Revisiting APOE : Insights from rare variants and the expanding landscape of Alzheimer's disease genetics

The centrality of apolipoprotein E (APOE) to Alzheimer's disease (AD) genetics is unparalleled. Among common complex diseases, no other locus has demonstrated such consistent replication across independent, multi-ancestry studies.^1–5 Historically, the APOE ε4 allele has been regarded as the strongest genetic risk factor for late-onset AD, its influence robust across European, Asian, and African populations.^6,7 Yet recent work, including that by Miyashita et al., ⁸ suggests the story is more intricate.

The study by Miyashita et al. examines rare missense variants (RMVs) within APOE in a Japanese cohort, identifying novel associations that underscore the importance of rare variant contributions to AD risk. ⁸ This finding aligns with the growing shift from the “common variant, common disease” paradigm toward acknowledging “rare variant, common disease” contributions. The work further emphasizes that APOE's influence is not solely through ε4 but also through a spectrum of rare, potentially functional variants.

A particularly compelling feature of the APOE locus is the divergence in risk conferred by its common alleles: ε2 is protective, while ε4 increases AD risk. Although ε4's role has been debated, whether inherently toxic or contextually modifiable, a recent consensus by a working group convened by the Alzheimer's Disease Sequencing Project (ADSP) concluded that APOE ε4 is definitively toxic, especially in African and European ancestries. ⁹ This reinforces ε4's central pathogenic role and calls for the development of targeted interventions. At the same time, these insights prompt a closer examination of other APOE variants that may exert risk-modifying effects, especially rare, population-specific alleles that might act independently of or interactively with ε4.

Within this framework, the identification by Miyashita et al. ⁸ of a rare missense variant associated with protection against AD adds a new dimension to APOE biology. Notably, APOE ε7, defined by two tightly linked missense variants (p.E262 K and p.E263 K), resides in the C-terminal lipid-binding domain ¹⁰ and appears to confer a protective effect against AD. However, as acknowledged by the authors, this protective association loses statistical significance when adjusted for ε4 status, emphasizing the inherent challenges in disentangling the independent effects of rare variants in the context of strongly penetrant common alleles. Such limitations underscore the importance of cautious interpretation and replication across cohorts.

This discovery expands the scope of functional scrutiny beyond the canonical ε2/ε3/ε4 isoform-defining residues. It raises the intriguing possibility that multiple other residues throughout the ApoE protein, particularly within structurally and functionally critical regions, may also exert modifying effects on disease risk. This repositions APOE not just as a gene marked by a few influential variants, but as a protein whose broader amino acid landscape could harbor both pathogenic and protective potentials.

This broadened perspective aligns with recent evidence showing that individuals harboring null mutations in APOE do not exhibit overt neuropathology. Rather than presenting a paradox, such cases reinforce the hypothesis that partial or complete inactivation of APOE, particularly ε4, may confer neuroprotection. Together, these insights underscore the urgency of investigating rare, function-affecting APOE variants—not only to clarify their roles in disease etiology but also to guide the design of precision therapeutic strategies grounded in naturally occurring protective mechanisms.

The emergence of rare APOE variants with substantial effects in non-European populations underscores the necessity of global genetic studies. Given that APOE ε7 has been observed in multiple ethnicities, cross-ancestry meta-analyses may provide sufficient power to resolve its true effect on AD risk. Furthermore, the identification of several risk-modifying mutations in the lipid-binding domain highlights this region as a promising target for therapeutic innovation. Such advances invite a reconsideration of therapeutic strategies targeting APOE, demanding nuanced approaches that account for functional heterogeneity rather than treating ε4 as a monolithic risk factor.

Miyashita et al. ⁸ have contributed an important piece to this evolving puzzle. Future research must aim to functionally characterize these rare variants, map their expression profiles in critical brain regions, and integrate these insights into a unified model of APOE biology in AD. Only through such efforts can we hope to resolve the contradictions and fully understand the dual nature of APOE as both friend and foe.