NEW YORK – Using transcriptome sequence data from more than 1,500 postmortem brain samples from individuals with or without Alzheimer's disease (AD), a team led by researchers at the Icahn School of Medicine at Mount Sinai has highlighted several expression-based AD subtypes.
"Understanding the genetic and molecular differences between molecular subtypes of AD within these data will provide novel insights into disease pathogenesis and offer new avenues for developing effective therapeutics," senior and corresponding author Bin Zhang, a researcher at the Icahn School of Medicine at Mount Sinai and director of Mount Sinai's Center for Transformative Disease Modeling, and his colleagues wrote in a study published in Science Advances on Wednesday.
The researchers relied on RNA sequencing data for some 1,543 postmortem brain samples from the Mount Sinai-JJ Peters Veteran Affairs Medical Center Brain Bank (MSBB-AD) and the Religious Orders Study-Memory Aging Project (ROSMAP), including samples spanning four brain regions from hundreds of individuals with or without AD.
The team's analyses on transcriptome data for more than 900 samples from the frontal pole (FP), superior temporal gyrus (STG), parahippocampal gyrus (PHG), and inferior frontal gyrus (IFG) brain regions in 364 MSBB-AD participants with or without AD or related dementia suggested that AD-related gene expression shifts appear to be most pronounced in the PHG brain region.
From there, the researchers focused in on differential gene expression patterns in the PHG, adjusting for AD stage and severity. Their results pointed to five PHG expression-based subtypes of AD, falling into three main clusters, along with related molecular signatures, clinical features, and potential driver genes.
Those subtypes were marked by altered expression of pathways related to everything from tau-mediated neurodegeneration and amyloid-beta neuroinflammation to synaptic signaling and myelination, the researchers reported — results that they shored up with data for postmortem brain samples from another 615 AD cases or controls in ROSMAP.
"Such differences strongly suggest there are subtypes of AD with different biological and molecular factors driving disease progression," Zhang said in a statement.
He added that while the current findings need to be verified in larger studies, the results so far "lay down a foundation for determining more effective biomarkers for early prediction of AD, studying causal mechanisms of AD, developing next-generation therapeutics for AD, and designing more effective and targeted clinical trials, ultimately leading to precision medicine for AD."
With the data at hand from the MSBB-AD cohort, for example, the investigators came up with an expression-based classifier to slot AD cases from ROSMAP into one of the three PHG subtypes, before exploring related expression-based subtypes experimentally in mouse models of AD.
The authors noted that some, but not all, of the brain expression shifts found in the AD mouse models lined up with those identified in the human samples. They suggested that those results "may partially explain why a vast majority of drugs that succeeded in specific mouse models do not align with generalized human trials across all AD subtypes," and called for additional research to better classify AD cases and personalize treatment for individuals with the disease.
"The remaining challenges for future research include replication of the findings in larger cohorts, validation of subtype-specific targets and mechanisms, identification of peripheral biomarkers, and clinical features associated with these molecular subtypes," Zhang said.