Editors' Note: Biological Subtypes of Alzheimer Disease: A Systematic Review and Meta-analysis

Abstract

In the article, “Biological Subtypes of Alzheimer Disease: A Systematic Review and Meta-analysis,” Dr. Ferreira et al. performed a systematic review and meta-analysis on Alzheimer disease (AD) subtype studies based on postmortem and neuroimaging data until July 2019. They identified 4 main subtypes of AD—typical, limbic-predominant, hippocampal-sparing (these 3 being based on the distribution of tau-related pathology and regional atrophy), and minimal atrophy AD. They concluded that 2 core dimensions of the heterogeneity were seen in AD—typicality and severity with further influence driven by a combination of protective factors, risk factors, and concomitant non-AD brain pathologies. In response, Dr. Uleman et al. identify 2 main points for expanding the work of this study—examining the details of the AD criteria used in the individual studies because this too would introduce heterogeneity and could potentially influence the review's conclusions and a reliance on 2 simple linear axes (severity and typicality), which do not capture the complex physiologic pathways of interaction that underlie additional aspects of resilience and resistance in AD. They suggest that drawing on insights about elucidating mechanisms of resilience and resistance to cognitive decline in concussion may be helpful in this regard. Responding to these comments, the authors counter that details on the AD criteria used in the studies included in their review are available in the supplementary material and briefly summarize the trends observed in these criteria. They note that inter-study variability in subtyping methods, rather than AD diagnostic criteria, were a major source of heterogeneity. They clarify that they cannot specify whether the severity and typicality axes are linear and postulate that most of the factors in their model have complex relationships to each other. The authors note the need to empirically test and expand their conceptual model to better understand heterogeneity in AD. This exchange provides a glimpse into the complexities of AD subtyping and the various interconnected factors that are at play. In the article, “Biological Subtypes of Alzheimer Disease: A Systematic Review and Meta-analysis,” Dr. Ferreira et al. performed a systematic review and meta-analysis on Alzheimer disease (AD) subtype studies based on postmortem and neuroimaging data until July 2019. They identified 4 main subtypes of AD—typical, limbic-predominant, hippocampal-sparing (these 3 being based on the distribution of tau-related pathology and regional atrophy), and minimal atrophy AD. They concluded that 2 core dimensions of the heterogeneity were seen in AD—typicality and severity with further influence driven by a combination of protective factors, risk factors, and concomitant non-AD brain pathologies. In response, Dr. Uleman et al. identify 2 main points for expanding the work of this study—examining the details of the AD criteria used in the individual studies because this too would introduce heterogeneity and could potentially influence the review's conclusions and a reliance on 2 simple linear axes (severity and typicality), which do not capture the complex physiologic pathways of interaction that underlie additional aspects of resilience and resistance in AD. They suggest that drawing on insights about elucidating mechanisms of resilience and resistance to cognitive decline in concussion may be helpful in this regard. Responding to these comments, the authors counter that details on the AD criteria used in the studies included in their review are available in the supplementary material and briefly summarize the trends observed in these criteria. They note that inter-study variability in subtyping methods, rather than AD diagnostic criteria, were a major source of heterogeneity. They clarify that they cannot specify whether the severity and typicality axes are linear and postulate that most of the factors in their model have complex relationships to each other. The authors note the need to empirically test and expand their conceptual model to better understand heterogeneity in AD. This exchange provides a glimpse into the complexities of AD subtyping and the various interconnected factors that are at play.