Articles
The effect of APOE and other common genetic variants on the onset of Alzheimer's disease and dementia: a community-based cohort study

https://doi.org/10.1016/S1474-4422(18)30053-XGet rights and content

Summary

Background

Alzheimer's disease is one of the most heritable diseases in elderly people and the most common type of dementia. In addition to the major genetic determinant of Alzheimer's disease, the APOE gene, 23 genetic variants have been associated with the disease. We assessed the effects of these variants and APOE on cumulative risk and age at onset of Alzheimer's disease and all-cause dementia.

Methods

We studied incident dementia in cognitively healthy participants (aged >45 years) from the community-based Rotterdam Study, an ongoing prospective cohort study based in Rotterdam, the Netherlands, focusing on neurological, cardiovascular, endocrine, and ophthalmological disorders, and other diseases in elderly people. The Rotterdam Study comprises participants in three baseline cohorts (initiated in 1990, 2000, and 2006), who are re-invited to the research centre every 3–4 years, and continuously monitored by records from general practitioners and medical specialists. Cumulative incidence curves up to age 100 years were calculated for Alzheimer's disease and dementia, taking into account mortality as a competing event. These risk curves were stratified by APOE genotypes, tertiles of a weighted genetic risk score (GRS) of 23 Alzheimer's disease-associated genetic variants, and parental history of dementia.

Findings

12 255 of 14 926 participants (58·5% women) from the Rotterdam Study were included in this study. During a median follow-up of 11·0 years (IQR 4·9–15·9; 133 123 person years), 1609 participants developed dementia, of whom 1262 (78%) were classified as having Alzheimer's disease; 3310 people died of causes other than dementia. Both APOE and the GRS significantly modified the risks of Alzheimer's disease and dementia. There was evidence for a significant interaction between APOE genotypes and the GRS for the association with Alzheimer's disease (p=0·03) and dementia (p=0·04); the risk for carriers homozygous for APOE ε4 was modified most by the GRS. In carriers homozygous for APOE ε4, the difference between the high-risk tertile and the low-risk tertile by age 85 years was 27·0% for Alzheimer's disease (p=8·5 × 10−3) and 37·2% for dementia (p=2·2 × 10−4), which translates to a 7–10 year difference in age at onset. Comparing the risk extremes, which were carriers homozygous for APOE ε2 or heterozygous with one copy each of the ε2 and ε3 alleles in the low-risk tertile of the GRS versus carriers homozygous for APOE ε4 in the high-risk tertile of the GRS, the difference in risk by age 85 years was 58·6% (4·1% vs 62·7%; p=6·2 × 10−13) for Alzheimer's disease, and 70·3% (7·2% vs 77·5%; p=3·0 × 10−23) for dementia. These risk differences translate into an 18–29 years difference in age at onset for Alzheimer's disease and an 18–23 year difference in age at onset dementia. This difference becomes more pronounced when parental history of dementia is considered (difference in risk 83·8%; p=1·1 × 10−20).

Interpretation

Common variants with small individual effects jointly modify the risk and age at onset of Alzheimer's disease and dementia, particularly in APOE ε4 carriers. These findings highlight the potential of common variants in determining Alzheimer's disease risk.

Funding

None.

Introduction

Alzheimer's disease is the most common form of dementia. It is a multi-factorial disease with a substantial genetic component (60–80%)1 and the APOE gene is the strongest common genetic risk factor for Alzheimer's disease.2 The gene has three common alleles: the protective allele APOE ε2, the neutral or reference allele APOE ε3, and the risk allele APOE ε4.3 The effect of APOE ε4 is best described by the high absolute risk carriers have of developing Alzheimer's disease and dementia. According to findings from case-control studies,4, 5 carriers homozygous for APOE ε4 have an estimated risk of Alzheimer's disease of over 50% by age 85 years, compared with less than 10% for non-carriers by this age. Because of this high risk, there is increasing interest in including carriers homozygous for APOE ε4 in presymptomatic Alzheimer's disease treatment or prevention trials to reduce the necessary duration of these costly studies.6, 7 However, the clinical onset of Alzheimer's disease and dementia varies widely,8 ranging from midlife to the ninth decade even within carriers homozygous for APOE ε4.4

In addition to APOE, 23 other genetic variants have been identified in the past decade that significantly modify the risk of Alzheimer's disease.9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19 Combining the effects of these 23 variants results in a polygenic risk score that is not only associated with risk of Alzheimer's disease,20, 21, 22 but also with neuropathological hallmarks of Alzheimer's disease,23 conversion of mild cognitive impairment to Alzheimer's disease,24, 25, 26 and the age at onset of disease in both APOE ε4 carriers and non-carriers.23 Since estimates of absolute risks of Alzheimer's disease can be overestimated as a result of selection bias and not taking into account competing risks (particularly of death by other causes in this case),27 these findings await validation in large community-based cohort studies.4, 23

Research in context

Evidence before this study

We searched PubMed for articles published in English between Jan 1, 1990, and June 1, 2017, with combinations of the search terms “Alzheimer's disease”, “dementia”, “cumulative incidence”, “age at onset”, “genetic risk score”, “APOE”, “risk”, and “lifetime risk”. We found that in case-control studies, genetic risk scores of common genetic variants were significantly associated with Alzheimer's disease and age at onset beyond APOE genotypes. In these studies, genetic risk scores improve classification of cases and controls and might facilitate the identification of individuals at high risk and their inclusion in presymptomatic Alzheimer's disease treatment or prevention trials. However, to translate findings into practice, the combined effect of these variants needs to be validated in large community-based cohort studies that assess incident Alzheimer's disease and all-cause dementia, while controlling for competing risk of death by other causes than dementia.

Added value of this study

Our study shows that the small effect of common genetic variants together significantly modify the risk of Alzheimer's disease and all-cause dementia, and partly account for the variability in the age at onset, above and beyond the APOE genotype. Shifts in age at onset translated into a higher risk of Alzheimer's disease and dementia at all ages. The joint effect of common genetic variants on cumulative risk of Alzheimer's disease is most profound in APOE ε4 carriers. On the basis of all variants, we were able to stratify individuals into risk groups from very low (4·1% by 85 years) to extremely high (62·7%) cumulative risk of Alzheimer's disease.

Implications of all the available evidence

To our knowledge, this is the first time that the the potential of genetic variants beyond APOE in determining cumulative risk of Alzheimer's disease has been shown in a large community-based cohort. Including genetic variants beyond APOE could contribute towards better risk stratification in the community, facilitating planning of feasible and efficient preventive and curative clinical trials.

In this study of a large community-based cohort followed up for up to 25 years for incident dementia, we measured the aggregated effect of common variants separately and in conjunction with APOE on the risk and age at onset of Alzheimer's disease and all-cause dementia. Since our knowledge of Alzheimer's disease genetics is far from complete and parental history of dementia adds to the prediction of Alzheimer's disease on top of APOE and the genetic risk score (GRS),28 age at onset curves were constructed stratified by family history.

Section snippets

Study population

This study included participants without dementia from the Rotterdam Study, a prospective community-based cohort study.29 In 1990, residents aged 55 years and older residing in Ommoord, a district of Rotterdam, the Netherlands, were invited to participate in the study. Of 10 215 invited inhabitants, 7983 agreed to participate in the baseline examinations. In 2000, 3011 participants (of 4472 invitees) who had reached 55 years of age or moved into the study district since the start of the study

Results

12 255 of 14 926 participants from the Rotterdam Study were included in this study; 2671 participants were excluded, 743 due to insufficient screening at baseline, 527 due to prevalent dementia at study entry, and 1401 because they did not contribute follow-up time beyond age 60 years. Median follow-up was 11·0 years (IQR 4·9–15·9; 133 123 person-years). Follow-up for dementia was complete for 92% of 144 738 potential person-years. Table 1 shows the characteristics at study entry of the 12 255

Discussion

In this large community-based study, a GRS of common genetic variants modified the risk and onset of Alzheimer's disease and dementia above and beyond the effect of APOE, when taking into account the competing risk of mortality. The risk modification by the joint effect of common variants was most pronounced in carriers homozygous for APOE ε4, in whom there was a difference of up to 10 years in onset age between those in the low-risk and those in the high-risk tertile of the GRS. This shift in

References (53)

  • AM Saunders et al.

    Association of apolipoprotein E allele epsilon 4 with late-onset familial and sporadic Alzheimer's disease

    Neurology

    (1993)
  • LA Farrer et al.

    Effects of age, sex, and ethnicity on the association between apolipoprotein E genotype and Alzheimer disease. A meta-analysis. APOE and Alzheimer Disease Meta Analysis Consortium

    JAMA

    (1997)
  • E Genin et al.

    APOE and Alzheimer disease: a major gene with semi-dominant inheritance

    Mol Psychiatry

    (2011)
  • S Seshadri et al.

    Apolipoprotein E epsilon 4 allele and the lifetime risk of Alzheimer's disease. What physicians know, and what they should know

    Arch Neurol

    (1995)
  • KL Moulder et al.

    Dominantly Inherited Alzheimer Network: facilitating research and clinical trials

    Alzheimers Res Ther

    (2013)
  • EM Reiman et al.

    Alzheimer's Prevention Initiative: a plan to accelerate the evaluation of presymptomatic treatments

    J Alzheimers Dis

    (2011)
  • JC Lambert et al.

    Genome-wide association study identifies variants at CLU and CR1 associated with Alzheimer's disease

    Nat Genet

    (2009)
  • P Hollingworth et al.

    Common variants at ABCA7, MS4A6A/MS4A4E, EPHA1, CD33 and CD2AP are associated with Alzheimer's disease

    Nat Genet

    (2011)
  • JC Lambert et al.

    Meta-analysis of 74,046 individuals identifies 11 new susceptibility loci for Alzheimer's disease

    Nat Genet

    (2013)
  • R Guerreiro et al.

    TREM2 variants in Alzheimer's disease

    N Engl J Med

    (2013)
  • AC Naj et al.

    Common variants at MS4A4/MS4A6E, CD2AP, CD33 and EPHA1 are associated with late-onset Alzheimer's disease

    Nat Genet

    (2011)
  • RS Desikan et al.

    Polygenic overlap between C-reactive protein, plasma lipids, and Alzheimer disease

    Circulation

    (2015)
  • G Jun et al.

    A novel Alzheimer disease locus located near the gene encoding tau protein

    Mol Psychiatry

    (2016)
  • S Seshadri et al.

    Genome-wide analysis of genetic loci associated with Alzheimer disease

    JAMA

    (2010)
  • T Jonsson et al.

    Variant of TREM2 associated with the risk of Alzheimer's disease

    N Engl J Med

    (2013)
  • D Harold et al.

    Genome-wide association study identifies variants at CLU and PICALM associated with Alzheimer's disease

    Nat Genet

    (2009)
  • Cited by (171)

    View all citing articles on Scopus
    View full text