Towards a better understanding of pseudoatrophy in the brain of multiple sclerosis patients

Progressive loss of brain volume is a constant feature of multiple sclerosis (MS) pathology. ¹ Different pathological substrates may be responsible for this process, including: (a) loss of volume within white matter (WM) lesions due to the loss of myelin, oligodendrocytes and axons and the contraction of astrocyte volume occurring during lesion maturation; (b) neuronal and glial loss in cortical grey matter (GM) lesions, which are, for the most part, invisible on magnetic resonance imaging (MRI); and (c) volume loss in normal-appearing brain tissue. The latter may itself have different pathological substrates including: Wallerian degeneration resulting from axonal transection in WM and GM lesions ¹ and slow, progressive axonal loss throughout the brain, due to diffuse inflammation and oxidative stress. ²

Since treatment of MS needs to target not only focal inflammatory lesions but also the neurodegeneration that occurs, MRI-based measurements of brain volume are routinely used to monitor the effects of disease-modifying therapies (DMTs) on brain volume. ³ Paradoxically, anti-inflammatory drugs used for the treatment of MS have often been associated with an acceleration of brain volume loss following the initiation of therapy. ³ This phenomenon, commonly referred to as ‘pseudoatrophy’, is generally assumed to be due to the resolution of inflammation. Fluid shifts (resolution of oedema) have been suggested as the basis for this. ⁴ However, evidence for decreased water content of brain, when looked for, ⁵ has not been found and changes in the volume of inflammatory cells, particularly glial cells, may be more relevant. The phenomenon of pseudoatrophy complicates the interpretation of treatment effects on brain volume and has led to some assessing the treatment effect of DMTs on brain atrophy only after the first year in clinical trials. ⁶ However, the assumption that pseudoatrophy occurs only during the first year of therapy is not necessarily valid, since the time course of pseudoatrophy is not completely understood.

In this issue of the Multiple Sclerosis Journal, Sastre-Garriga and colleagues ⁷ provide important insights into this topic by showing that MS patients who have gadolinium-enhancing lesions at baseline have a greater decrease in brain volume, in comparison to those without gadolinium-enhancing lesions, up to 24 months after initiation of therapy with natalizumab. While previous studies have shown an accelerated decrease in brain volume related to the presence of gadolinium-enhancing lesions,^8,9 the present paper is important because it demonstrates that accelerated atrophy associated with suppression of inflammation by natalizumab can continue beyond one year. This implies that the neuroprotective effect of natalizumab, as reflected by slowing of brain volume loss, is still being underestimated during the second year of therapy. Therefore, the two-year placebo-controlled observation period used in most phase III MS clinical trials, might not be long enough to fully evaluate this effect.

To fully understand the significance of brain volume loss after initiating anti-inflammatory therapy in MS, it is necessary to distinguish ‘true’ atrophy (i.e. irreversible brain tissue loss) from pseudoatrophy. This is not an easy task, since the presence of active focal inflammatory lesions associated with gadolinium-enhancement is likely to be associated with both greater pseudoatrophy and greater true brain atrophy, owing to the presence of inflammation-related neurodegeneration. Nevertheless, there have been encouraging developments.

Regional volumetric studies have suggested that pseudoatrophy is mostly confined to the WM, ⁹ where inflammatory infiltrates, glial activation and vasogenic oedema are more prominent compared to GM. If this is true, GM volume change might be a more reliable marker of the neuroprotective effects of treatment in the first year or two of therapy. However, there are technical issues related to the measurement of GM volume change that need to be worked out before GM volume change can be considered as a reliable outcome marker. Another approach is to use T2 relaxometry to differentiate true tissue loss from loss resolution of oedema. ¹⁰ However, oedema is not the only mechanism by which resolution of inflammation could be associated with decreases in brain volume, and this approach does not rule out decreases in the number or volume of inflammatory cells, such as microglia.

Despite advances in the understating of pseudoatrophy, there are important elements that still need to be elucidated. In particular, it is necessary to further investigate to what extent the pseudoatrophy in MS brains after the initiation of certain therapies may be related to the resolution of inflammation as opposed to neurodegeneration. Quantitative neuroimaging can help with this, eventually leading to a better understanding of the changes in brain volume that can be observed, even in a complex pathological condition such as MS.