Periventricular calcification added to the phenotypic repertoire of Alexander disease

Abstract

Presence of a periventricular rim is one of the diagnostic magnetic resonance imaging (MRI) findings characteristically observed in nearly all reported cases of Alexander disease, except for some juvenile- or adult-onset cases with lesions restricted to the brainstem and upper cervical cord. Previously, I suggested that the pathophysiology of periventricular lesions in Alexander disease differs from that of extensive cerebral white matter abnormalities with frontal predominance,1 because a subpopulation of astrocytes in the subventricular zone proliferate in vivo and behave as multipotent progenitor cells in vitro throughout life.2 Moreover, these astrocytes remain devoid of chain migration and are marked by the most intense production of GFAP-ε (rodent homologous GFAP-δ) that is one of the alternatively spliced isoforms and lacks the ability to self-assemble into intermediate filaments.3 However, GFAP-ε tends to form filament aggregation with the major isoform GFAP-α and αB-crystallin under an overproduced condition, suggesting its critical role in Rosenthal fiber formation around the periventricular wall in combination with mutated GFAP-α. Since the first report by Brenner et al. on heterozygous missense mutations of GFAP in 11 infantile cases and one juvenile case of Alexander disease, numerous GFAP mutations associated with Alexander disease have been reported expanding the phenotypic repertoire beyond the conventional description of leukodystrophy. Among the previous four exceptional cases of Alexander disease with atypical GFAP alterations other than missense mutations, the two juvenile-onset cases (p.R126_L127dup and p.K86_V87delinsEF) reported by van der Knaap et al.4 had a kind of garland along the ventricular wall and lacked extensive white matter abnormalities. Together with the present juvenile-onset Alexander disease with periventricular calcification associated with an unusual deletion of glutamic acid residue in GFAP, some types of non-missense minor alterations in GFAP may cause some pathognomonic reactions different from those of usual missense mutations in the subventricular population of astrocytes. Concerning the pathophysiological mechanism of the unusual periventricular calcification, the authors discussed the similarity of Alexander disease and low-grade glioma with respect to MR spectroscopy results and Rosenthal fiber formation. Then, they suggested the existence of some unknown mechanism of calcification common to Alexander disease and low-grade glioma. The explanation seems to be inadequate, because the authors made no remark about recent advances in basic research into brain calcification. Ramonet et al.5 showed evidence of a common pattern of brain calcification taking place in several human pathologies such as hypoxia, Alzheimer’s disease, Parkinson disease, astroglioma, and epilepsy. An imbalance between cellular signals of activity and energy availability for their execution is present in all these situations. According to their theory, I further speculate that a relative deficiency in vascularization and a metabolic overactivation of the periventricular astrocytes filled with Rosenthal fibers composed of GFAP with the novel non-missense mutation may be responsible for the unusual periventricular rim of calcification. The case report by Jefferson et al.6 calls on pediatric neurologists and neuroradiologists to pay attention to periventricular calcification as a possible computed tomography finding of Alexander disease, and at the same time to reconsider whether or not periventricular calcification actually occurs so rarely in Alexander disease. Since the MRI-based diagnostic criteria of Alexander disease proposed by van der Knaap et al. has been accepted widely, a clinical course of the disease is usually followed by MRI. Thus, there is a possibility that calcifying lesions of Alexander disease may be overlooked by a routine follow-up MRI examination. Therefore, the present report indicates a need for computed tomography examination at least on selected cases of late-onset Alexander disease that have a periventricular rim and brainstem abnormalities but lack extensive cerebral white matter abnormalities on MRI.