APOE4 Accelerates Development of Dementia after Stroke: Is There a Role for Cerebrovascular Dysfunction?

After showing that the incidence of dementia is nearly fifty times higher one year after major stroke compared to age-matched and sex-matched general population of stroke-free patients enrolled in the Oxford Vascular Study (OxVasc)¹, the same group has now studied how apolipoprotein E4 (APOE4), the major genetic risk factor for sporadic late-onset Alzheimer’s disease (AD), impacts the risk of dementia associated with stroke using the same longitudinal population-based cohort². The OxVasc is the first and only population-based study providing information on the incidence, causes, and outcomes of acute vascular events such as strokes and heart attacks in a population of nearly 100,000 people living in Oxfordshire, UK. In the present study, Pendlebury and colleagues reported that APOE4 homozygosity (ε4/ε4), but not heterozygosity (ε3/ε4), is strongly related to both pre- and post-event dementia in patients with transient ischemic attack (TIA) and stroke². Associations between pre- and post-event dementia and APOE genotype were studied using logistic regression and Cox regression models after statistically controlling for age, sex, and education, cerebrovascular burden (white matter disease, stroke history and severity), dysphasia, and vascular risk factors such as diabetes.

The present findings are important, but some additional outcome measures of cerebrovascular burden that have not been presently considered, as discussed in the paper, factors could also play a role in the overall data interpretation and may influence or change conclusions. This includes, for example, subtype of stroke (ischemic vs. hemorrhagic), treatment of stroke (thrombolysis or thrombectomy), metabolic factors (e.g., hypercholesterolemia), hypertension, pre- and post-TIA and pre- and post-stroke cerebrovascular integrity changes, post-ischemic cerebral blood flow recovery, and/or regenerative response during stroke recovery or after TIA. Additionally, the n numbers of APOE-ε4/ε4 carriers having pre- and post-dementia evaluations were still relatively low, i.e., 5 and 9, respectively, as the authors also pointed out in their discussion. For all the reasons above, the present findings should be accepted with at least some caution notes.

By interpreting their data, the authors seem to favor the “non-vascular” hypothesis as an underlying mechanism that could potentially explain the association between APOE4 genotype and pre- and post-event dementia and neurodegenerative changes after TIA and stroke such as Alzheimer’s pathology. For instance, the discussion elaborates that both TIA and stroke may accelerate cognitive decline in APOE4 homozygous carriers by facilitating aberrant processing of Alzheimer’s amyloid-β (Aβ) protein or via Aβ-mediated post-stroke inflammation. Despite that the Aβ pathway could play a role, still no conclusive data have been shown or reported to support this hypothesis either way in patients with TIA or stroke. As briefly discussed, the “vascular” effect cannot be ruled out either. For example, some other important cerebrovascular burden measures such as cerebral amyloid angiopathy (often found in ε3/ε4 and ε4/ε4 carriers), silent strokes (frequently found on MRI analysis in these age groups), microbleeds (frequent in ε3/ε4 and ε4/ε4 carriers), enlarged perivascular spaces, and/or loss of cerebrovascular integrity and blood-brain barrier (BBB) breakdown could also play a role.

In fact, BBB dysfunction could explain at least some of the present findings because previous studies have implicated APOE4 as a prime culprit in damaging brain blood vessels. Indeed, APOE4 exerts strong cerebrovascular toxic effects leading to accelerated BBB breakdown and degeneration of brain capillary pericytes that maintain BBB integrity^3–7, as shown by human neuropathological^8–11 and animal¹² studies. The latter study in transgenic APOE4 mice found that a pro-inflammatory cyclophilin A (CypA)-matrix metalloproteinase-9 (MMP9) pathway leads to degradation of BBB tight junction and basement membrane proteins causing BBB breakdown¹², that has been confirmed on post-mortem brain tissue analysis in both ε3/ε4 and ε4/ε4 carriers diagnosed with AD⁹. Disintegration of the vessel wall of small capillary blood vessels leads in turn to leakages and accumulation of toxic blood-derived proteins in the brain parenchyma, such as to name a few, fibrinogen⁶, albumin¹³, thrombin and plasminogen⁵, leading to white matter injury, neuronal and synaptic dysfunction, and neurodegenerative changes, respectively. Notably, this occurs in the absence of Aβ, and could also contribute to post-ischemic cognitive decline especially as ischemic injury leads to large BBB breakdowns that would likely be potentiated by the APOE4 gene.

Remarkably, genetic and pharmacologic manipulations of the CypA-MMP9 pathway at the BBB were able to restore the BBB integrity, as well as to improve the neuronal and synaptic function in APOE4 transgenic mice¹², suggesting this vascular pathway could potentially be a therapeutic target in APOE4 carriers, who have increased risk for AD-type dementia or post-TIA and post-stroke dementia. Breakdown of the BBB has been demonstrated in older adults with cognitive impairment¹⁴, and has been shown to be an early biomarker of human cognitive dysfunction independently of Alzheimer’s Aβ and tau biomarkers¹⁵. However, how the effects of APOE4 on the cerebrovascular system contribute to cognitive impairment, how they relate to other APOE4 effects on neurons, and Aβ and tau disorders, remains presently unknown. It also remains elusive whether APOE4 cerebrovascular effects can interact with Alzheimer’s Aβ and tau pathways¹⁶ after TIA and stroke to accelerate development of dementia. These interesting questions should be addressed by future studies focused to find out mechanistic links and define pathways underlying pathophysiological changes leading to post-ischemic cognitive impairment.