Skip to main content
Alzheimer's & Dementia logoLink to Alzheimer's & Dementia
. 2025 Dec 24;21(Suppl 1):e105970. doi: 10.1002/alz70855_105970

Peruvian Population Empowers Discovery of a TREM2 Missense Variant in Alzheimer Disease

Bilcag Akgun 1,✉, Mario Cornejo‐Olivas 2,3, Pedro R Mena 1, Larry D Adams 1, Katrina Celis 1, Patrice G Whitehead 4, Kara L Hamilton‐Nelson 1, Julia Rios‐Pinto 5, Angel Medina‐Colque 6, Clifton L Dalgard 7, Eden R Martin 1,8, Ivan Cornejo‐Herrera 9, Maryenela Illanes‐Manrique 2,3,10, Edward Ochoa‐Valle 11, Elison Sarapura‐Castro 2,3, Koni Mejia ‐ Rojas 12,13, Victoria Marca‐Ysabel 3, Sheila Castro‐Suarez 10,14, Anthony J Griswold 1,8, Rosario Isasi 1,8, Katalina F McInerney 15, Michael L Cuccaro 1,8, Jeffery M Vance 1,8,15, Farid Rajabli 1,8, Margaret Pericak‐Vance 1,8,15; Alzheimer's Disease Sequencing Project (ADSP)
PMCID: PMC12738110

Abstract

Background

Increasing ethnic/ancestral diversity in genetic studies is critical for defining the genetic architecture of Alzheimer disease (AD) and maximizing the opportunities for identifying genetic protective and risk alleles. The Peruvian population, with up to ∼80% of Amerindian ancestry, provides a unique opportunity to leverage Amerindian ancestry for deeper insights into AD mechanism. We performed the genome‐wide association study (GWAS) in the Peruvian population to characterize known AD genetic risk loci.

Method

567 individuals (215 AD; 352 cognitively unimpaired) were included in these analyses. We performed GWAS on the Whole Genome Sequencing (WGS) dataset using a generalized linear mixed model, adjusting for sex, age, and population substructure as fixed effects and the genetic relationship matrix as a random effect. For follow‐up analysis, we extracted all nominal significant known AD variants and calculated linkage disequilibrium (LD) with surrounding SNPs. Variants with an R2 ≥ 0.8 with the index SNP were annotated and further evaluated. To assess whether the associations were ancestry‐specific, we calculated the local ancestry of the regions corresponding to these variants.

Result

We replicated two known AD loci APOE4 allele and TREML2 marker (rs60755019). Fine‐mapping analysis at the TREML2 loci identified a TREM2 missense exonic marker (p.H157Y, OR=3.3 [1.9‐5.9], p = 4x10‐5, CADD=23.2) having strong LD with the TREML2 risk marker in Peruvians. LA analysis showed that p.H157Y is located on an Amerindian ancestral background. p.H157Y is extremely rare across frequency databases (gnomAD, HGDP, and 1000 Genomes) except in Admixed American population ∼0.02. Additionally, APOE showed a strong association with AD, with an effect size (OR=4.3 [3.0‐6.1], p = 4x10‐15) higher that was observed in non‐Hispanic White populations.

Conclusion

Peruvian GWAS identified Amerindian ancestry specific exonic AD susceptibility variant p.H157Y. This variant has been previously reported as AD marker in Han Chinese populations with extreme effect size, but its rarity and contradictory findings in other groups suggest a potential ancestry‐specific effect. Functional studies indicate that it may enhance TREM2 shedding, impacting microglial function and contributing to AD pathogenesis. These observations highlight the power of examining diverse populations to gain a more comprehensive view of the genetic architecture of AD.


graphic file with name ALZ-21-e105970-g001.jpg


Articles from Alzheimer's & Dementia are provided here courtesy of Wiley

RESOURCES