Alzheimer’s disease (AD) is the main neurodegenerative disease leading to dementia and cognitive impairment in the elderly worldwide. The classic pathophysiological hallmarks of AD are extracellular β-amyloid (Aβ) plaques and intracellular tau tangles, which eventually lead to the impairment of cognitive functions. These features progress slowly and are asymptomatic in the first stages of the disease. In fact, this fact hinders an acute premortem diagnosis if not aided by biological markers, as AD symptomatology may share similarities with other causes of dementia.

Besides cellular alterations, AD is also characterized by the presence of several vascular alterations, including small infarcts, or lacunes, due to the occlusion of branches of cerebral arteries, increases in the number of atrophic vessels and amount of vascular tortuosity (abnormal twists and turns in vessels), and decreases in microvascular density and length. Indeed, such brain vascular-associated alterations underlie many pathophysiological mechanisms of AD ^[1][2][1,2]. Accordingly, the two-hit vascular hypothesis points at the initial damage in cerebral vasculature (hit one) as the inducer of the accumulation of β-amyloid in the brain (hit two) [3]. Remarkably, cerebral blood–brain barrier (BBB) leakage and microbleeds are associated with cognitive decline in patients with mild cognitive impairment (MCI) and early AD, which opens the door to search for new biomarkers allowing for the diagnosis of AD before symptoms start. Importantly, many recent studies assessed the relationship between AD and stroke, another vascular-related neurological disease with a worldwide impact ^[4][5][4,5]. In this regard, a meta-analysis revealed that all stroke subtypes significantly increase the risk of developing AD [4]; and, more recently, that several differentially expressed genes and cellular pathways are shared by both stroke and AD [5]. Furthermore, theour group has highlighted that higher numbers of circulating endothelial progenitor cells (EPCs) within the first week following stroke have a positive impact on functional outcome ^[6][7][8][9][6,7,8,9], and several studies point to EPCs as a beneficial target for AD [1]. Altogether, EPCs may offer a new target to find a treatment for AD based on their recent molecular and genetic connections.

2. Vascular Alterations in AD Brains

The presence of vascular alterations entails an increased risk of developing dementia [10]. A meta-analysis of 2856 patients showed that a 60% prevalence of dementia was present in patients with macroinfarcts and lacunar disease; this percentage diminishes to 56% in small-vessel disease patients, and ranges from 57 to 70% in the presence of microinfarcts (depending on the number of lesions) [10]. More precisely, a post-mortem study found that 80% of AD patients had vascular pathology (e.g., presence of large infarcts, lacunes and multiple microinfarcts, hemorrhages, atherosclerosis, and arteriosclerosis), which has a higher prevalence than in other neurodegenerative diseases [11]. Furthermore, the presence of cerebrovascular disease, any condition affecting blood flow and blood vessel structure negatively, increases the risk of dementia in AD, with a more prominent effect in the earlier stages of the pathology [11]. Although microvascular alterations occur during normal ageing, they are especially prominent in neurodegenerative diseases, such as AD [12]. Some of these alterations include a decrease in the microvascular density, and the increase in both the number of atrophic vessels and string capillaries in AD patients ^[12][13][14][12,13,14]. In addition, fusiform dilatations, tortuosity, abnormal branching and fusions, as well as a reduction of the total length were detected in capillaries from AD samples ^[13][15][16][13,15,16]. The analysis of brain samples from AD patients has yielded a large amount of evidence indicating the presence of vascular alterations (Figure 1). For example, immunohistochemical analysis has detected the presence of potentially neurotoxic substances in the parenchyma due to their extravasation through the BBB, such as prothrombin, thrombin, fibrinogen, fibrin, albumin, and immunoglobulins ^{[15][17][18][19][20][21]}[15,17,18,19,20,21]. In the specific case of prothrombin, levels of this protein in the prefrontal cortex were higher as the Braak stage increased, pointing to a greater damage of the BBB throughout the disease progression [20]. Furthermore, there is an extravasation and accumulation of immune cells in AD brains that are mediated by the overexpression of molecules such as monocyte chemoattractant protein-1 (MCP-1), intercellular adhesion molecule 1 (ICAM-1), and integrins, promoting neuroinflammation as well as BBB damage ^{[17][22][23][24][25][26]}[17,22,23,24,25,26]. In addition, the release of many pro-inflammatory mediators, such as interleukin (IL)-1β (IL-1β), IL-6, IL-8, tumor necrosis factor α (TNF-α), TWEAK, and transforming growth factor β (TGF-β), into the vessels strengthen such marked neuroinflammation ^{[17][22][27][28]}[17,22,27,28]. Finally, accumulation of erythrocytes and hemosiderin deposits was also detected in AD brain samples, and this is consistent with the presence of microbleeds detected by magnetic resonance imaging (MRI) [29]. Pericytes are one of the most important cells in the brain as their role is keeping the structure of the BBB intact. In AD patients, there is a ~60% decrease in the number of pericytes in the capillaries of the cortex, and a 33% one in the hippocampus compared to controls [19]. Accordingly, decreased pericyte marker platelet-derived growth factor receptor-β (PDGFRβ) and marked degeneration of pericytes were observed ^[13][15][19][13,15,19]. Both signals would explain the increase in BBB leakage, since the presence of soluble PDGFRβ (sPDGFRβ) in cerebrospinal fluid (CSF) is a marker of BBB leakage [30]. Tight junctions (TJ) are fundamental elements for the correct maintenance of the BBB that connect endothelial cells with each other [31]. Three important proteins of TJ are claudin-5, occludin, and zonula occludens 1 (ZO-1), all of which are decreased in cortical areas of AD patients ^[32][33][34][32,33,34]. Reduced levels of claudin-5 and occludin were correlated with higher Braak states, Aβ₄₀ levels, and loss of synaptic markers, such as synaptophysin [34]. Interestingly, Aβ deposits downregulated the expression of occludin and ZO-1 [32]. Furthermore, the decrease in TJ levels could be also explained by the overexpression of cyclophilin A (CypA), metalloprotease-9 (MMP-9), and MMP-2, since this pathway promotes the TJ degradation ^[15][35][15,35]. Furthermore, increased levels of endothelin-1 (ET-1), a potent vasoconstrictor, are also associated with BBB leakage [36]. Alteration and damage of the BBB not only induce the extravasation of substances but also tissue hypoperfusion. Consistent with data from neuroimaging studies, biochemical evidence of cerebral hypoperfusion was found in AD brain tissue. For example, a decreased myelin-associated glycoprotein/proteolipid protein-1 ratio, which indicates hypoperfusion, was detected in AD brains ^[21][36][21,36]. Moreover, this decrease was associated positively with PDGFRβ levels, but negatively with ET-1 levels, Braak stages, Aβ, and plaque levels ^[21][36][21,36]. Furthermore, several molecular changes have also been observed in the cerebral endothelium of AD patients, many of which eventually lead to an increase in brain Aβ accumulation. For example, there is a lower expression of the lipoprotein receptor-related protein 1 (LRP-1) and a higher expression of the receptor for advanced glycation end products (RAGE) in the brain endothelium ^[32][37][38][32,37,38]. LRP-1 is a transporter that together with P-glycoprotein (P-gp) is involved in Aβ clearance, while RAGE is a receptor that uptakes Aβ from the circulation into the parenchyma ^[32][37][38][32,37,38]. Overall, there are robust data from AD brains pointing at a clear relationship between the vascular system and AD; however, most of these results come from post-mortem analysis, impeding their use for AD diagnosis but not for discovering new therapeutic targets; thus, it is necessary to study these molecules in the early stages of the disease and assess their potential to be used as biomarkers of AD beginning and/or evolution.  In the following sections, researchers will review and discuss the most recent and relevant works addressing the role of these molecules as potential AD biomarkers.