What this means
BACKGROUND AND APOE ɛ4 is the strongest genetic risk factor for late-onset Alzheimer disease, while ɛ2 is associated with decreased risk. However, it is unclear how these genotypes relate to cognitive decline among those aged older than 90+.
A plain-English summary is on its way. Until then, this is the start of the authors' own summary.
- Who it was tested in
- People
- Kind of research
- A study
The science
- Original title
- Role of APOE ɛ4 and ɛ2 With Longitudinal Cognitive Decline Among Those With Exceptional Longevity: The LifeAfter90 Study.
- Authors
- Colbeth HL, Posis AIB, Rizvi BM et al.
- Journal
- Neurology
- Published
- Oct 16, 2026
- Study type
- Study · in humans
- PMID
- 42886871 ↗
Abstract
Background
APOE ɛ4 is the strongest genetic risk factor for late-onset Alzheimer disease, while ɛ2 is associated with decreased risk. However, it is unclear how these genotypes relate to cognitive decline among those aged older than 90+. We examined associations of ɛ4 and ɛ2 with longitudinal change in 3 cognitive domains among individuals aged 90+.
What they did
LifeAfter90 is an ongoing longitudinal multiethnic cohort study of those aged 90+. APOE genotyping was conducted using DNA extracted from saliva. Two dichotomous exposures were defined as ɛ4 carriers (ɛ2/ɛ4, ɛ3/ɛ4, or ɛ4/ɛ4 genotypes) vs non-ɛ4-carriers with no ɛ4 allele and ɛ2 carriers (ɛ2/2, ɛ2/3 genotypes) vs non-ɛ2-carriers (ɛ3/ɛ3, ɛ2/ɛ4, ɛ3/ɛ4, or ɛ4/ɛ4). Executive function (EF), semantic memory (SM), and verbal episodic memory (VEM) were measured every 6 months on up to 11 occasions using the Spanish and English Neuropsychological Scales. Time to all-cause mortality was calculated from baseline to death. We fit linear mixed-effects (LME) models and joint models (LME and Cox proportional hazard submodels) to account for the competing risk of death. Models adjusted for age and sex.
What they found
We evaluated 445 participants (mean age at baseline 93 years, SD 2.31, range 90-104; 59% female). In total, 20% were ɛ4 carriers, 15% were ɛ2 carriers, and 54% died during follow-up. The ethnoracial composition was 21% African American/Black, 22% as Asian, 16% as Hispanic/Latinx, 9% as multiracial/other, and 32% as White. The mean follow-up time was 3.9 years (SD 1.63, range 0.68-6.67). Neither ɛ4 nor ɛ2 carriers averaged substantially greater or lesser mortality hazard than noncarriers. In joint models, ɛ4 carriers had faster decline in EF (β = -0.07, 95% CI -0.12 to -0.02), but similar decline in VEM (β = -0.01, 95% CI -0.08 to 0.05) and SM (β = -0.03; 95% CI -0.08 to 0.02). APOE ɛ2 carriers had similar decline in EF (β = 0.01, 95% CI -0.05 to 0.08) and SM (β = 0.01, 95% CI -0.05 to 0.08) but slower decline in VEM (β = 0.07, 95% CI 0.00-0.15) compared with noncarriers. Longitudinal results were similar for LME and joint models. After accounting for the competing risk of death, the negative consequences of ɛ4 on EF decline extend to those with exceptional longevity.
The authors' abstract, grouped under plain headings from its own section labels.
Colbeth HL, Posis AIB, Rizvi BM et al. Role of APOE ɛ4 and ɛ2 With Longitudinal Cognitive Decline Among Those With Exceptional Longevity: The LifeAfter90 Study. Neurology. 2026. doi:10.1212/WNL.0000000000218596. PMID: 42886871. https://pubmed.ncbi.nlm.nih.gov/42886871/