1
ASSOCIATION BETWEEN PATHOLOGICAL AND MRI FINDINGS IN MULTIPLE
SCLEROSIS: AN UPDATE
1,2
Massimo Filippi, MD,
3
Wolfgang Brück, MD,
4
Declan Chard, PhD FRCP,
5
Franz Fazekas, MD,
6
Jeroen J. Geurts, PhD,
5
Christian Enzinger, MD,
3
Simon Hametner, PhD,
7
Tanja Kuhlmann, MD,
1,2
Paolo Preziosa, MD,
8
Àlex Rovira, MD,
9,10
Klaus Schmierer, MD,
3
Christine Stadelmann, MD,
1,2
Maria A. Rocca, MD, for the Attendees of the Correlation between Pathological and MRI
findings in MS workshop*.
1
Neuroimaging Research Unit and
2
Dept. of Neurology, Institute of Experimental Neurology,
Division of Neuroscience, San Raffaele Scientific Institute, Vita-Salute San Raffaele University,
Milan, Italy;
3
Institute of Neuropathology, University Medical Center, Göttingen, Germany;
4
Queen
Square MS Centre, NMR Research Unit, Dept. of Neuroinflammation, UCL Institute of Neurology,
London, UK and National Institute for Health Research (NIHR) University College London
Hospitals (UCLH) Biomedical Research Centre, UK;
5
Dept. of Neurology, Medical University of
Graz, Graz, Austria;
6
Dept. of Anatomy & Neurosciences, MS Center Amsterdam, Amsterdam
Neuroscience, VU University Medical Center, Amsterdam, The Netherlands;
7
Institute of
Neuropathology, University Hospital Münster, Münster, Germany;
8
Section of Neuroradiology and
MRI Unit, Department of Radiology, Hospital Universitari Vall d'Hebron, Universitat Autonoma de
Barcelona, Barcelona, Spain;
9
The Royal London Hospital, Clinical Board Medicine
(Neuroscience), Barts Health NHS Trust, London, UK;
10
The Blizard Institute (Neuroscience),
Barts and The London School of Medicine & Dentistry, Queen Mary University of London,
London, UK.
Key words: Multiple Sclerosis; Magnetic Resonance Imaging; Pathology.
Word count: 4874.
References: 127.
Illustrations: 4. Appendix Table, Figure and Panel.
*The full list of participants is given in the Appendix.
Correspondence should be addressed to: Prof. Massimo Filippi, Neuroimaging Research Unit,
Institute of Experimental Neurology, Division of Neuroscience, San Raffaele Scientific Institute,
Vita-Salute San Raffaele University, Milan, Italy. Telephone number: #39-02-26433033; Fax
number: #39-02-26435972; E-mail address: filippi.massimo@hsr.it
2
Summary
Pathological evaluation is the gold standard for identifying multiple sclerosis (MS)-related
processes that explain clinical disease manifestations and for guiding the development of new
treatments. However, there are intrinsic limitations to the techniques employed, including the
limited amount of donors available, samples often representing uncommon cases, and impossibility
of follow-up. Correlative studies have demonstrated that MRI is sensitive to the different
pathological substrates of MS (inflammation, demyelination and neuro-axonal loss). The role of
MRI in evaluating other pathological processes, such as leptomeningeal involvement, central vein
and rim of lesions, microstructural abnormalities and iron accumulation as well as recovery
mechanisms, has been recently investigated. While techniques used for quantifying pathological
processes in different regions of the CNS have advanced the diagnosis and monitoring of MS
disease course and treatment, new perspectives and questions have emerged, including how
different pathological processes interact over the disease course and when remyelination might
occur. Addressing these questions will require longitudinal studies using MRI in large cohorts of
patients with different phenotypes.
3
Introduction
Pathological assessment is the gold standard in multiple sclerosis (MS) for understanding
the processes involved in the disease and its progression, and for the definition of possible future
therapeutic targets. Magnetic resonance imaging (MRI) is a strong partner technology for in vivo
evaluation of pathology, due to its high sensitivity in revealing MS-related abnormalities,
monitoring disease progression and the effects of treatment. To improve the specificity of MRI in
detecting pathological features in MS, association between pathology and MRI needs to be
continuously verified. Imaging methods with a high specificity for MS pathology could then be
applied longitudinally.
In 2012, a review paper
1
based on a workshop held by a group of international experts in
neuropathology and neuroimaging in MS summarized the evidence from correlative pathological-
MRI studies available at that time. The aim was to discuss concordant findings, but also to highlight
controversies, identifying emerging pathological and MRI findings and open questions for future
research, which included the need to improve imaging specificity, the acquisition of high-resolution
images, the combination of MRI methods and the assessment of specific processes (e.g.,
remyelination and iron abnormalities). Thanks to the improvement of MRI technologies and larger
availability of MRI scanners since 2012, substantial advancements in the field have been obtained.
This has allowed better understanding and MRI monitoring of MS-related specific pathological
processes, including the presence of central vein sign (CVS) and hypointense rim, the
heterogeneous damage in different CNS regions and iron accumulation, but also mechanisms of
tissue recovery.
This review provides a summary of the advances in the state-of-the-art of MRI techniques
over the past five years and how these correlate to pathological observations. In particular white
matter (WM) and gray matter (GM) lesions, normal-appearing brain tissue (NABT) abnormalities,
damage of relevant structures (e.g., spinal cord, thalamus, cerebellum, and hippocampus) and iron
accumulation were the main topic selected to be discussed during a second workshop (Milan, Italy;
4
November 2017), since they represent the main areas of progress in filling the gap between
pathology and MRI. Emerging pathological and MRI findings that, in combination, might enhance
our understanding of disease pathophysiology and help identify reliable in vivo markers for
monitoring different aspects of MS are also presented.
White matter lesions
Focal WM lesions, characterized by inflammation and demyelination, are the most obvious
hallmark of MS histopathology. The inflammatory infiltrate mainly consists of blood-derived
monocytes and microglia (for simplicity subsequently termed here as phagocytes), T- and B-cells.
2
In active and demyelinating lesions, commonly found at the beginning of the disease—as
documented in biopsy studies in which the onset of symptoms provides an estimate of lesion
development—phagocytes with a round and foamy morphology are the dominating inflammatory
cell population (Figure 1).
3,4
In the very initial stages of lesion formation, these phagocytes contain
myelin debris in their cytoplasm suggesting ongoing myelin breakdown as these myelin proteins are
degraded within days in vitro by monocytes.
4,5
During later disease stages, mixed active/inactive
lesions with a complete or partial rim of phagocytes and inactive lesions with an almost complete
lack of phagocytes become more prominent (Figure 1).
3,6
The percentage of mixed active/inactive
lesions is statistically significantly higher in patients with secondary (SP) and primary progressive
(PP) MS compared to relapsing-remitting (RR) MS and correlates with disease severity,
3,6
therefore
MRI correlates that are able to identify this lesions type might be especially valuable to predict the
prognosis. The histopathology in early active and demyelinating lesion stages is heterogeneous with
respect to the absence or presence of complement depositions, immunoglobulins and
oligodendroglial loss, suggesting that different pathological mechanisms may trigger lesion
formation. Whether these different histopathological patterns are patient- or lesion-stage specific is
still a matter of debate.
5
Axonal damage is closely associated with inflammatory demyelination. In the initial lesion
stages, axonal spheroids, indicators of transient or permanent axonal transport disturbance and thus
axonal damage, are already present. A retrospective study of brain tissue from 39 patients with MS
has demonstrated a close correlation between phagocytic infiltration and axonal damage.
7
The
formation of new myelin sheaths around demyelinated axons (remyelination) is frequent in active
lesions, and de- and remyelination may occur in parallel within the same lesion.
8
In a study of 3188
tissue blocks (7562 MS lesions) from the autopsy cohort of the Netherlands Brain Bank, the
percentage of remyelinated lesions was higher in RRMS than progressive MS patients.
6
A
prerequisite for remyelination is oligodendrocyte progenitor cells that mature to myelinating
oligodendrocytes upon demyelination. Whether there is ongoing remyelination in long-lasting MS
lesions is a matter of debate; however, results from animal and post mortem studies suggest that
remyelination capacity decreases with age and disease duration and that the higher proportion of
remyelinated lesions in patients with longer disease duration might be due to a less severe
disease.
6,9,10
Whether the decrease in remyelination capacity is due to decreased activity of
phagocytes or to age-associated intrinsic changes in oligodendrocytes has yet to be determined.
8,11
However, remyelination is relatively infrequently observed in MS patients with a disease duration
of more than 10 years. Only about 20% of the lesions are completely remyelinated (so-called
shadow plaques), whereas in the vast majority of lesions, remyelination is either absent or limited to
a small rim at the lesion border (Figure 1).
12
T2-weighted (including fluid-attenuated inversion recovery [FLAIR]) imaging is highly
sensitive in detecting focal WM MS lesions, enabling a prompt and accurate diagnosis of MS in
patients presenting with clinical syndromes typical of demyelination. However, overreliance on
MRI, particularly in the context of atypical clinical symptoms, frequently leads to overdiagnosis of
MS, mainly due to a lack of a careful exclusion of alternative diagnoses
13
and an inappropriate
application of diagnostic criteria to define dissemination in space and time.
14