Evaluation of non-invasive PET quantification instead of invasive gold-standard methods in preclinical and clinical applications

Kerstens, Maria (2024): Evaluation of non-invasive PET quantification instead of invasive gold-standard methods in preclinical and clinical applications. Dissertation, LMU München: Medizinische Fakultät

Abstract

Background Quantification of molecular processes by means of non-invasive positron emission tomography (PET) is a challenging task of nuclear medicine, involving oncological as well as neurological and cardiological questions. In this context, the pharmacokinetic and regional concentration of radiotracers are approximated using various models and methods and characterized in terms of specific parameters and parametric PET images. In order to develop such quantifying approaches, preclinical and clinical studies are necessary in which correlations and validations are performed with mostly invasive gold-standard methods. Purpose This thesis aimed at establishing and evaluating two approaches using non-invasive PET instead of invasive gold-standard methods to visualize cellular mechanisms in the context of radiation-induced cellular damage and cerebral processes in neurological issues. Methods First approach: Longitudinal PET imaging with the proliferation radiotracer 3'-[18F]fluoro-3'-deox- ythymidine ([18F]FLT) and apoptosis radiotracer 2-(5-[18F]fluoropentyl)-2-methyl malonic acid ([18F]ML-10) was performed over a six-month period at defined time points in wild-type mice irradiated with different irradiation doses (0 Gy, 0.5 Gy, 1 Gy, 3 Gy). On the one hand, this was correlated and validated with histological and immunohistochemical examinations with Hematoxylin-Eosin (HE), Ki-67 and cleaved Caspase3 staining. On the other hand with analyses of white blood cells (WBC), red blood cells (RBC) and platelets (PLT). Furthermore, biodistribution studies with the radiotracers [18F]FLT and [18F]ML-10 were performed. Second approach: In healthy controls and patients with Alzheimer’s disease (AD) and progressive supranuclear palsy (PSP), aggregated tau deposits were visualized by [18F]PI-2620 PET and, in this context, arterial input functions were obtained by continuous sampling of radial artery whole blood. These were validated with non-invasive image-derived input functions (IDIF) generated by manual and automated extraction of the carotid artery in the corresponding PET image. Volumes of distribution (VT) and volume of distribution ratios (VTr) were calculated with the input functions using Logan plots and compared with quantitative parameters such as standard uptake value ratios (SUVr) and distribution volume ratios (DVr) determined by simplified reference tissue modeling. Results First approach: The [18F]FLT signal of the hematopoietic bone marrow correlated strongly with blood parameters, especially WBC, and histological as well as immunohistochemical data. Regarding other organs, such as the gastrointestinal tract and thymus, there were some correlations between the data of [18F]FLT PET and invasive gold-standard methods, but also unexpected, non-correlating data that need further investigation, as it does the [18F]ML-10 signal of the hematopoietic bone marrow. The biodistribution data showed strong variations with high standard deviations, which were attributed to difficulties in the technical performance. Second approach: AIF highly correlated with IDIF regardless of the manual or automated extraction method. VT revealed considerable variance across groups, which was strongly reduced by calculating VTr. VTr and DVr outperformed VT and SUVr in detecting differences between healthy controls and PSP patients, whereas all quantification parameters performed similarly in comparison of healthy controls and AD patients. Conclusion With appropriate radiotracers, non-invasive PET can visualize and quantify biological processes such as proliferation and apoptosis or tau deposition in vivo. In the course of testing and evaluating these quantifications, there are often performed validations with invasive gold-standard methods: Histology is suitable for the validation of PET imaging cellular processes, the measurement of activity concentration in tissue or blood is suitable for the validation of image-based determinations of activity concentration. It is important to be aware of sources of error and limitations, even with gold-standard methods, as the example of biodistribution showed.

• Dokumententyp: Dissertationen (Dissertation, LMU München)
• Themengebiete: 600 Technik, Medizin, angewandte Wissenschaften ; 600 Technik, Medizin, angewandte Wissenschaften > 610 Medizin und Gesundheit
• Fakultäten: Medizinische Fakultät
• Sprache der Hochschulschrift: Englisch
• Datum der mündlichen Prüfung: 13. Dezember 2024
• 1. Bericht­erstatter:in: Ziegler, Sibylle