Latent atrophy factors related to phenotypical variants of posterior cortical atrophy

<p><b>Objective:</b> To determine whether atrophy relates to phenotypical variants of posterior cortical atrophy (PCA) recently proposed in clinical criteria; dorsal, ventral, dominant-parietal and caudal, we assessed associations between latent atrophy factors and cognition.</p> <p><b>Methods:</b> We employed a data-driven Bayesian modelling framework based on latent Dirichlet allocation to identify latent atrophy factors in a multi-center cohort of 119 individuals with PCA (age:64<span>±7</span>, 38% male, MMSE:21<span>±5, 71% amyloid-β-positive, 29% amyloid-β status unknown</span>). The model uses standardized gray matter density images as input (adjusted for age, sex, intracranial volume, field-strength and whole-brain gray matter volume) and provides voxelwise probabilistic maps for a predetermined number of atrophy factors, allowing every individual to express each factor to a degree without <i>a-priori</i> classification. Individual factor expressions were correlated to four PCA-specific cognitive domains (object-perception, space-perception, non-visual/parietal functions and primary visual processing) using general linear models.<b> </b></p> <p><b>Results: </b>The model revealed four distinct yet partially overlapping atrophy factors; right-dorsal, right-ventral, left-ventral, and limbic. We found that object-perception and primary visual processing were associated with atrophy that predominantly reflects the right-ventral factor. Furthermore, space-perception was associated with atrophy that predominantly represents the right-dorsal and right-ventral factors. However, individual participant profiles revealed that the vast majority expressed multiple atrophy factors and had mixed clinical profiles with impairments across multiple domains, rather than displaying a discrete clinical-radiological phenotype.</p> <p><b>Conclusion:</b> Our results indicate that particular brain-behavior networks are vulnerable in PCA, but most individuals display a constellation of affected brain-regions and symptoms, indicating that classification into four mutually exclusive variants is unlikely to be clinically useful.</p>