Identification of genetic modifiers of autosomal dominant Alzheimer's disease: a genome-wide association study

Abstract

Background: Individuals with autosomal dominant Alzheimer's disease (ADAD) arising from mutations in PSEN1, PSEN2, or APP exhibit variability in clinical presentation. Genetic studies of ADAD have shaped our understanding of the disease, and the discovery of genetic modifiers can inform therapeutic interventions and improve patient outcomes. We aimed to discover new genetic modifiers in individuals with mutations in the three ADAD genes. Methods: In this genome-wide association study, we analysed data from participants in three study cohorts (the Knight Alzheimer Disease Research Center [Knight-ADRC], the Dominantly Inherited Alzheimer Network [DIAN] observational study, and the Alzheimer Disease Sequencing Project [ADSP] R4). We did whole-genome sequencing on 101 unrelated, non-Hispanic, White, symptomatic participants with ADAD mutations and 5050 asymptomatic, unrelated control participants. Sensitivity analyses included related participants (148 cases and 5813 controls). We assessed the molecular mechanisms associated with each risk variant, including cis-regulatory effects, plasma protein levels (Knight-ADRC, 2338 participants), CSF concentrations of Alzheimer's disease biomarkers (DIAN, 64 participants), and neuroimaging data (MRI and PET; DIAN, 64 participants). We evaluated the association of risk variants with age at onset in ADAD and in 6177 participants with sporadic Alzheimer's disease (ADSP R5). Findings: Three genome-wide loci with significant risk were associated with ADAD risk, irrespective of the specific ADAD gene mutation. The CNIH4 locus association was driven by a missense variant (is caused by Gly54Ser, p<0·0001, odds ratio [OR] 11·99 [5·39-26·64]). The CCNG1 locus risk allele increased the risk of Alzheimer's disease (p<0·0001, OR 9·56 [4·29-21·24]) and reduced the age at dementia onset (p=0·0068, β=-10·15 [95% CI -17·31 to -2·77]). This allele was also positively associated with Tar DNA binding protein 43 (TDP-43) plasma protein levels and a larger gap between chronological age and structural MRI predicted brain age. The RHOJ risk allele (p<0·0001, OR 5·96 [3·42-10·36]) was associated with increased the risk of Alzheimer's disease, higher CSF total tau (p=0·0056, β=358·37) and phosphorated tau 181 (pTau181; p=0·0006, β=81·28), and lower Aβ42/Aβ40 ratio (p=0·016, β=-0·11) in DIAN ADAD participants, comparing those carrying the risk allele with those not carrying it. Interpretation: Our findings provide potential insights into disease biology, emphasising the role of Aβ, tau, TDP-43, astrocytes, and angiogenesis in Alzheimer's disease aetiology. This study offers invaluable insight for family genetic counselling and future clinical trial designs. Funding: National Institute of Health, National Institute on Aging, Alzheimer's Association, Hope Center Pilot 2025 Award, NGI Pilot Grant 2025 Award, BrightFocus Foundation, UK Dementia Research Institute at University College London, UK National Institutes for Health and Care Research University College London Hospitals Biomedical Research Centre, Dominantly Inherited Alzheimer Network, Freedom Together Foundation.