Rare variant associations with late‐life cognitive performance

Alexandra N. Regelson, Derek B. Archer, Alaina Durant, Shubhabrata Mukherjee, Michael L. Lee, Seo‐Eun Choi, Phoebe Scollard, Emily H. Trittschuh, Jesse Mez, William S. Bush, Amanda Kuzma, Michael Cuccaro, Carlos Cruchaga, Lindsay Aa. Farrer, Li‐San Wang, Gerald D. Schellenberg, Richard Mayeux, Walter W. Kukull, C. Dirk Keene, Andrew Joel Saykin, Sterling C Johnson, Corinne D. Engelman, David A. A. Bennett, Lisa Laverne Barnes, Eric B Larson, Kwangsik Nho, Alison M. Goate, Alan E. Renton, Edoardo Marcora, Brian Fulton‐Howard, Tulsi Patel, Shannon L Risacher, Anita L. DeStefano, Julie A Schneider, Mohamad Habes, Sudha Seshadri, Claudia L Satizabal, Pauline Maillard, Arthur W. Toga, Karen Crawford, Duygu Tosun, Jeffery M. M Vance, Elizabeth C. Mormino, Charles S. DeCarli, Thomas J. Montine, Gary W. Beecham, Sarah Biber, Philip L. De Jager, Vilas Menon, Annie Lee, Adam Brickman, Badri N. Vardarajan, Christiane Reitz, Jennifer J. Manly, Jason M Fletcher, Miguel Arce Rentería, Jason Song, Qiongshi Lu, Paul K Crane, Timothy J. Hohman, Logan Dumitrescu, The Alzheimer's Disease Neuroimaging Initiative (adni) and The Alzheimer's Disease Sequencing Project (adsp)Alzheimer 2025 · DOI

TL;DR

This study identifies novel rare genetic variants associated with cognitive performance in aging, moving beyond common variants to explain some of the "missing heritability" in Alzheimer's Disease. By analyzing whole genome sequencing data from a large cohort, they pinpointed genes linked to memory, language, and executive function decline, offering new avenues for understanding and potentially treating AD.

What Problem Does This Paper Address?

Alzheimer's Disease (AD) is known to be highly heritable, yet a significant portion of this heritability remains unexplained by common genetic variants. This "missing heritability" is thought to stem from the complex nature of AD, including its clinical and neuropathological heterogeneity, and potentially from the influence of rare genetic variants. These rare variants, while individually infrequent, can have substantial effects on gene function. This paper aims to address this gap by investigating the association of rare genetic variants with different domains of cognitive performance in late life, using sensitive longitudinal measures as endophenotypes.

Methods

The researchers employed a rare variant association analysis using whole genome sequencing data from the Alzheimer's Disease Sequencing Project (ADSP). They analyzed data from 8,481 participants, focusing on individuals of European ancestry. Key methodological steps included:

  • Harmonized Cognitive Measures: Longitudinal scores for memory, executive function, and language were derived using confirmatory factor analysis models and linear mixed model regressions. This approach allows for a more nuanced understanding of cognitive trajectories over time.
  • Rare Variant Filtering: Variants with a minor allele frequency (MAF) less than 0.01 were selected. These variants were further filtered based on their predicted impact using the Variant Effect Predictor (VEP), focusing on high or moderate impact SNPs.
  • Gene-Based Association Testing: The SKAT-O (Sequence Kernel Association Test - Optimal) method was used to test for associations between genes and cognitive phenotypes. This method aggregates the effects of multiple rare variants within a gene, increasing statistical power. Genes with at least two contributing variants and a minimum aggregate minor allele count > 10 were included.
  • Statistical Adjustments: All analyses were adjusted for covariates such as sex, baseline age, sequencing center and platform, and the first five principal components of genetic ancestry to control for potential confounding factors.
  • Multiple Comparison Correction: The false discovery rate (FDR) procedure was used to correct for multiple testing.

Datasets

The study utilized data from the Alzheimer's Disease Sequencing Project (ADSP), which comprises whole genome sequencing data from multiple cohorts focused on cognitive aging. The analysis included 8,481 participants of European ancestry.

Key Insights and Ideas

  • Rare Variants Matter for Cognition: The study provides strong evidence that rare genetic variants contribute to individual differences in cognitive performance and decline, helping to explain some of the "missing heritability" in AD research.
  • Specific Genes Implicated: The identification of specific genes (e.g., GAB1, PLA2G4A, SIGIRR) associated with executive function decline is a significant finding, offering concrete targets for further investigation.
  • Functional Relevance of Identified Genes: The paper highlights the potential biological relevance of some identified genes, such as GAB1 being a potential AD treatment target and PLA2G4A showing differential expression in AD patients and aging brains. This suggests these variants may have direct mechanistic links to cognitive processes.
  • Endophenotypes Enhance Discovery: Using sensitive longitudinal cognitive measures as endophenotypes (intermediate phenotypes) proved effective in identifying associations that might be missed with cross-sectional or less precise cognitive assessments.

Key Results

The study identified 9 genes associated with cognitive domains. Specifically:

  • Memory: Two genes, APOE and PSEN1, were associated with baseline memory performance (PFDR = 0.07).
  • Language: The gene PEDS1-UBE2V1 was associated with baseline language (PFDR = 0.01).
  • Executive Function Decline: Six genes were associated with executive function decline: HPN, HPN-AS1, GAB1, CXCL3, SIGIRR, and PLA2G4A. The PFDR values for these associations ranged from 0.01 to 0.08.
  • High Impact Variants: SIGIRR, PLA2G4A, and HPN showed associations driven by high-impact rare variants, further strengthening their role in executive function decline.

Limitations and Future Work

  • Ancestry Limitations: The current analysis was limited to participants of European ancestry. The authors acknowledge that future work will incorporate other ancestries to provide a more comprehensive understanding of genetic contributions across diverse populations.
  • Functional Validation: While some genes show promising biological relevance, further functional studies are needed to fully elucidate the mechanisms by which these rare variants impact cognitive performance and AD pathogenesis.
  • ADSP Scope: The study leveraged data from the ADSP. Expanding to other large-scale sequencing initiatives could further validate these findings and uncover additional associations.
  • Explaining Heritability: While this study contributes to explaining missing heritability, it is likely not the sole factor. Other genetic architectures (e.g., structural variants, complex interactions) and environmental factors also play a role.

Summary updated

Abstract

These results highlight novel rare variants associated with cognition. GAB1 is an AGORA nominated gene target for potential AD treatment. Decreased expression was found in cholinergic neurons in AD patients and decreased learning and memory in a mouse model of AD. PLA2G4A has increased expression in AD patients that is evident in early stages but is decreased in healthy aging brains. Future work will incorporate other ancestries.
BibTeX
@inproceedings{regelson2025rare,
  author = {Alexandra N. Regelson and Derek B. Archer and Alaina Durant and Shubhabrata Mukherjee and Michael L. Lee and Seo‐Eun Choi and Phoebe Scollard and Emily H. Trittschuh and Jesse Mez and William S. Bush and Amanda Kuzma and Michael Cuccaro and Carlos Cruchaga and Lindsay AA. Farrer and Li‐San Wang and Gerald D. Schellenberg and Richard Mayeux and Walter W. Kukull and C. Dirk Keene and Andrew Joel Saykin and Sterling C Johnson and Corinne D. Engelman and David A. A. Bennett and Lisa Laverne Barnes and Eric B Larson and Kwangsik Nho and Alison M. Goate and Alan E. Renton and Edoardo Marcora and Brian Fulton‐Howard and Tulsi Patel and Shannon L Risacher and Anita L. DeStefano and Julie A Schneider and Mohamad Habes and Sudha Seshadri and Claudia L Satizabal and Pauline Maillard and Arthur W. Toga and Karen Crawford and Duygu Tosun and Jeffery M. M Vance and Elizabeth C. Mormino and Charles S. DeCarli and Thomas J. Montine and Gary W. Beecham and Sarah Biber and Philip L. De Jager and Vilas Menon and Annie Lee and Adam Brickman and Badri N. Vardarajan and Christiane Reitz and Jennifer J. Manly and Jason M Fletcher and Miguel Arce Rentería and Jason Song and Qiongshi Lu and Paul K Crane and Timothy J. Hohman and Logan Dumitrescu and The Alzheimer's Disease Neuroimaging Initiative (ADNI) and The Alzheimer's Disease Sequencing Project (ADSP)},
  title = {{Rare variant associations with late‐life cognitive performance}},
  booktitle = {Alzheimer},
  year = {2025},
  doi = {10.1002/alz70855\_105742},
}
Figure