Fortea J, Pegueroles J, Alcolea D, Belbin O, Dols-Icardo O, Vaqué-Alcázar L, Videla L, Gispert JD, Suárez-Calvet M, Johnson SC, Sperling R, Bejanin A, Lleó A, Montal V. APOE4 homozygozity represents a distinct genetic form of Alzheimer's disease. Nat Med. 2024 May;30(5):1284-1291. Epub 2024 May 6 PubMed. Correction.
Recommends
Please login to recommend the paper.
Comments
Rouen University Hospital
This is a very appealing study on neuropathological changes and in vivo biomarkers in APOE4 homozygotes. A better knowledge on the history of biomarkers in different genetic contexts is indeed required for the future of AD prevention, beyond autosomal dominant forms, and APOE is clearly the main gene to focus on first, along with rare variants in more recent genes (SORL1, TREM2, ABCA7, etc.).
One of the messages in this paper, and which is always a useful reminder, is that AD is the same disease in autosomal dominant and non-autosomal dominant forms, with or without APOE4: Whatever the cause (monogenic versus complex, with or without APOE4), the course of biological alterations and CSF biomarkers is similar, only the rhythm of progression along the disease path differs. This adds more data to further demonstrate it.
The main message is about the biological penetrance of AD. The data show quite convincingly that AD biological processes are identifiable at high levels from age 60 in 75 percent of E4 homozygotes and at high or intermediate levels in all E4 homozygotes by age 90. In that respect, the exact claim that APOE4 homozygotes present near-full penetrance of AD biology is justified and very interesting.
Of note, it is important to remember that APOE4 homozygotes do not present near-full penetrance of AD dementia. Combining cross-sectional, age-category-specific odds ratios with data from cohorts allowed a first estimation of AD clinical penetrance per APOE genotype back in 2011 (Génin et al., 2011) and the penetrance was high in the E4-homozygote group (50-60 percent) but still far from full (and APOE genotypes were already split and not gathered as a whole group of E4+). This result was then confirmed in a nice way in the Rotterdam study, directly based on cohort data (Van der Lee et al., 2018), which is a gold-standard study design for assessing penetrance.
The interpretation of the results is thus to be related to the actual definition of AD. It has been a hot topic (for more than a century!), but defining it exclusively with high/medium neuropathological changes, or using the ATN classification, makes a focus on biological/mechanistic changes only. Although this is meaningful from one point of view, it is not sufficient to predict dementia, which remains the main clinical question. Indeed, the discussion of this article suggests the use of APOE genotyping for genetic counseling with predicted age at onset, and it is reminded that roughly 2 percent of the population is homozygous. To us, this message is a bit dangerous. We know from previous large prospective cohorts that 50 percent of E4 homozygotes will not develop AD-dementia in their lifetimes. Admittedly, 50 percent will, and this number is indeed huge. In the Génin et al. paper on the “semidominant” inheritance of AD with APOE4 (meaning increased risk in heterozygotes and even more in homozygotes, in ranges not so far away from what we see in some Mendelian disorders), adding “with incomplete penetrance” in the title would have been more accurate. Not all E4-E4 individuals develop AD dementia, and that is very important to keep in mind when dealing with such a concept in the clinic, especially with asymptomatic individuals. In other words, penetrance of AD biology is not penetrance of AD dementia.
Another important point is the prediction of the age of onset. In families, the age of onset of the transmitting parent can be used as a reference or an add-on to account for familial background risk. In the case of APOE4 homozygotes, it is obviously not possible. We already know that additional factors, including common variants in other genes and more importantly rare coding variants in recently identified genes (e.g., Schramm et al., 2022), drastically change the picture at the individual level, and that is without accounting for non-genetic factors. This is much less the case in dominantly inherited AD, where APOE genotype has only a modest effect on age at onset, and other modifier variants are only found in extremely rare outlier cases. We would thus disagree with the authors’ claim that ages of onset are predictable in APOE4-homozygous individuals. In addition, it appears questionable to predict ages of onset when we know that 50 percent of the E4 homozygotes do not develop AD dementia. How can the age at symptom onset be projected to be roughly between 50 and 80 years old for 95 percent of individuals (Fig 1d) of when 50 percent will not develop dementia by the age of 85? This appears paradoxical with the clinical cohort data.
Overall, these data are extremely interesting, especially those suggesting that APOE4 homozygotes all develop AD-related changes, even though it cannot be used to predict ages of onset, to our point of view, and it is important to remind that this work does not report clinical penetrance. APOE is already considered a major AD gene, and it is uncertain that adding a category that is already in the mind of people would be required.
—Camille Le Clézio Charbonnier is the co-author of this comment.
References:
Genin E, Hannequin D, Wallon D, Sleegers K, Hiltunen M, Combarros O, Bullido MJ, Engelborghs S, De Deyn P, Berr C, Pasquier F, Dubois B, Tognoni G, Fiévet N, Brouwers N, Bettens K, Arosio B, Coto E, Del Zompo M, Mateo I, Epelbaum J, Frank-Garcia A, Helisalmi S, Porcellini E, Pilotto A, Forti P, Ferri R, Scarpini E, Siciliano G, Solfrizzi V, Sorbi S, Spalletta G, Valdivieso F, Vepsäläinen S, Alvarez V, Bosco P, Mancuso M, Panza F, Nacmias B, Bossù P, Hanon O, Piccardi P, Annoni G, Seripa D, Galimberti D, Licastro F, Soininen H, Dartigues JF, Kamboh MI, Van Broeckhoven C, Lambert JC, Amouyel P, Campion D. APOE and Alzheimer disease: a major gene with semi-dominant inheritance. Mol Psychiatry. 2011 Sep;16(9):903-7. Epub 2011 May 10 PubMed.
van der Lee SJ, Wolters FJ, Ikram MK, Hofman A, Ikram MA, Amin N, van Duijn CM. The effect of APOE and other common genetic variants on the onset of Alzheimer's disease and dementia: a community-based cohort study. Lancet Neurol. 2018 May;17(5):434-444. Epub 2018 Mar 16 PubMed.
Schramm C, Charbonnier C, Zaréa A, Lacour M, Wallon D, CNRMAJ collaborators, Boland A, Deleuze JF, Olaso R, ADES consortium, Alarcon F, Campion D, Nuel G, Nicolas G. Penetrance estimation of Alzheimer disease in SORL1 loss-of-function variant carriers using a family-based strategy and stratification by APOE genotypes. Genome Med. 2022 Jun 28;14(1):69. PubMed. Correction.
View all comments by Gael NicolasBanner Alzheimer's Institute
Arizona Alzheimer's Consortium
We commend Fortea and colleagues for their analyses examining biomarker and clinical changes in a large number of cognitively impaired and unimpaired persons from NACC, Alfa plus, A4 trial, OASIS, ADNI, and WRAP. In general, the results from their retrospective case control study are consistent with those from other retrospective case control studies, showing much higher odds of Alzheimer’s pathophysiological changes and clinical decline, and younger ages at onset in those with two APOE4 alleles than those with other APOE genotypes.
Despite the value and consistency of results, we have concerns about the conclusions that 1) APOE4 homozygotes have a nearly certain likelihood of AD biology and symptom onset, 2) APOE4 homozygosity is categorically rather than quantitatively different from the other APOE genotypes, and 3) homozygotes should be evaluated separately from other groups in treatment and prevention trials.
We’d like to note the following:
Despite the high risk of developing AD in APOE4 homozygotes, we hope that this population shares our optimism about the chance to find effective disease-modifying and prevention therapies in the next few years, as well as the chance to develop and test investigational APOE-modifying treatments.
References:
Corder EH, Saunders AM, Strittmatter WJ, Schmechel DE, Gaskell PC, Small GW, Roses AD, Haines JL, Pericak-Vance MA. Gene dose of apolipoprotein E type 4 allele and the risk of Alzheimer's disease in late onset families. Science. 1993 Aug 13;261(5123):921-3. PubMed.
Reiman EM, Arboleda-Velasquez JF, Quiroz YT, Huentelman MJ, Beach TG, Caselli RJ, Chen Y, Su Y, Myers AJ, Hardy J, Paul Vonsattel J, Younkin SG, Bennett DA, De Jager PL, Larson EB, Crane PK, Keene CD, Kamboh MI, Kofler JK, Duque L, Gilbert JR, Gwirtsman HE, Buxbaum JD, Dickson DW, Frosch MP, Ghetti BF, Lunetta KL, Wang LS, Hyman BT, Kukull WA, Foroud T, Haines JL, Mayeux RP, Pericak-Vance MA, Schneider JA, Trojanowski JQ, Farrer LA, Schellenberg GD, Beecham GW, Montine TJ, Jun GR, Alzheimer’s Disease Genetics Consortium. Exceptionally low likelihood of Alzheimer's dementia in APOE2 homozygotes from a 5,000-person neuropathological study. Nat Commun. 2020 Feb 3;11(1):667. PubMed.
Qian J, Wolters FJ, Beiser A, Haan M, Ikram MA, Karlawish J, Langbaum JB, Neuhaus JM, Reiman EM, Roberts JS, Seshadri S, Tariot PN, Woods BM, Betensky RA, Blacker D. APOE-related risk of mild cognitive impairment and dementia for prevention trials: An analysis of four cohorts. PLoS Med. 2017 Mar;14(3):e1002254. Epub 2017 Mar 21 PubMed.
Ghisays V, Goradia DD, Protas H, Bauer RJ 3rd, Devadas V, Tariot PN, Lowe VJ, Knopman DS, Petersen RC, Jack CR Jr, Caselli RJ, Su Y, Chen K, Reiman EM. Brain imaging measurements of fibrillar amyloid-β burden, paired helical filament tau burden, and atrophy in cognitively unimpaired persons with two, one, and no copies of the APOE ε4 allele. Alzheimers Dement. 2020 Apr;16(4):598-609. Epub 2020 Jan 16 PubMed.
View all comments by Eric M. ReimanWashington University
The authors of this manuscript analyzed, together, data from a number of previously published studies and comment that: “Findings revealed that almost all APOE4 homozygotes exhibited AD pathology and had significantly higher levels of AD biomarkers from age 55 compared to APOE3 homozygotes. By age 65, nearly all had abnormal amyloid levels in cerebrospinal fluid, and 75% had positive amyloid scans, with the prevalence of these markers increasing with age, indicating near-full penetrance of AD biology in APOE4 homozygotes. The age of symptom onset was earlier in APOE4 homozygotes at 65.1, with a narrower 95% prediction interval than APOE3 homozygotes.”
Based on this re-analysis of published data, the authors concluded that APOE4 homozygosity is a distinct genetic form of AD. Their analysis is certainly correct from these previously published cohorts, and I agree that there is no question that APOE genotype is the strongest and most powerful genetic risk factor for late-onset AD. Further, there are some distinct features of AD pathology in the presence of APOE4 that is allele dose dependent, such as increase in CAA in addition to parenchymal amyloid deposition.
However, there is one major difference between the APOE4 heterozygous and homozygous state that differs from AD deterministic mutations in APP, PSEN1, PSEN2, and trisomy 21/trisomy for APP. The risk for AD biomarkers and dementia appears to be significantly different across race and ethnicity for APOE4 homozygotes, being in order of highest to lowest in East Asians, non-Hispanic whites, non-Hispanic blacks, and Hispanics (Belloy et al., 2023). Other work suggests similar findings. For example, see the figure below from the National Institute on Aging/Alzheimer’s Disease Sequencing Project Working Group on APOE (Vance et al., 2024).
As Belloy et al. pointed out, and certainly more work needs to be done on how different race and ethnic factors affect risk of AD conferred by APOE4 and other APOE genotypes, similar large changes in risk due to other differences across race and ethnicity have not been described for APP, PSEN1, PSEN2, or Down’s syndrome. Certainly, there are recent examples where a very rare variant in APOE itself (APOE3ch homozygosity) appears to delay age of onset of dementia in the setting of a specific PSEN1 mutation. However, the differential risk across race and ethnicity suggests that there must be factors, genetic and other, that can strongly affect risk for different aspects of AD due to APOE4. Given that, I’m not sure at this point it is appropriate to think of APOE homozygosity in the same way as autosomal-dominant AD mutations or Down’s syndrome. There are many reasons why both the genetics and biology of APOE and APOE4 homozygosity need to be studied even more than they are now, and to think about assessing APOE4 homozygotes in an individualized way for all clinical trials and treatments, and, at a minimum, to account for this group in relation to other groups in analyses.
References:
Belloy ME, Andrews SJ, Le Guen Y, Cuccaro M, Farrer LA, Napolioni V, Greicius MD. APOE Genotype and Alzheimer Disease Risk Across Age, Sex, and Population Ancestry. JAMA Neurol. 2023 Dec 1;80(12):1284-1294. PubMed.
Vance JM, Farrer LA, Huang Y, Cruchaga C, Hyman BT, Pericak-Vance MA, Goate AM, Greicius MD, Griswold AJ, Haines JL, Tcw J, Schellenberg GD, Tsai LH, Herz J, Holtzman DM. Report of the APOE4 National Institute on Aging/Alzheimer Disease Sequencing Project Consortium Working Group: Reducing APOE4 in Carriers is a Therapeutic Goal for Alzheimer's Disease. Ann Neurol. 2024 Apr;95(4):625-634. Epub 2024 Jan 5 PubMed.
View all comments by David HoltzmanStanford University
I think this paper is helpful in putting some biomarker and neuropathology heft behind the growing consensus in the field that E4 homozygosity, in European ancestry individuals at least, is all but deterministic for the development of Alzheimer’s disease by age 90.
One important argument against their reconceptualization, which they mention briefly in their discussion, is that the risk of Alzheimer’s disease in E4 homozygotes varies substantially across different genetic ancestries. In our recent publication, we replicated and expanded upon an earlier, seminal meta-analysis from Lindsay Farrer and colleagues (Belloy et al., 2023). European-ancestry individuals with two copies of APOE4 have a 13-fold increased risk for Alzheimer’s, whereas in African-ancestry individuals with two copies of APOE4, their risk is 6.5-fold, or half of that seen in Europeans. This has critical implications when counseling patients, in that we really need to provide ancestry-informed genetic risk assessments for Alzheimer’s disease. It also speaks to some yet-to-be-discovered genetics and biology that presumably drive this massive difference in risk. It is unlikely that their reconceptualization would hold up in an African-ancestry population.
References:
Belloy ME, Andrews SJ, Le Guen Y, Cuccaro M, Farrer LA, Napolioni V, Greicius MD. APOE Genotype and Alzheimer Disease Risk Across Age, Sex, and Population Ancestry. JAMA Neurol. 2023 Dec 1;80(12):1284-1294. PubMed.
View all comments by Michael GreiciusUniversity of Copenhagen
Perspective on the findings and their implications.
Age-related Alzheimer’s disease (AD) is a neurodegenerative disorder with a complex pattern of genetic and environmental factors contributing to its pathogenesis. One gene stands, however, central for risk of AD, namely the apolipoprotein E gene (APOE) where two common genetic variants combine into six common APOE genotypes, APOE2/2, 3/2, 4/2, 3/3, 4/3, and 4/4. The APOE4 allele is a major risk factor for AD, with risk estimates for 4/4 homozygotes around 8-10 in prospective studies of the general population and up to approximately 15-fold in case control studies. Insights into the APOE4 risk were already reported in the ’90s and have subsequently been shown in many populations around the world. In European populations, around 25 percent are APOE3/4 carriers and 2-3 percent are 4/4 carriers with a decreasing north to south APOE4 allele frequency gradient. Differences in frequency and effect size according to ethnicity are also observed.
In this paper, scientists led by Juan Fortea and Victor Montal suggest that homozygosity for APOE ε4 constitutes a genetically determinant form of AD similar to autosomal dominant forms of AD, and that their results will have compelling consequences for public health and genetic counseling.
Limitations to this claim are, however, important to emphasize. The current article builds on a case/control design, with cases identified through neuropathological evaluation or preclinical or established AD, recruited through AD-families, and with controls being cognitively intact. An inherent and well-established bias in case control studies is so-called referral bias, where APOE 4/4 carriers who display altered AD biomarkers and/or cognitive symptoms, are referred to studies more often than APOE4/4 carriers who do not have such changes. This naturally leads to a circular argument, with inflation of risk estimates and penetrance, which may at least partly explain the high penetrance of APOE 4 homozygosity on AD biomarkers observed in the present case control study. To determine the penetrance in an unbiased way for a common genotype as the APOE4, prospective studies of the general population, with long follow-up, are needed. Such a design minimizes the risk of referral bias, because individuals are picked at random from the background population. It is also well-established that APOE genotypes are involved in many complex biological interactions, both in the brain and in peripheral lipid metabolism, and that the effect of APOE4 can be substantially modulated by other genetic risk markers spread across the genome, as well as by traditional cardiovascular risk factors.
The risk associated with APOE4 homozygosity is important and should be taken seriously in clinical trial designs and in public health initiatives aiming at preserving brain health with early preventable measures. To categorize APOE4 homozygosity together with autosomal dominant forms of AD, as suggested by Fortea and Montal, is however an overly strong claim considering the current case control findings of a common genetic risk variant.
View all comments by Ruth Frikke-SchmidtUniversity of Melbourne
Is late-onset, “sporadic” Alzheimer’s disease genetic?
Is this a trick question? Ever since Strittmatter and Roses published the first evidence of the association between ApoE and Aβ-amyloid in 1993 (Strittmatter et al., 1993), and as long as he lived, Alan Roses (1943-2016) promoted the idea that “sporadic” AD was largely genetic, driven by the ε4 allele in a dose-dependent fashion: one copy about 50 percent penetrance, two copies nearly 99 percent penetrance. In today’s world of amyloid PET and biofluid biomarkers, we have a clearer understanding that the ApoE4 allele drives the onset of Aβ accumulation. Our most recent data shows that carriage of one or two alleles of APOE4 reduces the age of onset (judged by Aβ-PET) by 7.5 years (Burnham et al., 2024), but the rates of accumulation in APOE4-positive and APOE4-negative subjects are the same. The current formulation of how the APOE4 allele works points to ApoE-Aβ interactions in the microglial clearance pathways, mediated by a gain of function through the Cys-Arg changes at codons 112 and 158.
This paper by Fortea et al. serves to emphasize the ApoE4 gene-dosage effect: two copies are almost completely dominant in determining age at onset, but there are escapees. They compare E4 homozygotes with autosomal-dominant Alzheimer’s (ADAD) and Down’s syndrome AD (DSAD). But there are some subtle differences that we should remember. ADAD has a more aggressive presentation and regional accumulation of Aβ is more marked in the striatum, cerebellum (and in some cases in the cortico-spinal tracts, resulting in spastic paraparesis), which are not features of late-onset AD. DSAD is much more like late-onset AD, but with an earlier mean clinical onset at 55 years.
Nevertheless, APOE4 homozygotes should be categorized as a defined subset of late-onset “sporadic” AD, simply because they can be easily identified well before onset of Aβ accumulation, and should be dealt with as a separate group in primary and secondary prevention trials with Ab-targeting immunotherapies.
References:
Burnham SC, Cox T, Benzinger T, et al. When Does Alzheimer's Disease Start? Robust Estimates Based on Longitudinal Aβ-Amyloid-PET in Three Large International Cohorts. The Lancet Preprint, March 18, 2024 The Lancet Preprint
Strittmatter WJ, Saunders AM, Schmechel D, Pericak-Vance M, Enghild J, Salvesen GS, Roses AD. Apolipoprotein E: high-avidity binding to beta-amyloid and increased frequency of type 4 allele in late-onset familial Alzheimer disease. Proc Natl Acad Sci U S A. 1993 Mar 1;90(5):1977-81. PubMed.
View all comments by Colin MastersInstitute Pasteur de Lille
The APOE4 allele constitutes a unique case among common genetic variants of chronic diseases, exerting a major influence on Alzheimer’s disease (AD) risk. AD also associates with three other distinct genetic forms, featuring rare variants of high effect, such as APP, PS1, PS2, and SORL1, as well as low-frequency variants with intermediate effects such as in TREM2, but the majority of cases involve common variants with minor effects, such as CLU or BIN1. Génin et al. in 2011 characterized APOE as a "major gene with semi-dominant inheritance," with lifetime risks akin to BRCA1 in breast cancer. This research by Fortea et al. further solidifies this assertion by revealing full pathological and biological penetrance in APOE4 homozygotes, and showcasing earlier clinical and biomarker changes. These findings advocate for reclassifying the APOE gene from a mere "risk factor" to a "major codominant gene," particularly among populations of European ancestry where 2 percent are homozygous for the APOE4 allele.
Should we regard APOE4 homozygosity as a distinct genetic form of AD? Quite possibly. Recent analyses of anti-amyloid drugs, such as lecanemab, suggest contraindications in APOE4 homozygotes due to a 39 percent incidence of amyloid-related imaging abnormality microhemorrhages (ARIA-H), compared to 13 percent in other patients (Van Dyck et al., 2022). Conversely, some studies have identified genetic resilience to AD among APOE4 homozygotes, with some individuals not developing AD beyond the age of 75. Notably, CASP7 rs10553596 and SERPINA3 rs4934 A/A have emerged as two genetic variants offering significant protection against AD in APOE4 homozygotes (Huq et al., 2019). Therefore, this high-risk population warrants special attention, both due to the unclear pathophysiological processes underlying the effect of this major AD gene, and to the accessibility of APOE genotype testing through direct-to-consumer genomic tests. This may have adverse impacts on individuals' lives, as in the case of Chris Hemsworth, the famous interpreter of Thor in the Marvel films, who learned a few years ago that he carries two copies of the E4 allele.
This extensively documented paper by Fortea and colleagues underscores the heterogeneous nature of the etiology, akin to other chronic diseases, such as hypertension, necessitating various treatment approaches combining prevention measures and diverse treatments, many of which are yet undiscovered in AD. Hence, increased investment in basic research is imperative to fully comprehend how the APOE4 allele, identified over 30 years ago, precisely influences AD pathophysiology. Once elucidated, powerful new drugs, and a straightforward genetic screening test, could mitigate the lifetime risk of AD for APOE homozygotes. However, until such treatments materialize, ignorance of one's APOE genotype may be preferable.
References:
Genin E, Hannequin D, Wallon D, Sleegers K, Hiltunen M, Combarros O, Bullido MJ, Engelborghs S, De Deyn P, Berr C, Pasquier F, Dubois B, Tognoni G, Fiévet N, Brouwers N, Bettens K, Arosio B, Coto E, Del Zompo M, Mateo I, Epelbaum J, Frank-Garcia A, Helisalmi S, Porcellini E, Pilotto A, Forti P, Ferri R, Scarpini E, Siciliano G, Solfrizzi V, Sorbi S, Spalletta G, Valdivieso F, Vepsäläinen S, Alvarez V, Bosco P, Mancuso M, Panza F, Nacmias B, Bossù P, Hanon O, Piccardi P, Annoni G, Seripa D, Galimberti D, Licastro F, Soininen H, Dartigues JF, Kamboh MI, Van Broeckhoven C, Lambert JC, Amouyel P, Campion D. APOE and Alzheimer disease: a major gene with semi-dominant inheritance. Mol Psychiatry. 2011 Sep;16(9):903-7. Epub 2011 May 10 PubMed.
Fortea J, Pegueroles J, Alcolea D, Belbin O, Dols-Icardo O, Vaqué-Alcázar L, Videla L, Gispert JD, Suárez-Calvet M, Johnson SC, Sperling R, Bejanin A, Lleó A, Montal V. APOE4 homozygozity represents a distinct genetic form of Alzheimer's disease. Nat Med. 2024 May;30(5):1284-1291. Epub 2024 May 6 PubMed. Correction.
van Dyck CH, Swanson CJ, Aisen P, Bateman RJ, Chen C, Gee M, Kanekiyo M, Li D, Reyderman L, Cohen S, Froelich L, Katayama S, Sabbagh M, Vellas B, Watson D, Dhadda S, Irizarry M, Kramer LD, Iwatsubo T. Lecanemab in Early Alzheimer's Disease. N Engl J Med. 2023 Jan 5;388(1):9-21. Epub 2022 Nov 29 PubMed.
Huq AJ, Fransquet P, Laws SM, Ryan J, Sebra R, Masters CL, Winship IM, James PA, Lacaze P. Genetic resilience to Alzheimer's disease in APOE ε4 homozygotes: A systematic review. Alzheimers Dement. 2019 Dec;15(12):1612-1623. Epub 2019 Sep 7 PubMed.
View all comments by Philippe AmouyelMake a Comment
To make a comment you must login or register.