Graded multidimensional clinical and radiological variation in patients with Alzheimer’s disease and posterior cortical atrophy

Abstract

Alzheimer's disease spans a heterogeneous collection of typical and atypical phenotypes. Posterior cortical atrophy represents one of the most striking examples, characterised by prominent impairment in visual and other posterior functions in contrast to typical, predominantly amnestic Alzheimer's disease. Whilst putative posterior cortical atrophy subtypes include dorsal, ventral, caudal and dominant parietal presentations, investigations of posterior cortical atrophy subtypes are rare. The current study establishes how the similarities and differences of cognition and neural integrity within posterior cortical atrophy and typical Alzheimer's disease (and by extension other Alzheimer's disease variants), can be conceptualised as systematic variations across a unified transdiagnostic, graded multidimensional space. Principal component analysis was applied to detailed neuropsychological cognitive and visual data from a large cohort of posterior cortical atrophy (N=93) and typical Alzheimer's disease patients (N=58). Voxel-based morphometry was then used to explore neural correlates of the emergent transdiagnostic PCA phenotype dimensions. The principal component analysis for posterior cortical atrophy extracted three dimensions, reflecting general cognitive, visuoperceptual and visuospatial impairments. Projecting typical Alzheimer's disease cases into the posterior cortical atrophy derived multidimensional space, and vice versa, revealed graded, overlapping variations between cases along these dimensions, with no evidence for categorical-like clustering of the patients. Likewise, the relationship between neural integrity and scores on the extracted dimensions was overlapping for PCA and tAD. The results are consistent with phenotypic continua spanning posterior cortical atrophy and typical Alzheimer's disease, arising from systematic graded variations within a transdiagnostic, multidimensional neurocognitive geometry.