Introduction

Positron emission tomography uses radioligands to gather information about molecular and cellular processes in health and disease. Radioligands binding to the translocator protein (TSPO), e.g. PBR28, have proved to be useful to study inflammatory brain processes that are characterized by the upregulation of TSPO in glial cells¹. Relatedly, quantitative Magnetic Resonance Imaging (i.e. relaxometry, magnetization transfer) provides metrics of tissue-level microstructure, which permit to study the consequences of brain inflammation such as oedema, demyelination, tissue degeneration and iron accumulation¹. In this work, we use the recently-developed technology of integrated PET/MRI to investigate the relationship between PBR28 uptake and T1 and T2 relaxation times with regard to the microstructure of brain tissue.

Methods

We enrolled five patients with HIV infection on stable antiretroviral therapy (age 55 ± 4.2, 4 males 1 female), who exhibited variable degrees of glial activation compared to a population of five age, gender and genotype-matched healthy controls. Brain imaging was performed with a Siemens PET/MRI scanner² consisting of a dedicated brain avalanche photodiode-based PET scanner operating in the bore of a 3T TRIO magnetic resonance scanner-equipped with an 8-channel head coil. The use of integrated PET/MRI allowed us to collect structural MRI simultaneously with the PET data. Patients were scanned with in-house produced [11C]PBR28³ produced in-house. At the beginning of the scan, we acquired (i) a multi-echo MPRAGE, (ii) a whole-brain Magnetization Prepared 2 Rapid Acquisition Gradient Echoes MP2RAGE⁴ for T1 relaxometry and a T2 relaxometry sequence covering the whole brain till the level of the occipital lobe, which uses a new nonlinear inverse reconstruction algorithm that estimates a T2 map⁵ (Table 1). For the purpose of this study, we considered brain areas with significantly higher uptake of PBR28 in patients compared to controls i.e. the thalamus and the frontal cortex. Standardized uptake value maps (SUV i.e. mean radioactivity/injected dose/weight) were computed using in-house software and used as a measure of TSPO expression^6-8. SUV, T1 and T2 maps were normalized to MNI space using nonlinear registration (FNIRT, www.fmrib.ox.ac.uk/fsl/)¹¹. In addition, SUV maps only were also spatially smoothed (8 mm) to improve signal-to-noise ratio, and intensity-normalized to a mean of 1 (SUVRs) in order to account for global signal differences in PET signal across subjects (Figure 1), as previously reported^9,10 (Figure 1). The frontal cortical brain regions and thalamus were segmented using SPM and FreeSurfer on all T1 maps¹¹. Spearman correlations were performed between T1/T2 relaxation times and SUVRs for all voxels in the thalamus of the 5 patients and all voxels of frontal lobe area that showed increased uptake. In order to minimize the effect of tissue type in the correlation measurement, we removed the WM from the frontal-lobe region-of-interest.

Results

We found a significant negative correlation between T2 relaxation times and PBR28 uptake in the thalamus (Spearman Rho=-0.26, p-value<1e-10), but no correlation between T1 and PBR28 uptake (Spearman Rho =-0.01, p-value=0.26). The analysis of the frontal lobe showed that PBR28 uptake negatively correlated with T2 relaxation times (Spearman Rho =-0.14, p-value<1e-10), and positively with T1 relaxation times (Spearman Rho =0.16, p-value<1e-10) Figure 2.

Discussion and Conclusion

Our data indicate that increased [¹¹C]PBR28 SUVR corresponds to a decrease in T2 relaxation times in both the thalamus and the cortex, possibly due to an accumulation of paramagnetic ions (i.e. iron) occurring in activated glial cells^13,14. The lack of a correlation between PBR28 SUVR and T1 relaxation time in the thalamus as well as the presence of a positive correlation reported in the cortex of the frontal lobe is possibly due to the additional sensitivity of T1 relaxometry to myelin, which is an important determinant of the T1 contrasts in the frontal lobe cortex and less in the thalamus⁵. An increase of PBR28 SUVR correlates with an increase of T1 relaxation times in the cortex (compatible with decrease in myelin content), most probably due to cortical myelin damage provoked by activated glial cells^15,16. In summary, our preliminary data suggest an association between uptake of the glial marker [¹¹C]PBR28 and changes of the T1 and T2 metrics of brain tissue microstructure. These observations support the utility of quantitative T1/T2 relaxometry as a complementary tool to PET in investigations of neuroinflammation.

Acknowledgements

No acknowledgement found.

References

1. Albrecht et al. In vivo imaging of human neuroinflammation. ACS Chem Neurosci 2016 Apr 20; 7(4):470-83.; 2. Catana, C et al. "Simultaneous In Vivo Positron Emission Tomography and Magnetic Resonance Imaging." Proceedings of the National Academy of Sciences, March 11, 2008, Vol. 105 (10), pp. 3705-3710; 3. Imaizumi et al., Brain and whole-body imaging in nonhuman primates of [11C]PBR28, a promising PET radioligand for peripheral benzodiazepine receptors. NeuroImage 2008; 4. Marques et al., MP2RAGE, a self bias-field corrected sequence for improved segmentation and T1-mapping at high field. NeuroImage 2010; 5. Sumpf et al., Fast T2 Mapping with Improved Accuracy Using Undersampled Spin-echo MRI and Model-based Reconstructions with a Generating Function. Trans. Med. Imaging 2014; 6. Hirvonen et al., Increased in vivo expression of an inflammatory marker in temporal lobe epilepsy. J Nucl. Med. 2012; 7. Fujita et al., PET reveals inflammation around calcified Taenia solium granulomas with perilesional edema. PlosOne 2013; 8. Shao et al., Imaging of carrageenan-induced local inflammation and adjuvant-induced systemic arthritis with [(11)C]PBR28 PET. Nucl. Med. Biol. 2013; 9. Loggia M. et al., Evidence for brain glial activation in chronic pain patients. Brain 2015; 10. Zuercher N. et al., Increased in vivo glial activation in patients with amyotrophic lateral sclerosis: assessed with [(11)C]-PBR28. NeuroImage Clin.2015; 11. Fischl et al., Automatically Parcellating the Human Cerebral Cortex. Cerebral Cortex 2004. ; 11. Smith et al., 2004;12. Fischl et al., Automatically Parcellating the Human Cerebral Cortex. Cerebral Cortex 2004.; 13. Pellizzoni I et al., Biochimica et Biophysica Acta (BBA) - Molecular Basis of Disease 2013; 14. Lane DJ, Biochemistry 2010; 15. Helena S. Domingues, Front Cell Dev Biol. 2016; 16. Lucchinetti CF, NEJM 2011.