Glycation in Alzheimer’s Disease and Type 2 Diabetes: The Prospect of Dual Drug Approaches for Therapeutic Interventions

Positive Lesions

Positive lesions in AD are defined by the presence of Aβ deposits and NFTs, both of which play pivotal roles in the disease’s progression. Aβ peptides, generated through the cleavage of amyloid precursor protein (APP) by β-secretase and γ-secretase, aggregate to form plaques. Among these peptides, Aβ42 is particularly neurotoxic due to its higher propensity for fibrilization, and decreased solubility compared to Aβ40, which is more abundant but less harmful [9, 10]. The accumulation of these plaques leads to neuronal damage and cognitive decline.

NFTs, composed of hyperphosphorylated tau, also contribute to AD pathology. The hyperphosphorylation of tau disrupts its interaction with microtubules, destabilizing the cytoskeleton and impairing synaptic communication. This disruption, driven by kinase dysregulation, oxidative stress, and inflammation, correlates closely with dementia severity [11, 12]. Together, Aβ plaques and NFTs represent critical therapeutic targets for AD.

Negative Lesions

Neuronal and synaptic loss occurs early in AD, before the appearance of amyloid plaques and NFTs. This loss exceeds the rate of normal aging and contributes significantly to brain atrophy and cognitive impairment. Synaptic degradation and neuronal death are linked to impaired neurogenesis, especially in regions like the hippocampus, which exacerbates cognitive deficits [13, 14].

Further, the loss of allopregnanolone, a neurosteroid, has been shown to worsen frontal cortex dysfunction, contributing to impaired cognition [15]. The increased apoptosis rate in AD distinguishes it from the natural aging process, highlighting the urgency of early intervention to slow neuronal and synaptic loss [16]. Identifying these early negative lesions can help in developing preventative strategies for AD-related neurodegeneration.

Neuroinflammation

The accumulation of Aβ plaques and NFTs in AD initiates a cascade of neuroinflammatory responses. Microglia and astrocytes become activated, releasing pro-inflammatory cytokines that, in turn, contribute to neuronal damage. While some of these cytokines may have protective effects under normal circumstances, their excessive production exacerbates neurodegeneration. Vital inflammatory pathways, including nitric oxide, cyclooxygenase (COX), and caspase signaling, amplify cytokine release, further promoting the inflammatory cycle [17, 18].

In addition to the direct effects of glial activation, dysbiosis in the gut microbiome has been linked to neuroinflammation in AD. Imbalances in gut microbiota can produce harmful molecules like lipopolysaccharides (LPS) and amyloids, which penetrate the blood–brain barrier, activating the NFκB signaling pathway and accelerating neuroinflammation [19]. Furthermore, chronic hyperglycemia in T2DM promotes excessive formation of advanced glycation end-products (AGEs), which accumulate in the brain and bind to the receptor for AGEs (RAGE) on neurons, microglia, and endothelium [20]. RAGE engagement potently activates NF-κB and NADPH oxidase signaling, greatly amplifying oxidative stress (via ROS) and the release of pro-inflammatory cytokines (e.g., TNF-α, IL-6) [21]. This injurious oxidative–inflammatory milieu impairs synaptic function and drives neuronal apoptosis [22]. In parallel, hyperglycemia and RAGE-dependent signaling activate PKC/NF-κB cascades that disrupt tight junctions and blood–brain barrier integrity [23] permitting influx of peripheral immune factors and Aβ into the CNS. Together, these interlinked mechanisms directly link diabetic hyperglycemia to exacerbated AD neuroinflammation and accelerated neuron loss. This complex interplay between systemic inflammation, gut microbiota dysbiosis, and neurodegeneration underpins the multifaceted nature of AD and offers promising avenues for therapeutic intervention. Although targeting AGEs and RAGE shows some promise, it is unlikely to be sufficient alone to significantly alter Alzheimer’s disease AD progression. Recent studies support multi-targeted approaches, combining anti-glycation strategies with therapies against other pathological mechanisms, to achieve better cognitive outcomes and broader disease modification. While these strategies introduce greater complexity, they align with the multifactorial nature of AD, and current research increasingly favors integrated interventions over single-target therapies.

Glycation and Amyloid-Beta Aggregation

The interaction between glycation and Aβ is critical in understanding the progression of AD. Aβ oligomerization, a key AD feature, is accelerated by glycation, leading to increased amyloid plaque formation. High glucose concentrations have been shown to enhance the formation of AGEs, which in turn influence Aβ aggregation and stability. Glycated Aβ is more prone to forming toxic oligomers and amyloid fibrils, exacerbating neuronal toxicity [41]. These modified Aβ species resist proteasomal degradation, contributing to plaque accumulation [42].

Interestingly, glycation at specific lysine residues of Aβ destabilizes the fibrillar structure, promoting the formation of toxic soluble oligomers rather than stable fibrils. This shift is particularly concerning, as soluble oligomers are believed to be more neurotoxic than amyloid fibrils, contributing significantly to cognitive decline in AD [43].

RAGE and the AGE-Amyloid Axis

RAGEs play a pivotal role in mediating the toxic effects of AGEs and glycated Aβ in AD. RAGE facilitates the transport of Aβ across the blood–brain barrier, promoting amyloid plaque formation while exacerbating neuroinflammation and oxidative stress [44, 45]. This interaction establishes a detrimental cycle of AGE accumulation, Aβ aggregation, and neuronal damage, with RAGE activation triggering further pathological processes such as tau hyperphosphorylation through the glycogen synthase kinase-3 (GSK-3) pathway, contributing to neurofibrillary tangle formation and AD progression [46, 47].

RAGE activation also induces oxidative stress and inflammation via the production of reactive oxygen species (ROS) and the activation of NF-κB, which releases inflammatory cytokines and cellular apoptosis, further compromising neuronal health [48]. Since RAGE expression is elevated in AD brains, therapeutic strategies targeting the AGE-RAGE axis are promising for mitigating disease progression. Soluble RAGE (sRAGE), which acts as a decoy receptor, competes with membrane-bound RAGE for binding to Aβ and AGEs, potentially reducing inflammation and neuronal damage [48].

Oxidative Stress Role in Neurodegeneration

Oxidative stress plays a central role in the pathophysiology of T2DM and AD. In T2DM, hyperinsulinemia and impaired glucose metabolism increase NADH, FADH₂, and ATP levels, driving ROS production. Elevated ROS promotes non-enzymatic glycation and AGE formation, which activate RAGE—a pathway implicated in AD progression [49]. ROS also disrupts neuronal glucose metabolism by damaging key enzymes in glycolysis and the TCA cycle, reducing ATP synthesis [50]. In APP/PS1 models, metabolomics and redox proteomics have revealed widespread metabolic dysfunction, oxidative damage, and hyperglycemia, alongside oxidation of ATP synthase, aldolase, and α-enolase—proteins critical for energy homeostasis and synaptic function [51]. These impairments parallel AD features such as synaptic loss, Aβ deposition, and NFT accumulation. Additionally, T2DM exacerbates CNS insulin resistance and early Aβ accumulation, triggering a feedback loop of ROS, inflammation, and further IR [52]. This bidirectional relationship underscores oxidative stress as a shared driver of T2DM and AD, offering a potential therapeutic target.

Glycation and Tau Hyperphosphorylation

Glycation amplifies neurodegeneration and cognitive decline in AD. Cross-linked AGE-modified Aβ and tau aggregates accumulate and resist clearance, disrupting proteostasis and accelerating synaptic loss [53]. Chronic AGE–RAGE signaling in neurons and glia sustains oxidative stress, inflammation, and tau hyperphosphorylation through NF-κB and kinases such as GSK-3β [36, 54]. Experimental models show that blocking AGE formation or RAGE activation with agents like aminoguanidine preserves synaptic proteins and rescues memory deficits [36, 55]. Clinically, conditions that elevate glycation, including aging and diabetes, roughly double the risk of developing AD, reinforcing the pathogenic role of AGEs in disease progression.

Glycation promotes tau hyperphosphorylation through RAGE-mediated activation of GSK-3, leading to NFTs [35]. NFTs, composed of hyperphosphorylated tau, disrupt neuronal microtubule dynamics and are closely linked to synaptic dysfunction and cognitive decline in AD [56].

AGEs colocalize with tau in AD brain tissues, amplifying oxidative stress and enhancing tau aggregation. This interaction further destabilizes neuronal function, promoting the progression of AD (Chaudhuri et al., 2018). The combined impact of AGE-induced Aβ aggregation and tau hyperphosphorylation underscores the importance of glycation in driving neurodegeneration. AGEs contribute to AD pathology by disrupting insulin signaling and promoting oxidative stress, both of which drive tau hyperphosphorylation and Aβ accumulation. In T2DM, AGEs impair insulin pathways and reduce PP2A activity [57], a key regulator that normally prevents tau hyperphosphorylation. This dysfunction—worsened by hypothermia commonly observed in T2DM—leads to tau misfolding and aggregation into NFTs [58]. These aggregates resist autophagic clearance, causing intracellular buildup, ROS generation, and mitochondrial dysfunction. The resulting oxidative stress triggers apoptotic pathways and synaptic degeneration, both critical in AD progression. Additionally, insulin resistance affects MAPK and AKT signaling [59] further destabilizing tau. Insulin oligomers accumulate in neurons with hyperphosphorylated tau, a common feature in AD and other tauopathies [60]. Diabetic AD models and CSF data reveal distinct tau phosphorylation patterns, especially in APOE ε4 carriers [61]. Elevated tau levels and reduced Aβ1–42 have also been observed in individuals with both T2DM and cognitive impairment, reinforcing the link between AGEs, tau, and Aβ pathologies [62].

The Mechanistic Link Between AGEs, T2DM, and AD

The pathological link between T2DM, glycation, and AD involves impaired insulin signaling, hyperglycemia, and inflammation—all of which drive neurodegeneration. In T2DM, brain insulin resistance disrupts PI3K/AKT/GSK-3β signaling, leading to altered APP processing, increased Aβ accumulation, and tau hyperphosphorylation—core features of AD [63, 64]. Reduced insulin transport across the BBB, driven by systemic IR, aging, and inflammation, contributes to neuronal insulin deficiency and signaling failure. Hyperglycemia accelerates non-enzymatic glycation, resulting in AGE formation, which promotes protein misfolding and aggregation (Aβ, tau) and activates RAGE, triggering ROS and chronic neuroinflammation [65]. IR also impairs IDE activity, limiting Aβ clearance [66]. These changes, along with mitochondrial dysfunction, ceramide accumulation, and cytokine release, result in synaptic loss and cognitive decline. Together, these mechanisms underscore how T2DM and glycation contribute to AD progression and support targeting insulin-related and glycation pathways as potential therapeutic strategies.

Biomarkers of Glycation in AD

Several glycation-related molecules have been proposed as potential biomarkers for AD. AGEs such as CML, CEL, and pentosidine accumulate in AD brains. Serum pentosidine levels are significantly elevated in AD compared to controls, and colocalize with Aβ plaques and tau tangles [67, 68]. CML is prominently detected in AD brain lesions and correlates with tau pathology. Circulating extracellular vesicle (EV) levels of carboxymethyllysine (CML) have been shown to distinguish early to moderate stages of Alzheimer’s disease. Within the brain, CML-modified neuronal proteins may spread intercellularly through EV-mediated transfer CML-modified proteins are more susceptible to oxidation and crosslinking, exacerbating neuronal dysfunction [69]. For instance, a small study reported detectable CSF pentosidine levels, which were decreased in AD compared to controls, suggesting altered AGE metabolism [70]. Notably, pentosidine levels distinguished AD patients from diabetic and age-matched healthy controls despite comparable renal function and were unrelated to conventional glycemic indicators such as HbA1c and fructosamine. This highlights pentosidine’s potential diagnostic value in Alzheimer’s disease beyond traditional markers of glycation [67]. In addition, elevated methylglyoxal promotes Alzheimer’s disease progression by enhancing AGE accumulation, oxidative stress, and tau hyperphosphorylation, ultimately driving neuronal dysfunction; thus, targeting MG detoxification and improving glucose metabolism may represent promising therapeutic strategies to reduce carbonyl stress and cognitive decline in AD [54]. Other circulating glycation markers, such as fructosamine, glycated albumin, and HbA1c, are also elevated in AD cohorts but lack specificity for distinguishing AD from DM. RAGE and its soluble form (sRAGE) have also been investigated as biomarkers. Plasma levels of RAGE ligands, including S100A12 (EN-RAGE) and HMGB1, correlate with dementia risk. A large cohort study (n ≈ 3890) found plasma RAGE ligands associated with prevalent dementia and skin AGE levels (measured by autofluorescence) correlated with cognitive impairment [71]. However, RAGE-based measures are not yet validated for diagnosis and are mainly explored as therapeutic targets. Experimental RAGE-PET imaging probes are under development for early detection [22].

Additional markers include TAGE and oxidative glycation products. Elevated MG-derived adducts, such as MG-H1, have been observed in AD, with age-related declines in Glo1/2 contributing to MG and AGE accumulation [39]. Although no glycation markers are part of routine AD diagnostics, they show promise for monitoring disease progression. Higher pentosidine levels have predicted faster cognitive decline [68]. However, no glycation biomarker is FDA-approved for AD, and they are currently under investigation as adjuncts to classical biomarkers (Aβ, tau) for early detection and tracking [67].

Implications and Therapeutic Opportunities

The significance of glycation in AD highlights the critical role of metabolic health in neurodegeneration. Glycation is accelerated in conditions like T2DM, where chronic hyperglycemia increases AGE formation, leading to a heightened risk of AD. Chronic hyperglycemia increases the production of ROS, leading to oxidative damage in neurons. This stress environment contributes to both direct neuronal injury and activation of inflammatory pathways, setting the stage for neurodegeneration [72]. Elevated glucose levels promote the formation of AGEs, which trigger inflammation via the RAGE receptor. Human studies show that individuals with both T2DM and AD have the highest AGE levels, linking glycation stress to cognitive decline and brain inflammation [73]. T2DM-associated insulin resistance affects the brain’s ability to use glucose efficiently. This results in neuronal energy deprivation, impaired protective signaling, and increased vulnerability to AD pathology—supporting the idea of AD as a “type 3 diabetes” [74]. Together, oxidative stress, AGEs, and impaired metabolism drive chronic neuroinflammation. This leads to microglial activation, cytokine release, and ultimately neuronal and synaptic loss. Imaging studies confirm greater brain atrophy in individuals with T2DM, linking diabetes to accelerated neurodegeneration in AD [75].

Targeting the glycation process, particularly the AGE-RAGE axis, offers promising therapeutic potential for mitigating AD progression. Lifestyle interventions that reduce AGE formation, such as dietary modifications, improved glycemic control, and RAGE inhibitors, may provide viable strategies for slowing cognitive decline and synaptic dysfunction in AD [76, 77]. Further research into glycation’s role in AD, particularly its impact on Aβ and tau pathology, is crucial for developing targeted therapies to restore proteostasis, reduce oxidative stress, and prevent neuroinflammation.

Regional Vulnerability to Insulin Resistance and Metabolic Dysfunction in AD

Insulin-related cognitive decline often affects brain regions and functions similar to those impacted in AD, but it can emerge prior to the development of overt pathology. Memory deficits are particularly common, as the hippocampus and medial temporal lobe (MTL)—regions heavily reliant on insulin signaling for glucose uptake and synaptic plasticity—are among the first affected. Insulin resistance (IR) in late-middle age has been associated with reduced hippocampal volume and impaired episodic memory performance [84]. Even cognitively normal individuals with elevated IR demonstrate poorer verbal memory scores [85]. Executive functions and processing speed may also decline, with midlife hyperinsulinemia linked to accelerated deterioration in verbal fluency—a frontal-executive task—independent of amyloid burden [86]. Cross-sectional studies by Benedict et al. further showed that IR correlates with reduced grey matter (GM) volume in the temporal and frontal lobes and with lower verbal fluency in healthy older adults. These findings suggest that insulin dysregulation primarily affects memory, fluency, and attention, implicating the hippocampus and frontal cortex as key vulnerable regions. Importantly, such cognitive impairments can manifest in individuals without significant amyloid or tau pathology; for instance, patients with diabetes mellitus or prediabetes often experience mild cognitive deficits or “brain fog” driven by metabolic dysfunction rather than classical AD lesions. In individuals with established AD, comorbid IR appears to exacerbate global cognitive decline and brain atrophy beyond what would be expected based on amyloid or tau pathology alone [87]. Overall, insulin-related dysfunction seems to selectively impair memory circuits and associative cortical areas, paralleling the cognitive profile of AD but through distinct metabolic mechanisms.

Production of AGEs in T2D

Recent research highlights the critical role of AGEs in inflammation and metabolic health. Elevated levels of AGEs are associated with increased inflammatory markers, independent of age, gender, body mass index (BMI), or smoking status [113]. This association extends to nondiabetic individuals after adjusting for BMI, waist circumference, and hemoglobin A1C, suggesting that AGEs may influence insulin resistance and overall metabolic health even in those without diabetes [114].

Glycation affects insulin sensitivity through several mechanisms (Fig. 3). It modifies insulin at the phenylalanine residue on the beta chain, reducing its effectiveness by approximately 20% and necessitating higher doses for effective glucose management [115, 116]. Furthermore, glycation of proteins such as albumin elevates tumor necrosis factor-alpha (TNF-α) levels, which induces inflammation and suppresses insulin signaling [117, 118]. Additionally, methylglyoxal (MG), an intermediate in AGE formation, disrupts AMP-activated protein kinase (AMPK) function, impairing hepatic insulin metabolism and increasing lipogenesis and gluconeogenesis, thereby exacerbating insulin resistance [91]. These findings underscore glycation’s complex and significant role in insulin resistance and metabolic dysfunction.

Insulin Signaling Impairment in AD Pathogenesis

Research reveals a strong link between changes in energy metabolism and AD development.

Brain insulin resistance and impaired insulin/IGF‑1 signaling have emerged as independent drivers of cognitive decline in AD, even aside from amyloid-β and tau pathology. In AD brains insulin/IGF signaling deficits mirror diabetes‐like dysfunctions across neurons, glia, and vasculature, leading to multifactorial damage [119]. For example, reduced insulin-mediated PI3K/Akt activity and unchecked GSK‑3β in insulin‑resistant neurons impair synaptic plasticity and neuronal survival [120]. These signaling disturbances also provoke chronic neuroinflammation, oxidative stress, and energy failure—manifesting as decreased glucose uptake, lower ATP production, synaptic weakening, and myelin loss [121]. Notably, brain regions rich in insulin receptors (such as the hippocampus and medial temporal lobe) show disrupted insulin signaling in AD, and the degree of hippocampal insulin resistance predicts worse cognitive performance [122]. Thus, a “type‑3 diabetes”–like state of brain insulin resistance can directly compromise neuronal metabolism, synaptic function, and plasticity, driving memory loss and dementia independently of Aβ/tau aggregates [120].

Furthemore, Aβ oligomers inhibit insulin signaling by blocking the auto-phosphorylation of insulin receptors on hippocampal neurons, contributing to impairments in learning, memory, and overall neurological function [123]. These oligomers also disrupt the tumor necrosis factor-alpha (TNF-α)/c-Jun N-terminal kinase (JNK) pathway, leading to neuroinflammation and further AD progression. By inhibiting insulin receptor substrate-1 (IRS-1), Aβ oligomers induce insulin resistance, which is linked to metabolic dysfunction and accelerates AD progression [124]. Insulin resistance exacerbates β-amyloid accumulation and plaque formation, creating a vicious cycle that promotes AD. Studies indicate that high-fat diets increase the levels of Aβ oligomers and result in memory impairment. Still, insulin injections can counteract these effects, underscoring the cyclical relationship between insulin resistance and Aβ oligomers [106]. Additionally, mice lacking insulin-degrading enzymes exhibit hyperinsulinemia, glucose intolerance, and elevated cerebral β-amyloid levels, further connecting insulin resistance to AD [125].

Peripheral and brain insulin resistance, while mechanistically related, represent distinct pathophysiological entities. Peripheral insulin resistance in T2DM affects classical insulin-responsive tissues such as muscle, liver, and adipose tissue, leading to systemic hyperglycemia and compensatory hyperinsulinemia. In contrast, brain insulin resistance refers to impaired insulin signaling within the CNS, which can occur independently of peripheral metabolic dysfunction. This central impairment disrupts neuronal glucose uptake, synaptic maintenance, and neuroplasticity, contributing to cognitive decline in AD [126, 127]. These distinctions complicate the therapeutic repurposing of antidiabetic agents for AD. One of the primary challenges is BBB permeability, as many systemic antidiabetic drugs—including insulin, GLP-1 analogs, and thiazolidinediones—have limited ability to cross into the CNS [128]. Intranasal insulin delivery has shown promise in overcoming this barrier while avoiding peripheral hypoglycemia. Moreover, achieving effective central drug concentrations may necessitate altered dosing regimens, which increases the risk of systemic adverse effects, particularly in older patients. The pharmacodynamics of these agents may also differ in the brain, potentially producing cognitive effects that vary with individual factors such as APOE genotype. For example, PPARγ agonists have shown cognitive benefit only in certain subgroups of AD patients [129]. These considerations highlight the need for precise delivery systems, dose optimization, and patient stratification when targeting insulin signaling pathways in AD. This interplay highlights potential therapeutic targets for addressing both metabolic dysfunction and neurodegeneration.

Insulin Signaling and Tau Hyperphosphorylation

The tau protein, which is crucial for stabilizing microtubules in axons, becomes hyperphosphorylated due to impaired insulin signaling, leading to the formation of NFTs and subsequent neuronal degeneration in AD [130]. Increased activity of glycogen synthase kinase three beta (GSK-3β), resulting from reduced Akt signaling, promotes tau hyperphosphorylation and NFT formation [131]. Research suggests that inhibiting GSK-3β and enhancing insulin or insulin-like growth factor 1 (IGF-1) signaling can help regulate tau phosphorylation and prevent neurodegeneration [132]. The link between AD and T2DM is particularly notable due to impaired insulin signaling. Elevated GSK-3β activity in AD increases Aβ production and tau phosphorylation, exacerbating neuronal damage and synaptic impairments [20, 133]. Targeting insulin signaling and GSK-3β activity presents a promising strategy for treating neurodegeneration in both AD and related metabolic disorders.

Glycation and Gut Microbes in AD and T2DM

Imbalances in the gut microbiome can significantly influence levels of AGEs and the progression of AD (Fig. 4). Certain gut bacteria convert dietary components into AGE precursors, and their dysbiosis can lead to inflammation and neurodegeneration. Maintaining gut health is crucial for managing AD, particularly by focusing on short-chain fatty acids (SCFAs), the impact LPS, and the effects of aging on microbial populations.

SCFAs, produced from dietary fibers, play a vital role in gut-brain communication, reducing neuroinflammation and supporting the integrity of the gut barrier. Other microbial metabolites, such as dopamine and serotonin, also contribute to brain health, underscoring the necessity of a balanced gut microbiota and sufficient fiber intake.

Microbial imbalances can increase gut permeability, allowing bacterial products like LPS to enter systemic circulation, leading to inflammation and enhanced AGE production and accumulation. Elevated levels of LPS and other microbial components have been observed in AD patients, linking gut health to cognitive impairment [134–137].

Shifts in gut bacteria associated with AGEs can exacerbate inflammation and damage the intestinal barrier, potentially accelerating the progression of AD [138]. Aging alters gut microbiota composition, increasing oxidative stress and mitochondrial dysfunction [139], while germ-free mice exhibit reduced oxidative stress [140]. Persistent gut inflammation resulting from aging can lead to neuroinflammation and compromise the blood–brain barrier [141, 142].

Microbiota transfer from aged to young rats impairs cognitive function while modifying gut microbiota in AD models can reduce disease markers. For instance, antibiotic treatment in 5xFAD mice decreased Lactobacillaceae, reduced plaque formation, and lowered RAGE expression [143]. These findings highlight the significant impact of gut microbiota changes on AGE accumulation and neurodegeneration, suggesting that dietary and probiotic interventions could improve mental health and mitigate age-related diseases.

Therapeutic modulation of the gut microbiota using probiotics (Lactobacillus, Bifidobacterium), prebiotics (e.g., inulin), and dietary fiber has shown promise in restoring microbial balance and mitigating inflammatory processes. Additionally, fermented foods such as kefir and pharmacologic agents like metformin have been linked to improved cognitive outcomes and favorable microbiome alteration. Twelve weeks of probiotic supplementation have been shown to improve oxidative stress markers, reduce inflammation, and enhance both quality of life and physical activity in individuals with mild to moderate Alzheimer’s disease [144] Furthermore, probiotic strains like Lactobacillus acidophilus have demonstrated antidiabetic effects and support intestinal health [145]. These findings support the potential of gut microbiota-targeted interventions as adjunctive strategies in managing metabolic and neurodegenerative disorders.

Lifestyle as a Link Between AD and T2DM

AD and T2DM share several pathophysiological features, as T2DM—often associated with obesity and sedentary lifestyles—leads to chronic hyperglycemia that can damage the brain. Notably, both conditions exhibit insulin resistance, impaired glucose metabolism, and amyloid-beta aggregation, which has prompted some to label AD as “type 3 diabetes” or the “diabetes of the brain” [146–148]. Data from large-scale studies, including UK Biobank neuroimaging and cognitive assessments, reveal that T2DM is linked to deficits in executive function and processing speed, along with gray matter atrophy in regions such as the ventral striatum, cerebellum, and putamen—changes that mimic, yet appear earlier than, normal aging, with greater neurodegeneration associated with longer disease duration. Importantly, metformin treatment did not yield cognitive improvements, suggesting that mere glycemic control may be inadequate for preserving brain health [149]. Additionally, research involving 5653 participants demonstrated that diabetic individuals experienced a 45% faster decline in cognitive domains like memory and reasoning compared to a 29% decline in normoglycemic individuals, highlighting the significant role of both disease duration and glycemic control in accelerating neurocognitive decline [150].

Dietary Intake Implications in AD and T2DM

Diet plays a crucial role in regulating glucose levels and supporting cognitive health in individuals with T2DM and AD. Both excessive caloric intake and overly restrictive diets can disrupt metabolic balance, but consistent dietary monitoring can improve glycemic control and may reduce neurodegeneration by limiting metabolic stress and AGE formation [2]. For elderly patients with both T2DM and AD, a less aggressive approach to glucose control—allowing slightly higher blood sugar targets—has shown better outcomes, possibly by minimizing oxidative stress from strict glycemic management [151].

In terms of diet, a balanced intake of macronutrients and low-glycemic index foods can help regulate glucose spikes and reduce AGE accumulation, which is crucial in managing both conditions [152]. The Mediterranean diet has demonstrated particular benefits, improving insulin sensitivity, supporting cardiovascular and cognitive health, and reducing AGE-related damage by emphasizing fruits, vegetables, legumes, whole grains, olive oil, and moderate fish and dairy intake, while limiting red meat [153]. The DASH and MIND diets have also shown promise in reducing AD risk, although the evidence is less extensive [154].

The ketogenic diet, by restricting carbohydrate intake and increasing fat consumption, may decrease endogenous AGE formation and improve mitochondrial function, reducing neuroinflammation and oxidative stress. These effects, along with improvements in weight, lipid profiles, and HbA1c, are beneficial for T2DM patients [155].

Genetics

Recent genomic studies have begun to illuminate shared molecular underpinnings between T2DM and AD, offering new perspectives on risk stratification and targeted therapies. One study utilized non-negative matrix factorization (NMF) to analyze gene expression profiles and identified 241 differentially expressed genes (DEGs), including ACTB, SERINC3, and ZMIZ1, which are involved in immune-related pathways such as T cell activation and chemokine signaling—both of which are increasingly recognized in the pathology of T2DM and AD [156]. Similarly, another investigation applied NMF to cortical neuron datasets and, through protein–protein interaction analysis, highlighted overlapping gene networks involving CDKN1A, COL22A1, EIF4A, GFAP, SLC1A1, and VIM, further supporting the hypothesis of genetic convergence in these two diseases [157].

In parallel, integrative genome-wide association study (GWAS) analyses have provided complementary evidence. Using a conditional false discovery rate framework, Wang et al. [158] identified 78 AD-associated single-nucleotide polymorphisms (SNPs), many of which are also implicated in T2DM. Among these, genes such as TP53INP1, TOMM40, and NDUFAF6 emerged as key players in mitochondrial regulation and oxidative stress—pathways known to bridge metabolic dysfunction and neurodegeneration. These findings reinforce the notion of a shared genetic architecture and highlight novel targets for future interventions that address the metabolic-inflammatory axis linking T2DM and AD.

Modulation of the Glycation Pathway

Most anti-diabetic drugs show mechanistic promise for reducing glycation stress. Preclinical studies consistently show metformin, GLP-1 agonists, DPP-4i, SGLT2i, and TZDs suppress AGE/RAGE signaling, oxidative stress, and neuroinflammation [162]. Observational cohorts suggest that GLP-1RAs, SGLT2i, and DPP-4i are associated with lower dementia incidence in T2D patients [163]. Small trials report cognitive benefits with liraglutide or intranasal insulin [39]. However, large RCTs are sparse or negative: exenatide and TZD trials showed no benefit [164], and intranasal insulin trials have mixed results [165]. Thus, despite strong mechanistic rationale, clinical evidence for AD prevention/treatment with these drugs remains preliminary. Anti-diabetic drugs may have a “dual” effect on diabetes and AD by mitigating glycation-related toxicity. Mechanistic studies indicate that many classes (metformin, incretin therapies, SGLT2 inhibitors, TZDs) attenuate AGE formation or RAGE activation, thereby reducing downstream inflammation and neurodegenerative processes [166]. Some human data support cognitive benefits (especially for GLP-1 agonists and perhaps SGLT2i [167]. Nevertheless, definitive proof in AD populations is lacking. Future work should fill gaps with well-designed trials of these agents in AD and clarify how much of their neuroprotective effect comes from direct anti-glycation actions versus general metabolic control.

Intranasal Insulin and Its Benefits for AD Cognitive Improvement

Insulin, a hormone produced by the pancreas, regulates blood glucose levels by facilitating glucose uptake into cells for energy production or storage. In the context of AD, insulin treatment has been shown to mitigate the harmful effects of proteins such as Aβ and tau, which are critical contributors to the formation of amyloid plaques and neurofibrillary tangles—hallmarks of AD pathology. By modulating these proteins, insulin may reduce their aggregation and toxicity, potentially alleviating some neurodegenerative processes associated with the disease [168].

Research indicates that increased insulin levels can enhance memory in AD patients, even without high blood sugar, highlighting insulin’s role in cognitive function [169, 170]. Intranasal insulin delivery, which bypasses the risk of hypoglycemia, has been shown to rapidly reach key brain areas, such as the cortex and hippocampus, within 15–30 min [171]. A pilot study involving 24 patients with mild cognitive impairment (MCI) or early AD demonstrated that a 3-week intranasal insulin intervention improved working memory and cognitive skills [172]. Furthermore, a more extensive study with 104 patients experiencing MCI or mild to moderate AD reported cognitive and functional improvements after 4 months of chronic intranasal insulin administration [128]. These improvements were associated with changes in β-amyloid levels and the cerebrospinal fluid (CSF) β-amyloid/tau protein ratio, suggesting that intranasal insulin may be a viable therapeutic approach for AD without significant side effects.

A recent study investigating the effects of intranasal insulin detemir on patients with MCI and AD revealed enhancements in cognitive function, including verbal and audiovisual memory [169]. However, the study noted that insulin’s effects on cognitive function may vary depending on the ApoE genotype. Specifically, insulin appeared to benefit AD patients lacking the ApoE4 allele, suggesting a potential therapeutic advantage for this subgroup. Conversely, insulin administration could be less effective or even exacerbate symptoms in patients with the ApoE4 allele. This indicates the need for genotype-specific approaches when considering insulin treatment for cognitive decline in AD [173].

In summary, insulin therapy, mainly through intranasal delivery, holds promise for enhancing cognitive function and modulating harmful protein activities in AD. However, its efficacy may be influenced by genetic factors, such as the presence of the ApoE4 allele, emphasizing the importance of personalized treatment strategies.

Metformin in AD

Metformin is commonly prescribed for T2DM due to its effectiveness in managing hemoglobin A1c levels, promoting weight management, and reducing cardiovascular risk, making it a preferred initial treatment choice [174]. Metformin reduces hepatic glucose production, improves insulin sensitivity, delays glucose absorption, and inhibits mitochondrial functions, lowering NAD + and ATP levels. Long-term use of metformin has yielded mixed results: Infante linked it to cognitive impairment, potentially due to vitamin B12 deficiency [175]. Ng found that it reduced the risk of cognitive decline after four years [43]. Notably, metformin use in T2DM patients is associated with a 24% reduction in the risk of dementia compared to those not on antidiabetic medications [176]. It has shown benefits for executive function, memory, and attention, although it does not appear to affect cerebrospinal fluid AD biomarkers [177]. Some studies suggest a slight increase in AD risk with long-term metformin use, possibly related to vitamin B12 deficiency [177]. Overall, while metformin is effective for managing T2DM, its effects on cognitive function and dementia risk remain complex and warrant further investigation.

GLP-1 Agonists (Incretin) in AD

Glucagon-like peptide-1 (GLP-1) and glucose-dependent insulinotropic polypeptide (GIP) are incretins that stimulate insulin secretion, enhance beta-cell proliferation, and regulate glucose levels [178]. GLP-1, in particular, may reduce RAGE expression, thereby preventing inflammation and AGE accumulation. GLP-1 binds to its receptor (GLP-1R) in various tissues, activating the cAMP pathway to increase insulin release, inhibit glucagon secretion, and slow gastric emptying [179]. This mechanism also modulates inflammatory responses and reduces oxidative stress.

By decreasing RAGE expression, incretins can prevent chronic inflammation and tissue damage caused by AGEs, highlighting GLP-1’s potential in managing T2DM complications. Despite promising results in animal studies, current research has not demonstrated a reversal of AD pathology in humans. Further studies are needed to clarify the effects of incretins on AD across different stages in human subjects.

Neuroprotective Mechanisms of Antidiabetic Agents Beyond Glycemic Control

Neuroinflammatory processes are increasingly recognized as central drivers of neurodegeneration. In Alzheimer’s and Parkinson’s diseases, prolonged activation of central immune responses, particularly involving microglia and pro-inflammatory cytokines, contributes to progressive neuronal loss. Some antidiabetic drugs have been shown to counteract these effects by dampening key inflammatory signals [180]. Thiazolidinediones and similar agents are capable of downregulating major inflammatory mediators, including TNF-α and IL-6—cytokines strongly implicated in neurodegenerative progression. This cytokine suppression may help interrupt chronic inflammatory cycles in the brain and protect vulnerable neurons from immune-mediated injury [181].

The ability of antidiabetic drugs to influence microglial polarization has also emerged as a relevant mechanism. By encouraging a shift from pro-inflammatory (M1) to anti-inflammatory (M2-like) microglial states, these agents may reshape the neuroimmune landscape toward one more conducive to repair and regeneration [182]. Together, these anti-inflammatory actions suggest that certain antidiabetic medications could serve dual roles: metabolic regulators and modulators of harmful neuroinflammation.

Neurons rely heavily on mitochondrial function for energy production, redox balance, and calcium regulation. Impairments in these systems are common in neurodegenerative disorders. Antidiabetic drugs have been reported to support mitochondrial health through multiple mechanisms [183]. Certain agents, particularly those that activate the PGC-1α signaling axis, have been shown to promote the synthesis of new mitochondria. This biogenic effect enhances both the quantity and efficiency of mitochondria within neurons, supporting their metabolic demands [184]. By improving mitochondrial respiration and ATP synthesis, antidiabetic medications contribute to a more stable and resilient neuronal energy supply. These improvements help neurons maintain function in the face of metabolic stress and may limit progression of energy-deficiency-driven neurodegeneration [185]. In essence, the ability of these agents to stabilize mitochondrial performance offers an additional layer of protection against the cellular dysfunctions underlying neurodegenerative diseases.

Implications of Repurposing T2DM Treatment for AD

Despite growing interest in repurposing antidiabetic medications for the treatment of AD, several important limitations and risks must be considered. Agents such as GLP-1 receptor agonists and SGLT2 inhibitors have shown neuroprotective effects in T2DM populations; however, their efficacy in non-diabetic AD cohorts remains inconclusive, as most clinical data are drawn from diabetic patients [186]. Furthermore, the heterogeneity of AD pathology presents a challenge—these drugs primarily address insulin resistance and metabolic dysfunction, but may not directly influence Aβ or tau-related pathways. Trials investigating agents like metformin or intranasal insulin have yielded mixed results, with some cognitive benefits observed, while others show limited impact [186]. Safety concerns also exist, particularly in older adults with cognitive impairment. TZDs have been associated with fluid retention and weight gain, while sulfonylureas and insulin increase the risk of hypoglycemia, which can be detrimental in cognitively vulnerable individuals [161]. Additionally, polypharmacy in AD populations raises the risk of drug–drug interactions.

Similarly, targeting glycation pathways—another shared feature of AD and T2DM—must be approached with caution. While AGE inhibition has therapeutic potential, some agents such as aminoguanidine have shown hepatotoxicity and gastrointestinal side effects in trials [187]. Excessive suppression of AGEs may interfere with normal physiological roles in immune regulation and tissue repair [188], and chelating anti-glycation compounds may impair micronutrient absorption. Furthermore, modulating glycation could alter drug metabolism or interact with concurrent medications, complicating polypharmacy regimens. Natural compounds like polyphenols offer a safer alternative but often suffer from low bioavailability and uncertain long-term effects [80]. Together, these observations highlight the importance of designing targeted, selective interventions with rigorous clinical validation to ensure safety and therapeutic value in AD populations.

Clinical Trials and Anti-glycation Therapies

Targeting glycation in neurodegeneration presents significant challenges. AGE formation is non-enzymatic, largely irreversible, and chemically heterogeneous, making it difficult for inhibitors to comprehensively block their formation [22]. Anti-AGE drugs, such as aminoguanidine, have shown safety concerns, including liver dysfunction and vasculitis at higher doses [189]. Effective CNS delivery is another obstacle, as many AGE inhibitors and detoxifying agents have poor brain penetration, and RAGE blockade may interfere with its normal functions. Clinical failures, such as the Phase III trial of azeliragon, further highlight these barriers. Moreover, glycation damage accumulates slowly over time, meaning significant injury may already be established by the time AD is diagnosed. Thus, despite its promise, anti-glycation therapy faces substantial biological and translational hurdles (Pasten, 2021).

The oral small-molecule inhibitor azeliragon (TTP488, PF-04494700) was the lead RAGE inhibitor anti-glycation drug tested. A Phase 2 safety trial (10 weeks) found azeliragon was well-tolerated but produced no change in amyloid or cognition (Sabbagh, 2011). A larger multicenter Phase 2 trial (n≈399, 18 months) evaluated low vs. high doses; the high dose was dropped early due to safety issues and no cognitive benefit, and the low dose showed futility (Galasko, 2014). Ultimately, the 18-month Phase 3 “STEADFAST” trial (880 mild AD patients, 5 mg/day vs. placebo) was terminated early in 2018 after failing both primary cognitive endpoints. A follow-up Phase 2/3 trial in AD patients with type-2 diabetes (targeting those with high HbA1c) was also halted in 2020 for lack of efficacy. In sum, RAGE blockade with azeliragon did not slow AD progression; high-dose azeliragon even seemed to accelerate decline. These negative results led Pfizer/vTv to stop development of RAGE inhibitors for AD.

Thiamine-derived anti-glycation (Benfotiamine): Benfotiamine is a high-bioavailability vitamin B1 analog that shunts glycolysis intermediates away from AGE formation. In a 12-month Phase IIa trial (71 amyloid-positive subjects with MCI/mild AD), daily 600 mg benfotiamine significantly reduced the rise of blood AGE levels and slowed cognitive decline. ADAS-Cog scores increased 43% less in the benfotiamine group (trend, p≈0.13) and CDR-SB increased 77% less (p = 0.034) compared to placebo [190]. No drug-related adverse events occurred. These results suggest benfotiamine may slow AD-related decline and lower glycation [190]. A larger Phase II trial (18 months, 406 mild-AD patients, dose-ranging) is now underway (adaptive design, completion ~ 2027) [191]. Notably, a Chinese add-on trial in AD (n = 302 on donepezil) saw no overall benefit on ADAS-Cog, though a subgroup with moderate AD showed dose-dependent slowing of decline [192]. Furthermore, a medicinal chemistry study designed phenyl benzoxazole derivatives with dual AChE inhibitory and antioxidant activity. The most potent analog (compound 34) had an AChE IC₅₀ of 0.363 μM and significantly higher antioxidant effect than donepezil. In vivo and ex vivo assessments showed compound 34 markedly reversed scopolamine-induced memory deficits and oxidative stress at 5 mg/kg, effects comparable to donepezil [193].

Diabetes drugs (Metformin, GLP-1 agonists, insulin) are sometimes considered “indirect” anti-glycation therapies. A pilot 12-month RCT of metformin (1000 mg BID) in 80 overweight adults with amnestic MCI found memory improvement on one test (SRT total recall) favoring metformin (9.7 vs. 5.3 point gain, p = 0.02) [194], though other measures were unchanged. About 10–15% of subjects could not tolerate full dose [194]. Larger metformin trials are in progress (e.g., the MAP study, NCT04098666) to test its preventive effects on cognition. Similarly, GLP-1 receptor agonists (e.g., liraglutide, semaglutide) and intranasal insulin are being tested in AD (due to their metabolic/anti-inflammatory effects) though results are pending. These agents improve glucose metabolism and could reduce glycation, but they are not specifically AGE-targeting.

Several other anti-glycation agents like AGE breakers and carbonyl scavengers (e.g., aminoguanidine, pyridoxamine, alagebrium) have shown benefit in animal models of AD, but no large AD trials in humans have been completed. Carnosine supplementation (a natural glycation scavenger) improved cognition in some small studies of elderly subjects, but its role in AD has not been established. Overall, to date, no anti-glycation therapy has proven efficacy in AD. Completed trials (azeliragon, benfotiamine, small metformin pilot) have yielded mixed or null results. Trials in progress (high-dose benfotiamine, larger metformin studies, GLP-1 trials) will clarify if targeting metabolism/glycation can slow AD.

Ethics Approval and Consent to Participate

Not applicable.

Consent for Publication

Not applicable.

Competing interests

The authors declare no competing interests.