Cognitive and behavioural but not motor impairment increases brain age in amyotrophic lateral sclerosis

Abstract

Age is the most important single risk factor of sporadic amyotrophic lateral sclerosis. Neuroimaging together with machine learning algorithms allows estimating individuals’ brain age. Deviations from normal brain ageing trajectories (so called predicted brain age difference) were reported for a number of neuropsychiatric disorders. While all of them showed increased predicted brain age difference, there is surprisingly few data yet on it in motor neurodegenerative diseases. In this observational study we made use of previously trained algorithms of 3377 healthy individuals and derived predicted brain age differences from volumetric MRI scans of 112 amyotrophic lateral sclerosis patients and 70 healthy controls. We correlated predicted brain age difference scores with voxel-based morphometry data and multiple different motoric disease characteristics as well as cognitive/behavioral changes categorized according to Strong and Rascovscy. Against our primary hypothesis, there was no higher predicted brain age difference in the amyotrophic lateral sclerosis patients as a group. None of the motoric phenotypes/characteristics influenced predicted brain age difference. However, cognitive/behavioral impairment led to significantly increased predicted brain age difference, while slowly progressive as well as cognitive/behavioral normal amyotrophic lateral sclerosis patients had even younger brain ages than healthy controls. Of note, the cognitive/behavioral normal amyotrophic lateral sclerosis patients were identified to have increased cerebellar brain volume as potential resilience factor. Younger brain age was associated with longer survival. Our results raise the question whether younger brain age in amyotrophic lateral sclerosis with only motor impairment provides a cerebral reserve against cognitive and/or behavioral impairment and faster disease progression. This new conclusion needs to be tested in subsequent samples. In addition, it will be interesting to test whether a potential effect of cerebral reserve is specific for amyotrophic lateral sclerosis or can also be found in other neurodegenerative diseases with primary motor impairment.