection and dosimetry predictions were assessed with DDG-PET and DDG-CT.
Methods: Twelve patients undergoing Ga-68-PET/CT for initial staging and assessment for Lu-177-RPT were included. A low-dose cine-CT was acquired in misregistered regions to extract end-expiration (EE)-phase CT (DDG-CT). For DDG-PET, EE-phase data was derived from 50% of PET data at 30% from end-inspiration. Baseline PET/CT, PET/DDG-CT, DDG-PET, and DDG-PET/DDG-CT were all compared. Organs were segmented on both standard CT and DDG-CT, while tumors were segmented using an SUV-based threshold. Changes in volumes, SUV, and lesion glycolysis (LG=SUV·volume) were assessed. SUV per unit volume (SUVvol) was tracked to explore combined effects for dosimetry predictions.
Results: A total of 114 tumors, 8 kidney-pairs, 9 spleens, and 12 livers were within the corrected PET/CT regions. Median (range) changes in SUVmax were 4% (-19–126%), 19% (-10–94%), and 28% (-7–217%) for PET/DDG-CT, DDG-PET, and DDG-PET/DDG-CT, respectively. The combined effect of SUV and volume in LG was generally small, but SUVvol changes were 8% (-48–358%), 36% (-64–671%), and 56% (-28–900%). Effects for organs were minimal in general, with consistent increases in both SUV and volume from the PET and CT corrections primarily for the spleen.
Conclusion: Both motion and misregistration should be addressed when quantitative metrics are used in patient selection, dosimetry, or treatment response assessment in RPT. The combined effects of increased SUV and decreased volumes for tumors likely impact dosimetry predictions in a clinically relevant manner. These changes are consistent and significant, but can be dramatic (up to 900% change) for select patients and tumors.