Demyelination and Remyelination in Multiple Sclerosis (MS) Detected by Brain Amyloid PET-CT

Multiple sclerosis (MS) is an autoimmune disease affecting the central nervous system with consequent onset of various sensory, motor, sphincter and sometimes cognitive symptoms. It is characterised by the presence of lympho-monocytic inflammatory infiltrates, demyelination, axonal damage, activation of innate immunity, astrogliosis and remyelination. In the last few years, Magnetic Resonance Imaging (MRI) has made enormous progress in the diagnosis and follow-up of patients, highlighting encephalic and spinal lesions, their number and relative lesion burden, location and cortical involvement as well as the presence of blood-brain barrier alterations in the earliest stages of the disease, associated with the activation of acquired immunity with subsequent inflammatory damage.

The demyelination plaques have on their outer edge a thin wall of iron-containing microglial cells that can be detected by high-resolution magnetic susceptibility MRI sequences and techniques. Such plaques, known by neuropathologists as 'smoldering' plaques, are characterised by a concentrically widening periphery causing myelin and neuronal damage while in the centre of the lesion the process of myelin loss becomes complete. Smoldering plaques, more frequent in progressive forms of the disease, are peculiar of MS compared to other demyelinating diseases and are associated with a negative prognostic significance.

Positron Emission Tomography (PET) is a technique that allows to analyze the in vivo metabolism of certain components of the central nervous system, such as neurons and glia, and, in particular cases, to highlight synaptic function and abnormal protein accumulations. In recent years, PET has been used in MS especially for research purposes: it allows to assess the degree of demyelination and remyelination by using amyloid tracers that bind to the white substance, as well as the degree of microglial activation or synaptic impairment applying particular tracer radio isotopes (TPSO ligands or the glycoprotein 2A of synaptic vesicles). The combination of MRI and PET could provide detailed information not only on the degree and extent of the nervous system impairment, but also on the most affected cell populations, on the metabolic activity of the affected encephalic areas and on the degree of success of the damage repair attempts.

The aim of this research project is to study, applying MRI and PET techniques, a population of progressive MS patients (PMS) with a relatively significant impairment and a population of patients with a relapsing-remitting form of the disease (RRMS) with a less significant neurological impairment. The purpose is to study chronically active or 'smoldering' plaques, with magnetic susceptibility MRI techniques and PET techniques with amyloid tracers, chronically inactive plaques and more recent T2/Flair positive plaques, with or without contrast enhancement, in order to verify and quantify their presence and to correlate their number, morphology and location with the degree of disability, the clinical course, the evolution of the disease and the laboratory data, evaluating in each case the extent of demyelination and remyelination process and the degree of activation of innate immunity. The data obtained will be processed and evaluated to identify MRI and PET differences in the various disease courses and to obtain information on the different desease development paths in patients.

Study design Enrolled patients will undergo a brain MRI, a brain PET scan with amyloid tracer and functional assessment scales at the time of recruitment.

All patients will undergo clinical evaluation and various tests:

• EDSS and MS Functional Composite Scale
• Mini-BESTest
• TUG test, MSWS-12, Berg Balance Scale (BBS)
• The nine hole peg test (9-HPT) for right and left upper limb.
• Cognitive functions assessed by Brief International Cognitive Assessment for MS (BICAMS) and MMSE

Laboratory analysis of serum markers will be performed. Specifically, neuronal and glial cytoskeletal proteins (e.g. Nf-L, pN-FH, GFAP) as brain damage indices (axonal and cellular damage), will be considered and quantified.

The levels of specific neurotrophic factors (e.g. BDNF, GDNF) and cytokines (e.g. TNFα, IL-6, IL-1β, interferon) will be determined; there is evidence of their involvement in synaptic plasticity, cognitive and motor functions, and in the neuroinflammatory processes typical of MS.

MRI protocol Examinations will be performed on a 3T MRI scanner using a 32-channel head coil with sequences optimised and harmonised by the IRCCS Neuroscience Network.

The sequences that will be acquired will be:

1. T1-weighted sagittal volumetric image (MP-RAGE, 1 mm resolution) to study the morphology and encephalon volumetry
2. T2-FLAIR weighted sagittal volumetric image (1 mm resolution) for lesion localisation and contouring
3. a sequence of diffusion-weighted images (EPI, 2.5 mm resolution, b=1000 and 2000 s/mm2, 30 directions per shell) for the study of tissue microstructure and for the identification of pathology-relevant white matter bundles by tractography
4. a 3D multi-echo gradient echo sequence for quantitative susceptibility mapping (1 mm resolution, 8 echoes) for the study of local susceptibility
5. an fMRI resting-state sequence (EPI, 3 mm resolution, 200 volumes with a temporal resolution of 2.4 s) for the study of functional resting state networks.

PET protocol

The PET/CT study will be performed using a dual time acquisition protocol:

• Early acquisition Early acquisition starts with the patient already positioned on the PET/CT bed and venous access available.

A low-dose CT scan is performed for attenuation correction, the PET acquisition is started in list mode and a couple of seconds after the start, 300 MBq of 18F-Florbetaben is administered followed by washing with 10 ml of saline.

Early acquisition will last 30 minutes.

• Late acquisition A low-dose CT scan is performed to correct for attenuation and anatomical localisation, followed by a 20-minute list mode PET acquisition so that images are acquired after 90 minutes (tolerance: +10 minutes) of radiopharmaceutical uptake.

All PET reconstructions will have the following characteristics:

• 256 x 256 matrix, FOV 30 cm
• 3D-OSEM reconstruction algorithm with TOF and PSF and with standard corrections for decay, attenuation, scatter, dead time
• 8 iterations, 32 subsets, Gaussian filter with FWHM of 5 mm. Raw data will be available for possible different iterations, decided a posteriori, to improve image quality.

Patients do not need to be fasting or suspending current therapies.

MRI and PET evaluations MRI

1. Automated encephalic morpho-volumetric study (Freesurfer), by acquisition of conventional sequences (T2w, FLAIR 3D, T1w-3D), harmonised according to RIN protocol:
2. cortical thickness;
3. global brain atrophy;
4. lesions burden;
5. Normal-Appearing White Matter (NAWM).
6. Detection and study of chronic "smoldering" lesions using the QSM technique, which allows both qualitative and quantitative assessments of the extent of intralesional iron accumulation:
7. Qualitative assessment:

• Identification of Rim+ lesions, defined as "non-enhancing lesions with hyperintense perilesional rim", with iso-ipointense core;
• Calculation of the global volume of Rim+ lesions (in mm3), with subsequent separate measurement of core and Rim.

1. Quantitative assessment

• calculation of the global susceptibility of the Rim+ lesions (expressed in ppb: parts per billion), compared to the CSF reference value;
• calculation of susceptibility of Rim and core, separately.

1. Active lesions evaluation by assessing:
2. Number of lesions showing contrastographic enhancement (qualitative assessment);
3. Type of enhancement (nodular or shell-enhancement). The various parameters will be evaluated by three operators (radiologists with expertise in neuroradiology), with consensus agreement and a double check at 4 months after the first evaluation.

PET A slightly abbreviated protocol will be used with a 30-minute dynamic acquisition immediately after the injection of 18F-Florbetaben followed by a 20-minute delayed acquisition (this performed between 90 and 110 minutes after injection). The tracer binds to NAWM and in minor extent to the actively demyelinating plaques, which will be quantified. In addition, a semi-quantitative objective methodology will be used to obtain data on the degree of demyelination of the different cases examined by comparison of the data of the region of interest under investigation with NAWM uptake and a reference area uptake.

In particular, the semi-quantification foreseen is a SUVr where the uptake ROIs are the lesions pre-segmented by the neuroradiologist on MRI (FLAIR and T1 sequences), while for the reference ROIs several possibilities will be evaluated: the contralateral uptake ROIs (i.e. in the white matter if apparently normal on MRI), the white matter hot-spots in the late acquisition, the white matter average with the exclusion of the uptake ROIs. The dynamic uptake profile from the early acquisition will also be assessed, using different temporal reconstructions.

This semi-quantitative evaluation will be carried out in collaboration with colleagues from Nuclear Medicine and the National Institute of Nuclear Physics in Genoa, who have extensive experience in the field and have recently published semi-quantitative methods for the evaluation of PET investigation in Alzheimer's disease.

Sample size

This is an exploratory study of correlation between clinical, laboratory, MRI and PET findings. Considerations on sample size are omitted in the current study and may be included and perfectionated in future confirmatory studies.

Expected results The main objective of the study will be to verify the presence of correlations between PET and MRI images indicative of chronic inflammatory activity (smoldering plaques), of apparent absence of inflammatory activity (silent plaques without microglial Rim) or indicative of more recent inflammatory activity, in contrast-enhancing areas or in T2/Flair positive areas of not distant onset.

PET data will be correlated with conventional MRI parameters, and with clinical and laboratory data. The finding of significant correlations would support the role of PET as a useful tool for evaluating demyelination and remyelination activity, which may be characteristic of a certain disease phase or even possibly of specific types of individual patients.

Particular attention will be paid to study chronic lesions considering the promising results reported in recent studies showing the possibility to distinguish silent lesions from active chronic lesions (termed "smoldering") characterised by a core of complete demyelination and a periphery with activated microglial/macrophage elements.

The presence of this 'peripheral rim' can be quantitatively assessed in vivo by QSM (Quantitative Susceptibility Mapping) magnetic resonance sequences with 3T magnetic field study with good reliability compared to the histological correlate.

The association and fusion of structural MRI imaging data (related to both areas of demyelination and iron deposits) with PET data (related to amyloid deposits) may better allow to identify, within a cohort of patients with chronic demyelinating lesions, smoldering type lesions where demyelination process is still in progress and which therefore, in addition to indicating a predictive factor for the failure of the remyelination process, correlate with a longer duration of the disease and a more progressive course.

Dosimetry PET/CT examination with 18F-Florbetaben is justified in view of the risk/benefit ratio for the patient. Following administration of the radiopharmaceutical, the patient will inevitably receive a certain dose of radiation. The activity to be administered was set at 300 MBq, which is below the upper limit of 360 MBq, in order to obtain a PET examination of good diagnostic accuracy with a reasonable radiation dose.