Caterina Mainero Talks About Cortical Pathology in Multiple Sclerosis

November 9, 2009


Caterina Mainero, MD, PhD, Instructor in Radiology at Harvard Medical School, discusses her paper, "In vivo imaging of cortical pathology in multiple sclerosis using ultra-high field MRI" which was recently published in Neurology® (2009;73:941-948). She spoke with José G. Merino, MD, Science Editor of Can you briefly summarize the methodology and major findings of the study?

Mainero: Using 7T MRI, we compared 16 patients with multiple sclerosis (MS) and eight age-matched controls in order to study the characteristics of cortical lesions and investigate the relationship between these lesions, white matter lesion load, and clinical features. We found that, with phased array images, we could detect and characterize the four types of lesions described in neuropathological studies, and that the presence of certain types of lesions (discussed below) is associated with worse outcome. Our findings suggest that the use of 7T MRI is an important tool for the in-vivo study of cortical pathology in MS and its impact on the progression of disease. What are the neuropathological features of the gray matter plaques?

Mainero: Pathologic studies have shown that cortical lesions constitute a substantial proportion of the total brain lesion load in MS, and that these lesions are characterized by extensive demyelination and substantial neuronal, glial, and synaptic loss. Most pathological studies describe three major types of cortical lesions:

  • Type I (leukocortical) lesions that extend across both white matter (WM) and gray matter (GM)
  • Type II (intracortical) lesions that are located within the cerebral cortex but do not extend to the brain surface or to the subcortical WM
  • Type III (subpial) lesions that extend from the pial surface to cortical layers 3 and 4

Subpial lesions may appear as circumscribed, focal lesions, or extend across multiple adjacent gyri leading to a phenomenon termed "general subpial demyelination." Some studies describe a fourth kind of cortical lesions (type IV)—these lesions affect the entire width of the cortex from the pial surface but do not extend to the subcortical WM. With the exception of leukocortical lesions, cortical plaques lack the apparent tissue destruction, gliosis, and perivascular infiltration that characterize WM lesions. In your series, how prevalent were GM plaques?

Mainero: All of our patients had at least two cortical lesions. Overall, we detected 199 cortical lesions. Subpial lesions were the most frequent type of cortical plaques observed (50.2 percent), followed by leukocortical lesions (36.2 percent) and intracortical lesions (13.6 percent). The ratio of lesion types seen on 7T scans was nearly identical to that documented in neuropathological studies. What is the etiology of the gray matter lesions in MS?

Mainero: Pathological studies show that cortical lesions in patients with MS are often characterized by demyelination and axonal and neuronal loss. In our study, we did not find an association between WM lesion load and cortical lesions, but a few previous studies reported such an association. The spatial and temporal connectivity between cortical and WM disease has not been well studied in neuropathological or MRI studies. It is unclear whether cortical disease is due to axonal damage secondary to WM demyelination or a direct target of the disease process. Evidence from neuropathological studies, at least in some phenotypes of MS, suggests that the GM pathology may occur independently of WM lesions, and may even precede the development of WM pathology. Some researchers postulate that inflammatory demyelination of the WM may be triggered by an early GM-based process as seen in some viral models of MS (the "inside-out" model of MS). These two processes might have different clinical implications and respond to different therapeutic approaches. In your study, was there an association of GM plaque load or location and disease severity and progression?

Mainero: Yes, we did find that the higher the number of type III/IV cortical lesions, the greater the disability. Previous pathology data showed that patients with a higher number of type III lesions became wheelchair dependent at a younger age than patients with fewer lesions, suggesting that the frequency and extent of intracortical and subpial demyelination may contribute to disease progression in MS. Recent data in primary progressive MS show that cortical lesions are able to predict clinical disability over a medium time period. MS is also associated with GM atrophy. What is the cause of this?

Mainero: Autopsy data show that the prevalence of cortical lesions increases with the duration of disease. MRI studies, however, show cortical atrophy even in the earliest stages of the disease. Whether cortical lesions and cortical atrophy reflect two different aspects of the same degenerative process, or have different biological substrates (primary vs. secondary to WM damage), has not yet been clarified. Few studies have examined this question, and have led to contrasting results. The use of 7T MRI may allow us to identify cortical lesions in the earlier stages of MS. We are currently investigating the relationship between the different types of cortical lesions detected at 7T and cortical thickness measures in the same sample of patients with MS. Why are the studies done in routine clinical practice using 1.5T and 3T magnets relatively insensitive to the presence of GM plaques?

Mainero: Standard MRI techniques are relatively insensitive to cortical plaques because the lesions are small and there is low contrast between the plaques and the surrounding normal cortex. Compared with standard 3T scanners, ultra-high field systems (7T to 9.4T) allow a two- to three-fold improvement in image signal-to-noise ratio (SNR). Advances in multichannel radiofrequency (RF) technology provide an additional two- to six-fold improvement in SNR, with the greatest gains in the cortex. The high-resolution images acquired at 7T enable us to characterize specific lesion locations inside the cortical ribbon, as is done for histopathologic lesion typing, and to significantly reduce partial volume effects, especially in areas adjacent to the cerebrospinal fluid (CSF). The use of double-inversion recovery (DIR) sequences in lower field strength scanners has improved the detection of cortical lesions, but clear differentiation between leukocortical, intracortical, and subpial pathology has not been possible. Of course, it is possible that the combination of the higher SNR at 7T with DIR methods may increase cortical lesion conspicuity even more, once technical issues are addressed at such high field strength. How can 7T MRI help us address questions unresolved with standard imaging techniques or scanners?

Mainero: Our study shows that ultra-high field 7T MRI can visualize the cortical structure in great detail and assess and quantify cortical MS pathology, including subpial disease, in vivo. While neuropathology is still the "gold standard," in vivo studies can be repeated over time and allow whole brain assessment and quantification of cortical lesions and atrophy. They can be used to study the association of GM and WM pathologies and to evaluate the impact of cortical lesions on clinical outcome. We are planning to extend our MR studies to ex-vivo MS brain tissue samples to understand the relationship between the histological and MRI properties of cortical lesions in MS, and to determine in which conditions MRI is a valid surrogate marker for pathologically visible cortical disease. What are the implications of your findings for clinical neurologists?

Mainero: The development and application of imaging tools able to reliably detect cortical MS lesions will allow us to evaluate the impact of such lesions on disease outcome, providing a more sensitive and early marker of disease progression than other MR measures of GM involvement, including cortical atrophy. Understanding cortical lesion patterns can provide more information about differences in disease between individuals, enabling a more accurate characterization of the different MS phenotypes. This could be useful to direct future phenotype-genotype correlations, for evaluating the therapeutic response to specific treatment, and could potentially facilitate development of novel therapeutic approaches.

Author Disclosures

Dr. Mainero has nothing to disclose.

Dr. Merino performed a one-time consultation with staff from Bell, Falla and Associates.