Multi-organ kinetic modeling for Na[18F]F pre-clinical total-body PET studies

Jose Benitez-Aurioles; Paul S Clegg; Carlos J Alcaide-Corral; Catriona Wimberley; Adriana A S Tavares

doi:10.1002/mp.17499

Multi-organ kinetic modeling for Na[¹⁸F]F pre-clinical total-body PET studies

Med Phys. 2024 Nov 5. doi: 10.1002/mp.17499. Online ahead of print.

Authors

Jose Benitez-Aurioles^{1

2}, Paul S Clegg¹, Carlos J Alcaide-Corral^{3

4}, Catriona Wimberley^{1

3

5}, Adriana A S Tavares^{3

4}

Affiliations

¹ School of Physics & Astronomy, University of Edinburgh, Edinburgh, UK.
² University of Manchester, Division of Informatics, Imaging and Data Science, Manchester, UK.
³ Edinburgh Imaging, University of Edinburgh, Edinburgh, UK.
⁴ University/BHF Centre for Cardiovascular Science, University of Edinburgh, Edinburgh, UK.
⁵ Centre for Clinical Brain Sciences, University of Edinburgh, Edinburgh, UK.

PMID: 39499788
DOI: 10.1002/mp.17499

Abstract

Background: Total-body positron emission tomography (PET), already well-established in the pre-clinical setting, makes it possible to study multi-parameters in biological systems as a whole, rather than focusing on single tissues analysis. Simultaneous kinetic analysis of multiple organs poses some daunting new challenges.

Purpose: To explore quantifying the pharmacokinetics of Na[¹⁸F]F in multiple dissimilar murine organs simultaneously in vivo with total-body PET imaging using different compartmental models for each organ and a shared cardiovascular system.

Methods: Six mice underwent a 60-min total-body PET scan following intravenous bolus injection of Na[¹⁸F]F. Compartmental models were constructed for each organ (heart, lungs, liver, kidneys, and bone) using an image derived input function. Non-linear least squares fitting of a model that connects the five organs to a shared cardiovascular system was used to analyze both the first 3 min of data and the full hour. Analysis was repeated 5000 times using different initial parameter values for each duration, permitting analysis of correlations between parameters.

Results: The models give a good qualitative account of the activity curves irrespective of the duration of the data; however, the quality of the fits to 3 min of data (average $χ_{ν}^{2}$ is 2.72) was generally better. Comparison of perfusion values to literature values was possible for the liver and lungs with the former (liver, 0.540 ± 0.177 mL/ml/min) being well-above expectations and the latter (lungs, 0.184 ± 0.413 mL/ml/min) in rough agreement. Correlations between microparameter values (especially affecting k₂) caused very noticeable problems for data modeling from both the kidneys and the femur.

Conclusion: The present study demonstrates an approach to performing kinetic modeling for multiple organs simultaneously with Na[¹⁸F]F. The observed correlations between microparameter values remain a challenge. Nonetheless, many microparameters can be estimated reliably with a quantitative analysis of perfusion being possible for some organs.

Keywords: Na[18F]F; calcification; kinetic modeling; perfusion; total‐body PET.

Abstract

Grants and funding