Personalized Computational Approach Identifies Four Alzheimer’s Subtypes

Researchers presenting at this year’s Society of Nuclear Medicine and Molecular Imaging (SNMMI) Annual Meeting demonstrated that a computational model can accurately identify four subtypes of Alzheimer’s disease, which may help generate insights into the condition’s underlying biology and personalize future treatment methods.

The model utilizes a combination of genomic and tau positron emission tomography (PET) imaging data, which are analyzed using a sparse canonical correlation analysis (SCCA)-based clustering framework to identify genome variations associated with Alzheimer’s and flag disease subtypes.

Studying the genes associated with Alzheimer's subtypes is critical to inform diagnostic and treatment approaches, as the disease is genetically complex, the researchers indicated. Pathological markers of Alzheimer's, such as PET imaging of amyloid plaques and tau neurofibrillary tangles, can further help characterize the condition.

“By identifying different subtypes of Alzheimer’s disease using both imaging and genomic information, researchers could gain potential new insights into the underlying biology of the disease and its progression,” said Joyita Dutta, PhD, associate professor in the Department of Biomedical Engineering at the University of Massachusetts Amherst, in a press release discussing the research. “Understanding the specific genetic associations for each subtype could also lead to the development of personalized treatment approaches in the future.”

In the study, the research team analyzed imaging and genomics data from participants in the Alzheimer's Disease Neuroimaging Initiative (ADNI) who had undergone 18F-flortaucipir PET and single nucleotide polymorphism (SNP) genotyping.

The study cohort was comprised of 541 individuals in total, 334 of whom were cognitively normal and 207 who were cognitively impaired.

From the tau PET imaging, the researchers computed tau PET standardized uptake value ratios from ten broad brain regions. From the SNP genotyping, the research team captured 145 genome variations associated with Alzheimer's disease.

After applying the SCCA-clustering framework to these datasets, the model identified four Alzheimer's disease subtypes: medial temporal lobe (MTL)-dominant, posterior, MTL-sparing, and lateral-temporal.

Top genes associated with each subtype were also identified in the analysis.

These findings may have significant implications for future research, diagnostics, and treatments for not only Alzheimer's disease, but also other conditions, the research team concluded.

“Genomics- and imaging-guided individualized subtyping is vital for Alzheimer’s disease because different subtypes may also have distinct rates and profiles of cognitive decline, potentially affecting clinical trial outcomes and treatment response,” stated Dutta. “By combining molecular imaging information with genomics, we have created a diagnostic technique that could be truly personalized for each patient. This has potential for broad diagnostic utility across many disease types, not only Alzheimer’s disease.”

This research adds to a growing body of investigations looking at how genomic data and advanced analytics technologies can support precision medicine efforts for diseases like Alzheimer’s.

Many of these efforts are driven by the knowledge that early diagnosis can minimize the impact of Alzheimer’s disease and other dementias, resulting in advancements in blood diagnostics, pharmacological interventions, and deep brain stimulation.

However, these interventions are often still in the research and development phase, or not widely accessible to patients, leading health systems to opt for other approaches.

One such approach is artificial intelligence (AI)-based digital screening in the primary care setting to help flag early cognitive decline.

Indiana University School of Medicine and Indiana University Health are piloting this approach in collaboration with the Davos Alzheimer’s Collaborative (DAC), in an attempt to shift cognitive care from reactive to proactive.

In March, Jared Brosch, MD, a neurologist at IU Health and assistant professor of clinical neurology at IU School of Medicine, and Phyllis Ferrell, global head of external engagement for Alzheimer’s disease at Eli Lilly & Company and director of the DAC Healthcare System Preparedness initiative, spoke with HealthITAnalytics about the pilot and how it may impact outcomes for patients with Alzheimer's disease and other dementias.