PET imaging in MS: a marker of acute demyelination

Eur J Nucl Med Mol Imaging. 2018 Oct 21. doi: 10.1007/s00259-018-4182-1. [Epub ahead of print]

Amyloid PET as a marker of normal-appearing white matter early damage in multiple sclerosis: correlation with CSF β-amyloid levels and brain volumes.

Pietroboni AM, Carandini T, Colombi A, Mercurio M, Ghezzi L, Giulietti G, Scarioni M, Arighi A, Fenoglio C, De Riz MA, Fumagalli GG, Basilico P, Serpente M, Bozzali M, Scarpini E, Galimberti D, Marotta G.

Abstract

PURPOSE:

The disease course of multiple sclerosis (MS) is unpredictable, and reliable prognostic biomarkers are needed. Positron emission tomography (PET) with β-amyloid tracers is a promising tool for evaluating white matter (WM) damage and repair. Our aim was to investigate amyloid uptake in damaged (DWM) and normal-appearing WM (NAWM) of MS patients, and to evaluate possible correlations between cerebrospinal fluid (CSF) β-amyloid1-42 (Aβ) levels, amyloid tracer uptake, and brain volumes.

METHODS:

Twelve MS patients were recruited and divided according to their disease activity into active and non-active groups. All participants underwent neurological examination, neuropsychological testing, lumbar puncture, brain magnetic resonance (MRI) imaging, and 18F-florbetapir PET. Aβ levels were determined in CSF samples from all patients. MRI and PET images were co-registered, and mean standardized uptake values (SUV) were calculated for each patient in the NAWM and in the DWM. To calculate brain volumes, brain segmentation was performed using statistical parametric mapping software. Nonparametric statistical analyses for between-group comparisons and regression analyses were conducted.

RESULTS:

We found a lower SUV in DWM compared to NAWM (p < 0.001) in all patients. Decreased NAWM-SUV was observed in the active compared to non-active group (p < 0.05). Considering only active patients, NAWM volume correlated with NAWM-SUV (p = 0.01). Interestingly, CSF Aβ concentration was a predictor of both NAWM-SUV (r = 0.79; p = 0.01) and NAWM volume (r = 0.81, p = 0.01).

CONCLUSIONS:

The correlation between CSF Aβ levels and NAWM-SUV suggests that the predictive role of β-amyloid may be linked to early myelin damage and may reflect disease activity and clinical progression.


Figure: a T1-weighted MRI; b co-registered 18F-florbetapir PET and c T1-weighted MRI/PET fusion images (using the hot iron map); d T2-weighted FLAIR MRI; e lesion elaboration from the Lesion Segmentation Tool (yellow); f normal-appearing white matter (blue) and damaged white matter (red) segmentation on T2-weighted FLAIR MRI


Traditionally, majority of us when we hear the terms Aβ or PET imaging (positron emission tomography; developed to image amyloid deposition), we immediately think of the dementia disorders, such as, Alzheimer's disease. However, PET imaging has recently also been repurposed as an imaging marker in MS to study myelin loss and repair, utilizing its non-specific binding to the white matter - where demyelination (loss of myelin) leads to a reduction in uptake. Not surprisingly, it may be a more sensitive marker of acute white matter lesions than conventional MRI.

Here the investigators progress our understanding of amyloid markers further by asking whether a) amyloid uptake in white matter lesions and normal appearing white matter (i.e. the normal looking white matter on MRI scans) can be split according to disease activity and b) if there is a relationship between amyloid uptake and spinal fluid levels of Aβ, white matter brain volumes and disease progression?

As expected, they found that amyloid uptake in the largest white matter MS lesion for each participant was reduced compared to their normal appearing white matter; confirming its usefulness as a marker of myelin loss (this finding has been confirmed by other investigators also and is not a new finding). However, the tracer uptake was reduced in those with active disease than in non-active participants, suggesting that white matter myelin is damaged during the early course of disease activation. Moreover, this reduced uptake was not only visible in the diseased white matter but also in the normal appearing white matter - it is not unusual in pathology studies to see immune activity even in macroscopically normal looking brain! 

The normal appearing white matter amyloid uptake showed a positive correlation with white matter volume (i.e both increased in tandem), but although there was a trend toward a negative correlation with diseased white matter volume (when one was high, the other was low), it was not found to be statistically significant. There was also a positive correlation between spinal fluid Aβ levels and tracer uptake in the normal appearing white matter (the lower the CSF Aβ level, the lower the tracer uptake in the normal appearing white matter). CSF Aβ levels showed a positive correlation with PASAT2 (an assessment of cognition), but not with other measures of disability. The latter findings are not surprising as the study is small.

Overall, a nice piece of work. But one of the major limitations of PET, in my opinion, is the cost of PET relative to MRI alone. Moreover, CSF Aβ, aside from performing a lumbar puncture, is again a fraction of this cost and one can measure several other biomarkers in tandem if necessary.

Labels: , , ,