DENVER—Is multiple sclerosis one disease? If it’s really several diseases, does that explain why immune therapies work only in some patients? Or do patients have lesions that evolve over time, requiring alternate types of treatment at different stages of the disease?

In an effort to confront these questions, Claudia Lucchinetti, MD, Associate Professor of Neurology at the Mayo Clinic, Rochester, Minnesota, analyzed the patterns of demyelination in a series of autopsy and biopsy specimens obtained with the cooperation of Prof. Hans Lassmann, from the Brain Research Institute in Vienna, and Prof. Wolfgang Bruck, from the Department of Neuropathology, Charité Hospital in Berlin. Their initial study, based on a large pathology sample of patients with multiple sclerosis (32 autopsies, 51 biopsies) revealed “profound heterogeneity in the immunopathological appearance of active multiple sclerosis lesions.” The group identified four different patterns, based on: 1) distribution of myelin protein loss, 2) plaque geography, 3) extent and pattern of oligodendrocyte destruction, and 4) evidence for immunoglobulin G and complement activation. According to Dr. Lucchinetti, “all the lesions from a single patient looked identical, but lesion patterns differed from person to person.”

INVESTIGATING THE PATTERNS

Pattern 3, a “distal oligodendrogliopathy,” has diffuse lesions with variable inflammation and pronounced microglial activation. Pattern 3 lesions do not directly surround blood vessels and typically retain a rim of preserved myelin around the vessels. Pattern 3 lesions are also characterized by a “striking loss of myelin-associated glycoprotein [MAG].” This preferential loss of MAG in comparison to other myelin proteins may reflect primary oligodendrocyte injury in demyelinating conditions. Unlike patterns 1 and 2, pattern 3 demonstrates limited re-myelination, with evidence of oligodendrocyte apoptosis. Pattern 3 has no complement activation.

Pattern 4 lesions have sharp macrophage borders and evidence of oligodendrocyte degeneration within a rim of normal-appearing white matter along the edge of the demyelinated plaque. Oligodendrocytes are largely absent from the lesion center, and there is minimal evidence of repair. But unlike pattern 3, pattern 4 shows no MAG loss, and unlike pattern 2, it has no complement activation.

PATTERN BEHAVIOR

In 145 patient samples, Dr. Lucchinetti and colleagues identified 19% of patients with pattern 1, 53% with pattern 2, 26% with pattern 3, and only 2% with pattern 4. Several observations suggest that clinical syndromes may correlate with specific pathologic patterns, she observed. All patients with Devic’s neuromyelitis optica had antibody/complement-mediated tissue damage indicative of pattern 2, while all patients with Balo’s concentric sclerosis had lesions of distal oligodendrogliopathy associated with pattern 3, and a small subset of patients with primary progressive disease had the primary oligodendrocyte degeneration of pattern 4. Classic multiple sclerosis, regardless of clinical course, was seen with all four patterns, she noted.

Dr. Lucchinetti speculated that the distribution of patients among the four patterns might account for the observed heterogeneity in treatment response. In a randomized, sham-controlled, double-blind study of plasma exchange in steroid-unresponsive acute attacks of central nervous system inflammatory demyelinating disease, including multiple sclerosis, only 42% of patients had a treatment response. Similarly, Keegan et al found that 44% of patients responded to plasma exchange. Dr. Lucchinetti suggested that one could possibly “predict that patients with antibody/complement-mediated disease may be more likely to respond to plasma exchange since this procedure could remove pathogenic antibodies.” If so, the best candidates for plasma exchange would be pattern 2 patients, she added. In support of this theory, her preliminary, unpublished findings correlating pathology with plasma exchange response in 11 multiple sclerosis patients revealed that all treatment responders (n = 6) had pattern 2 lesions. These findings remain to be corroborated in a larger series of cases.

ON THE TRACK OF A TREATMENT

A pilot study of 54 patients supported this hypothesis; Dr. Lucchinetti and colleagues found a strong correlation of lesion patterns 1 and 2 with T2-hypointense rims and gadolinium ring enhancement on MRI. Pattern 3 lesions may have a problem in the oligodendrocytes that precedes blood-brain barrier breakdown which may be responsible for the absence of early gadolinium enhancement.

NR

Return to table of contents