A High-Resolution In Vivo Atlas of the Human Brain’s Cyclooxygenase-2 (COX-2) System

Poster No:

2211 

Submission Type:

Abstract Submission 

Authors:

Martin Norgaard1,2, Anthony Galassi2, Nafiseh Ghazanfari2, Adrian Lee2, Jeih-San Liow2, Sami Zoghbi2, Paul Wighton3, Granville Matheson4, Cyril Pernet5, Gitte Knudsen5, Melanie Ganz5,1, Adam Thomas2, Douglas Greve3, Paolo Zanotti Fregonara2, Robert Innis2

Institutions:

1University of Copenhagen, Copenhagen, Denmark, 2NIMH Intramural Research Program, Bethesda, MD, 3Martinos Center for Biomedical Imaging at MGH, Boston, MA, 4Columbia University / Karolinska Institutet, Solna, Stockholms län, 5Neurobiology Research Unit, Copenhagen, Denmark

First Author:

Martin Norgaard  
University of Copenhagen|NIMH Intramural Research Program
Copenhagen, Denmark|Bethesda, MD

Co-Author(s):

Anthony Galassi  
NIMH Intramural Research Program
Bethesda, MD
Nafiseh Ghazanfari  
NIMH Intramural Research Program
Bethesda, MD
Adrian Lee  
NIMH Intramural Research Program
Bethesda, MD
Jeih-San Liow  
NIMH Intramural Research Program
Bethesda, MD
Sami Zoghbi  
NIMH Intramural Research Program
Bethesda, MD
Paul Wighton  
Martinos Center for Biomedical Imaging at MGH
Boston, MA
Granville Matheson  
Columbia University / Karolinska Institutet
Solna, Stockholms län
Cyril Pernet, PhD  
Neurobiology Research Unit
Copenhagen, Denmark
Gitte Knudsen, MD, DMSc  
Neurobiology Research Unit
Copenhagen, Denmark
Melanie Ganz  
Neurobiology Research Unit|University of Copenhagen
Copenhagen, Denmark|Copenhagen, Denmark
Adam Thomas  
NIMH Intramural Research Program
Bethesda, MD
Douglas Greve  
Martinos Center for Biomedical Imaging at MGH
Boston, MA
Paolo Zanotti Fregonara  
NIMH Intramural Research Program
Bethesda, MD
Robert Innis  
NIMH Intramural Research Program
Bethesda, MD

Introduction:

Cyclooxygenase-2 (COX-2) is one of the main enzymes activated by inflammatory stimuli and is a key target for anti-inflammatory drugs [1]. Here, we present a quantitative high-resolution in vivo atlas of the spatial distribution of COX-2 in the healthy human brain, obtained using Positron Emission Tomography (PET) data. The atlas will be made publicly available to the community as part of the OpenNeuroPET project (https://openneuropet.github.io/).

Methods:

Twenty-seven healthy participants (12 males, 15 females; mean age: 36±10.36 years, range: 22-56) were scanned using a Siemens Biograph mCT PET scanner 120 min after a bolus injection of the radioligand [11C]MC1. Arterial sampling was done to determine the input function. Ten of the participants also had a blocking scan to determine levels of receptor occupancy and non-displaceable distribution volume (VND). An isotropic 3D T1-weighted MRI was acquired for all participants using a Philips Achieva 3T MRI-scanner. The PET data were head motion corrected using the BIDS application PETPrep_hmc (v.0.0.7) [2]. The MR data were processed using FreeSurfer (v.7.4.1) [3] and co-registered to the PET data using a rigid transformation and a normalized mutual information cost function. The PET data were quantified to estimate total distribution volume (VT) for each region using Logan graphical analysis in PETSurfer [4, 5] and the radiometabolite-corrected plasma curve as input function estimated via the bloodstream BIDS application [6]. Occupancy and VND were estimated using the Lassen plot [7], from which the specific distribution volume (VS) was calculated. Parametric images of distribution volumes VT and VS were created in the MNI152 (CVS volume space from FreeSurfer) as well as in surface space (fsaverage), smoothed with a 6 mm and 10 mm smoothing kernel in the volume and surface, respectively [8].

Results:

The main outcome of this work is the generation of a brain atlas of the COX-2 VT and VS (Figure 1). The regional VT estimates ranged from 1.05 to 3.43 mL/cm3 across participants and scans (average = 2.14 ± 0.4 mL/cm3) and were mostly expressed in the cortex (Figure 2). VS estimates (VS = VT - VND) were generally low in subcortical regions (Figure 2), whereas measures in cortical regions largely ranged between 0.25-0.75 mL/cm3. The estimated average occupancy across participants from the baseline-blocking condition was 72 ± 17% (range: 37% - 89%), and the average VND was 1.81 ± 0.15 mL/cm3 (range: 1.64 - 2.1 mL/cm3).
Supporting Image: cox2_atlas_fig1_ohbm2024.png
Supporting Image: cox2_atlas_fig2_ohbm2024.png
 

Conclusions:

This quantitative in vivo brain atlas of the spatial organization of COX-2 distribution volumes in the healthy human brain provides a valuable tool for investigating the COX system; in particular, the atlas can be used as a reference to patient groups with alterations in their inflammatory system or to understand the effects of pharmacological interventions that act on the COX-2 system.

Modeling and Analysis Methods:

PET Modeling and Analysis 2
Segmentation and Parcellation

Neuroinformatics and Data Sharing:

Brain Atlases 1

Novel Imaging Acquisition Methods:

PET

Physiology, Metabolism and Neurotransmission :

Pharmacology and Neurotransmission

Keywords:

Data analysis
Pain
Positron Emission Tomography (PET)
STRUCTURAL MRI
Other - Inflammation

1|2Indicates the priority used for review

Provide references using author date format

[1] Shrestha, S., Kim, M. J., …, & Innis, R. B. (2020). PET measurement of cyclooxygenase-2 using a novel radioligand: Upregulation in primate neuroinflammation and first-in-human study. J Neuroinflammation, 17(1), 140. https://doi.org/10.1186/s12974-020-01804-6

[2] Norgaard, M., Beliveau, V., Galassi, A. (2023). mnoergaard/petprep_hmc: v.0.0.7 (v.0.0.7). Zenodo. https://doi.org/10.5281/zenodo.10102262

[3] Fischl, B. (2012). FreeSurfer. NeuroImage, 62(2), 774–781. https://doi.org/10.1016/j.neuroimage.2012.01.021

[4] Logan, J., Fowler, J. S., … , & Alexoff, D. L. (1996). Distribution Volume Ratios Without Blood Sampling from Graphical Analysis of PET Data. JCereb Blood Flow Metab, 16, 834–840. https://doi.org/10.1097/00004647-199609000-00008

[5] Greve, D. N., Svarer, C., … , & Knudsen, G. M. (2014). Cortical surface-based analysis reduces bias and variance in kinetic modeling of brain PET data. NeuroImage, 92, 225–236. https://doi.org/10.1016/j.neuroimage.2013.12.021

[6] Matheson, G. J. (2019). Kinfitr: Reproducible PET Pharmacokinetic Modelling in R. bioRxiv: 755751. https://doi.org/10.1101/755751

[7] Lassen N.A., Bartenstein P.A., … , & Andersen J.V. (1995). Benzodiazepine receptor quantification in vivo in humans using [11C]flumazenil and PET: application of the steady-state principle. J Cereb Blood Flow Metab, 15,152–165. https://doi.org/10.1038/jcbfm.1995.17

[8] Nørgaard, M., Beliveau, V., … , & Knudsen, G. M. (2021). A high-resolution in vivo atlas of the human brain’s benzodiazepine binding site of GABAA receptors. NeuroImage, 232, 117878. https://doi.org/10.1016/j.neuroimage.2021.117878