New developments in incretin-based treatments for memory and thinking problems linked to type 2 diabetes: studies in animals and humans

The pathophysiology of cognitive dysfunction in diabetes is not completely clear. Many evidence supported a potential causative role for vascular disease, hyperglycemia, hypoglycemia, insulin resistance and amyloid deposition. Individuals with diabetes have a 2–6 times higher risk of thrombotic stroke (22), and vascular disease appears to contribute to abnormalities in cognition in such patients. In patients with diabetes, an overall decrease in cerebral blood flow has also been demonstrated, the degree of severity is related to the duration of the disease (23). As in other organs, also in the brain hyperglycemia can induce alterations by increasing polyol pathway activation, formation of advanced glycation end products (AGEs), diacylglycerol activation of protein kinase C, and shifting glucose into the hexosamine pathway (24, 25). Animal studies showed an increased expression of AGEs and receptors for AGE (RAGEs) in glial cells and neurons and damage to white matter and myelin in diabetic mice (26). In post-mortem human studies, individuals with diabetes and Alzheimer's disease showed to have greater N-carboxymethyllysine (a type of AGE) staining on brain slices (27), and had a difference in the quantity of AGE-like glycated protein rich neurofibrillary tangles and senile plaques (28). Aragnoet al. (29) showed that galectin-3 (a proatherogenic molecule), RAGEs and the polyol pathway activation were increased in rat brains, while S-100 protein, a marker for brain injury that can bind to RAGEs, and Nuclear factor κB (NF-κB) transcription factors, a proinflammatory gene marker up-regulated by AGEs, were both up-regulated in the hippocampus. Moreover, in diabetic rats neurochemical changes have been observed, including an impairment of long-term potentiation (30), decreased dopamine and acetylcholine activity (31), decreased serotonin turnover and increased norepinephrine (32), suggesting a worsen neurotransmitter function. It is known that also repetitive episodes of moderate to severe hypoglycemia play a possible role in etiology of cognitive dysfunction in the young (33). In human patients who died of hypoglycemia, laminar necrosis and gliosis found in cortex, basal ganglia, and hippocampus (34). In other post-mortem human studies, multifocal or diffuse necrosis of the cerebral cortex has been demonstrated (35). Also insulin resistance contribute in impairment of cognitive function. In fact insulin plays an important role in brain, it is involved in regulation of glucose metabolism, growth of neurons, cognitive function. Alterations of insulin signalling may accelerate brain aging, impair plasticity, induce possibly neurodegeneration, alter permeability to insulin of the blood–brain barrier (BBB) (36), and downregulate the level of insulin receptors on the endothelium of the cerebral blood vessels (37). Moreover, it is now known that insulin has a neuroprotective action, preventing the formation of toxic beta-amyloid (Aβ), a component of amyloid plaques, a hallmark of AD (38). Several data suggested that insulin resistance leads to an Aβ accumulation, through increasing the activity of β-site amyloid precursor protein cleaving enzyme -1 (BACE-1), which is responsible for the proteolysis of amyloid precursor protein (APP) (39 –41), reducing the activity of insulin-degrading enzyme (IDE), which is involved in the degradation of β-amyloid and the intracellular domain of APP. B-amyloid accumulation maintains insulin resistance, through serine phosphorylation of insulin receptor substrate- 1 (IRS-1) (5). Moreover, solubilized amyloid-β can activate pro-inflammatory pathways, that amplify vasoconstriction and vascular inflammation (42). Insulin resistance promotes the hyperphosphorylation of the tau protein, which aggregated in intracellular neurofibrillary tangles (NFTs), resulting in progressive neuronal loss and neurodegeneration, especially in the hippocampus and olfactory epithelium (43). Several studies demonstrated that brain insulin resistance occur also in patients with AD and without T2DM, suggesting that insulin resistance in brain depends on glucose independent mechanisms (44). In the post-mortem brains from AD patients, De la Monte et al. (45, 46) showed decreased expression of insulin receptors, insulin-like growth factors 1 and 2 (IGF1 and IGF2), Insulin receptor substrate 1(IRS1), phosphatidylinositol 3-kinase/Akt (PI3K/Akt) and increased Glycogen Synthase Kinase 3 beta (GSK-3β) level. These alterations changes were inversely correlated with severity of AD, which indicates the relationship between impaired insulin signalling and progression of the disease (47).

Several evidences indicate that GLP-1receptor agonist can cross the blood–brain barrier and GLP-1Rs are localized is stated in the central nervous systems, including the frontal and occipital lobes, cerebellum, substantia nigra, the hippocampus, thalamus and hypothalamus (49 –51). Activation of GLP-1R in the brain can regulate numerous cellular processes such as neuroinflammation, oxidative stress, apoptosis, and mitochondrial dysfunction, and may promote cell survival, restore insulin signalling and improve neuronal functions (52 –54). Neuroinflammation is a central pathophysiological factor of neurodegenerative disorders and GLP-1 can mitigate inflammation response in the brain by reducing levels of pro-inflammatory cytokines such as Tumor necrosis factor-α (TNF-α) and Interleukin-1β (IL-1β) liberated by microglia and astrocytes (55, 56). The last cells are the most abundant type in adult brain tissue, and it has newly been documented for their crucial part in controlling glucose and energy homeostasis (57 –60). Therefore, the loss of GLP-1R in astrocytes can harm mitochondrial function, because GLP-1 prevents glucose uptake in astrocytes and stimulates beta-oxidation, that is important for energy balance in the brain and mitochondrial integrity (59). Furthermore, studies suggest that GLP1ra diminished oxidative stress via stimulation of antioxidant enzymes such as superoxide dismutase (SOD) and catalase through nuclear factor erythroid 2-related factor 2 (Nrf2) pathway (60). Moreover, being GLP-1 a growth factor, it stimulates growth of neurites, repair and replacement of cells, increases cell metabolism, and reduces apoptosis (61). In fact, inhibition of apoptosis is mediated by increase of anti-apoptotic B-cell lymphoma 2 (Bcl-2) and reduction of caspase-3 activation (62). Additionally, GLP1ra stimulates mitochondrial biogenesis and function via modulation of peroxisome proliferator-activated receptor gamma coactivator 1-alpha (PGC-1α) and sirtuin 1 (SIRT1) pathway, enhancing energy metabolism and decreasing reactive oxygen species (ROS) production (63). These drugs can induce neurogenesis and synaptic plasticity through cAMP-response element binding protein (CREB)/brain derived neurotrophic factor (BDNF) signalling pathway, thus promoting neuronal survival and synaptic functions (64, 65). Moreover, the neuroprotective effects of GLP-1RAs can be a consequence of the increased insulin signalling in the brain cells, leading to improved insulin sensitivity in the neurons (52). In the brain, GLP-1 can induce insulin release, through over-expression of the internalized insulin and IGF-1 receptors (55). Exendin-4 (Ex-4), a GLP-1RA, improved insulin resistance in neurons through insulin receptor substrate-1 (IRS-1), AKT, and GSK-3β pathways. GSK-3β participates to neuroprotection by reducing the levels of beta-amyloid and tau hyperphosphorylation in AD pathology (66). Tau protein is crucial for microtubule functions and glucose metabolism, but when it becomes hyperphosphorylated by specific kinases such as GSK-3β and cyclin-dependent kinase 5 (CDK5), loses its capacity to bind to microtubules, with consequent aggregation within neurons and in the extracellular space (Figure 2) (67 –70). And above all, the Ex-4 modulates the tau hyperphosphorylation and cognitive dysfunction by raising the Ins2-derived brain insulin via Wnt/β-catenin/NeuroD1 pathway in T2DM (71). Lastly, GLP-1RAs can improve memory and learning (72).

The biological action of GIP is promoted through binding to the GIP receptor (GIPR), a class B G-protein-coupled receptor that is similar to GLP-1R and belongs to the glucagon receptor family. The GIPR is expressed in the endocrine pancreas, in adipocytes, in myeloid cells, in the endothelium of the heart and blood vessels, the inner layers of the adrenal cortex and in the central nervous system (48).

Indeed, similar to GLP-1R, GIP receptors have been detected in many areas of the brain including the hippocampus and hippocampal progenitor cells. GIP and its agonists can cross the blood brain barrier and present significant neuroprotective effects on synapse function and numbers. They can stimulate neuronal proliferation, decreasing both amyloid plaques in the cortex and the chronic inflammation (73). Additionally, GIP analogues can act on memory formation (74 –76) and synaptic plasticity and, being GIP a neurotrophic factor, it can inhibit apoptosis of cerebellar cells (77). A novel long-acting GIP analogue exhibited to alleviate central pathological development in AD mice, with the underlying mechanism concerning to the inhibition of neuroinflammation and the upregulation of cAMP-/PKA/CREB signalling pathway (78). Systemic anti-inflammatory effects of GIP are related to reduced mRNA expression of macrophage chemoattractant protein 1 (MCP-1), vascular cell adhesion molecule 1 (VCAM-1), and intercellular adhesion molecule 1 (ICAM) together with decreased circulating levels of interleukin IL-1β and TNFα (79). The effect of GIP on redox balance is showed as decreased reactive oxygen species and release of nitric oxide (Figure 3) (80).

However, GLP-1 and GIP are rapidly degraded in plasma by dipeptidyl peptidase IV (DPP-4), resulting in a very short half-life of approximately 1.5 minutes (81). DPP-4 is a type-II integral transmembrane glycoprotein with enzymatic activity; it also exists as a soluble form in the plasma, lacking the cytoplasmic and transmembrane domains. The primary substrates for DPP-4's catalytic activity are incretins responsible for glucose metabolism, including GIP, GLP-1 and GLP-2 (82, 83). Consequently, DPP-4 inhibitors (DPP4-i) may have a pivotal role in prolonging the circulating half-life of GLP-1 and GIP, thereby preventing their cleavage (84).

At the preclinical level, in animal models and laboratory settings, there is some promising evidence and many interesting insights. Several studies have shown that DPP4-i can cross the blood-brain barrier, allowing them to act directly within the brain, and influence neuroinflammatory processes, the aggregation of pathological proteins, and neurodegeneration (87). It is well known in literature that DPP4-i have neuroprotective effects and modulate pathological pathway.

Kosaraju et al. demonstrate that the DPP4-i saxagliptin reduces beta-amyloid, tau accumulation (main culprits of neurodegeneration) and inflammatory markers and increases hippocampal GLP-1 and memory retention in mouse models of Alzheimer's disease. This occurs through modulation of clearance and production processes of these proteins. This effect is due to the improvement of GLP-1 levels and inverts behavioural deficits observed in AD (88).

On the side of inflammation, Zhuge et al. (2024) investigated the effects of the DPP-4 inhibitor linagliptin on mild cognitive impairment in mice, focusing on the regulation of microglia. They treated mice with either linagliptin or metformin for a defined period and analysed brain tissue for DPP-4 and markers of neuroinflammation levels and microglial polarization states. First of all, they proved that DPP-4 activity was enhanced in the hippocampus of middle-aged mice, primarily in microglia, but treatment with Linagliptin reduced the activation of M1 Macrofage-type (pro-inflammatory) microglia and promoted M2 Macrofage-type (anti-inflammatory) polarization. These protective effects appear to be partially mediated by macrophage inflammatory protein-1α (MIP-1α, a DPP-4 substrate) because conversely if MIP-1α was knocked out they were lost (87).

Interestingly, linagliptin does not easily cross the blood-brain barrier (BBB), but its neuroprotective effects may be partly due to increased GLP-1 activity, which can cross the BBB effectively (89). Studies in animal models of AD have shown that treatment with DPP-4 inhibitors (such as saxagliptin, vildagliptin, linagliptin, and sitagliptin) increases GLP-1 expression in the hippocampus and other various brain regions (88 –91). In these areas, GLP-1 bind to its receptor (GLP-1R), reducing amyloid deposition, tau levels, microglial activation, and memory decline (92). The preclinical studies performed in type 2 diabetes mellitus using the DPP-4 inhibitors are summarized in the Table 1.

Besides preclinical studies, in literature we found many clinical evidence regarding protective neuronal effect of DPP4-i. Firstly a retrospective study demonstrated that two years of DPP-4 inhibitor use was associated with reduced cognitive decline in diabetic patients with mild cognitive impairment (MCI). Although the study didn't include patients with Alzheimer's, it provided important evidence for a potential neuroprotective effect of DPP-i (93). Another retrospective study in 2021 highlighted that DPP-4i use reduces amyloid deposition and improves long-term cognitive outcome in diabetic patients with diabetic Alzheimer disease–related cognitive impairment (85). Similarly a prospective observational study evidenced that Diabetic patients with Alzheimer's treated with sitagliptin had significantly higher MMSE scores compared to those treated with metformin. This suggests a possible cognitive benefit of sitagliptin in this population (94). Furthermore, a study searched to explain the possible biological mechanism behind DPP-4 inhibitor linagliptin effect on AD. Serum Sirt1 levels are found to be significantly reduced in Alzheimer's patients, and a potential role for Sirt1 as a biomarker in this condition has been proposed (95). Linagliptin treatment increased Sirt1 expression in the peripheral blood leukocytes of diabetic patients with Alzheimer's disease (96). Sirt1 is an enzyme with known neuroprotective properties: enhances synaptic plasticity and memory (97), reduces Aβaccumulation, suppresses oxidative stress and neuronal loss (98). The role of DPP4-i has been compared also with sulfonylureas in a Korean diabetic patients' cohort. Data analysis demonstrated that DPP4-i administration was associated with a 34% lower risk of all-cause dementia compared with sulfonylureas in older patients with type 2 diabetes. In particular DPP4-i use reduced risk of Alzheimer's disease, but not vascular dementia, compared with sulfonylureas (99). Wu CY et al., in their study, searched to define which type of patients could profit from the use of DPP4-i. They evaluated how the use of metformin or DPP−4 inhibitors influences memory decline in patients with type 2 diabetes, comparing groups with: normal cognitive function (NC) and AD. Moreover, the role of the Apolipoprotein ϵ4 (APOE ϵ4) genotype was considered as a potential modifying factor. Indeed, Metformin administration was associated with better memory performance in normal cognitive function population, while in AD dementia population, DPP4 inhibitor use was associated with a slowdown of memory decline. Furthermore, APOE ϵ4 carrier status may predict better results of DPP4 inhibitors in normal cognitive individuals, and lower benefit of metformin in people with AD. So, these results also confirmed the need of a personalized antidiabetic therapy (100). The clinical studies performed in type 2 diabetes mellitus using the DPP-4 inhibitors are summarized in the Table 1.

Interest in GLP-1 agonists neurological effect begins in 2010 when McClean P.L. et al., demonstrated that liraglutide influences brain neurotransmission in rats and directly stimulates synaptic transmission in the hippocampus, enhancing long-term potentiation (LTP) (103). Similarly, Han et al. demonstrated that pre-treatment with liraglutide protects cognitive function (learning and memory) and reduces deficit of L-LTP induced by the bilateral intrahippocampal injection of amyloid- protein (25-35) in rats that had been administered amyloid-β (104). Also McClean P.L. et al. had previously demonstrated that periferical injection of liraglutide in Alzheimer mouse model (APP/PS1 mice) prevented memory deficits, reduced total amyloid plaques and dense-core plaques by 40–50%, and soluble oligomers by 25%, prevented synapse loss in the hippocampus and reduced the inflammation measured by microglial activation (reduced by 50%). These results suggested that use of liraglutide may be a promising strategy to prevent and ameliorate the cognitive function observed in AD. The effects of liraglutide was evaluated also on later stages of the AD (105). McClean P.L. et al. highlighted that liraglutide can reverse cognitive impairment, restore synaptic function, and reduce amyloid plaque burden even in later stages of the disease in aged APP/PS1 mice (mouse model of Alzheimer disease) (106). Moreover, it has been demonstrated that chronic liraglutide treatment stimulates neurogenesis in the hippocampus of Alzheimer's model mice, suggesting it may help restore neuronal populations lost due to neurodegeneration (107). Long-Smith et al. showed that intraperitoneal injection of liraglutide for 8 weeks in mice improves impaired brain insulin signalling reduces amyloid plaques, neuroinflammation and insulin resistance in mouse model of AD (108). More recently, Palleria et al. treated rats, with diabetes induced by streptozotocin (STZ), with liraglutide for 6 weeks evaluating cognitive function with behavioural tests, metabolic parameters, brain oxidative stress, neuroinflammation, and neurotrophic factors (109). Liraglutide significantly improved learning and memory in diabetic rats, as shown by better performance in behavioural tests. These cognitive benefits occurred without major changes in blood glucose or body weight, indicating the effect was central (brain-related) rather than due to improved metabolic control. In addition, liraglutide reduced oxidative stress and inflammation in the brain and increased levels of BDNF (brain-derived neurotrophic factor), which supports neuronal health. Interestingly, several studies evaluated the effects of treatment with liraglutide on neuroinflammation, stress oxidative. Liraglutide uses another molecular pathway to ameliorate brain insulin resistance, in particular inhibiting the c-Jun N-terminal kinase (JNK) pathway and activation of Bcl2 gene in T2DM mice (110). A study, using in vitro and in vivo AD models, evaluated the effect of liraglutide in SH-SY5Y human neuroblastoma cell line by investigating tau activation and Beta-site amyloid precursor protein cleaving enzyme 1 (BACE-1) expression (BACE-1 is a transmembrane aspartate protease, namely a key enzyme in beta-amyloid formation). Liraglutide protects SH−SY5Y neuronal cells from okadaic acid-induced apoptosis, reduces tau activation, and decreases the expression of BACE1 (111). Studies conducted in SH-SY5Y human neuroblastoma cells, a model of neurodegeneration, demonstrated that GLP-1 treatment has multiple effects: both neurotrophic and anti-inflammatory effects and reduction of neurodegeneration. Indeed GLP−1 enhances the survival of SH−SY5Y cells under glutamate- or ROS-induced stress, and it increases the activity of cyclic adenosine monophosphate(cAMP), protein kinase A (PKA), and AMP- activated protein kinase (AMPK). Moreover, it decreases IL−6 levels and TNF−α levels beyond that provides significant protection against toxicity induced by α−synuclein and β−amyloid (112). Using an AD mouse model, Park J.S. et al. highlighted that NLY01, a GLP-1 RA molecule, effectively inhibited microglia (that causes neuronal apoptosis) improving memory and survival of neurons (113). Regarding neuroinflammation, Iwai T. et al., developed a model of neuroinflammation induced by lipopolysaccharide, IL-1β, and hydrogen peroxide (H2O2) and demonstrated that GLP-1 treatment improves synaptic plasticity, preserves cognitive functions and has anti-inflammatory effect (114).

Exenatide: Also the exenatide, another GLP1-RA, seems to show effects on neuroinflammation and neuroprotection in animal models of AD. In 5 × FAD transgenic mice treated with exenatide was observed an improvement in cognitive function, a reduction in neuroinflammation and a reduction of oxidative stress. Exenatide exerts anti-inflammatory and neuroprotective effects likely through the inhibition of the NLR family pyrin domain containing 2 (NLRP2) inflammasome in astrocytes. This results in cognitive improvements in animal models of AD and a less pro-inflammatory brain environment (115). Similarly Qian Z. et al. demonstrated that GLP-1R activation reduced microglia-induced neuroinflammation in vitro and in vivo. Activation of GLP−1R in microglia represents an effective mechanism to reduce neuroinflammation and prevent glial scar formation after central nervous system injuries. The therapeutic effect is partly mediated by the restoration of the PI3K−ARAP3−RhoA pathway, which regulates glial remodelling and inflammatory responses. These findings suggest the potential use of GLP−1R agonists, such as exendin−4, in therapies for spinal cord injuries and possible chronic neuroinflammatory disorders (116). GLP-1 has also effects on neurogenesis. Indeed, in an AD model mice GLP-1 analogues promote neurogenesis in the hippocampus via mitogen activated protein kinases (MAPKs) (117). Zhang L.Q. et al., using a murine hippocampus model with memory deficits caused by neuropathic pain, demonstrated that exenatide acetate (Ex-4) ameliorates memory deficit, reduces inflammatory state by inhibiting the phosphorylation of NF-kB, decreasing the expression of several cytokines (IL-1β and TNF-α), and improving postsynaptic density protein 95 (PSD95), that stimulates synaptic plasticity (118, 119). Two recent reviews have summarized the action of GLP1-RA on neuroinflammation, oxidative stress and neuroprotective functions. Activation of GLP−1R in the brain by GLP-1RA represents an effective intervention against neuroinflammation: it inhibits microgliosis and astrogliosis, suppresses NF−κB and inflammasome activity, reduces peripheral immune cell recruitment, protects against oxidative stress, and promotes neuroprotective signalling (120). Similarly, in a very recent review, De Giorgi R. et al., further confirms that GLP-1 receptor agonists cross the blood–brain barrier and exhibit neuroprotective effects, including: reduction of oxidative stress, improvement of brain insulin signalling, decrease in neuroinflammation, protection against neuronal degeneration. Studies in animal models of Alzheimer's disease confirm improvements in memory, as well as reductions in amyloid plaques and phosphorylated tau (121). Lastly the neuroprotective effects of GLP-1RA by modulating systemic and brain inflammation were demonstrated also in a recent post hoc analysis of the randomized, placebo-controlled EXSCEL trials. In particular treatment with exenatide significantly reduced levels of inflammatory proteins associated with Alzheimer's disease in patients with type 2 diabetes (122). Moreover a recent systematic review and meta-analysis assessed the impact of different drugs for the treatment of diabetes on the risk of developing dementia, mild cognitive impairment, or cognitive decline highlighting that incretin-based therapies, particularly GLP-1 receptor agonists, were associated with a significant reduction in the risk of cognitive impairment compared to other antidiabetic treatments. This neuroprotective effect is thought to result not only from improved glycemic control but also from direct mechanisms involving modulation of brain inflammation, neuronal protection, and enhanced synaptic plasticity. Metformin, sulfonylureas, insulin, and DPP-4 i studies had inconsistent results. This heterogeneity among included studies emphasized the need for further controlled clinical trials to confirm these findings and explain underlying mechanisms (123).

Semaglutide: Very interesting are the preclinical data on cognitive function obtained using the semaglutide, a new GLP-1RA. A recent study investigated the effect of semaglutide on anxious and depressive behaviours as well as cognitive deficits in a mouse model of type 2 diabetes. Administration of semaglutide significantly reduced anxious and depressive behaviours in diabetic mice. It also improved impaired cognitive functions, such as memory and learning. The beneficial effects were mediated through modulation of the microbiota-gut-brain axis. Indeed, semaglutide positively changes the composition of the gut microbiota, reducing systemic and brain inflammation. This contributed to restoring neurochemical balance and brain function (124). The effects of semaglutide on cognitive deficits and glucose metabolism dysfunction in a mouse model of Alzheimer's diseasewas examined in another preclinical study. Results demonstrated that semaglutide significantly improved the cognitive performance of the mice and corrected alterations in brain glucose metabolism. The beneficial effectswere mediated through the GLP-1R/SIRT1/GLUT4 pathway, which regulates glucose uptake and utilization in the brain, as well as influencing neuroprotective processes (125). A recent study in APP/PS1/tau mice showed that intraperitoneal administration of semaglutide improves working and spatial memory and reduces brain inflammation by promoting the shift of microglia from the pro-inflammatory M1 state to the anti-inflammatory M2 state. Treatment also decreased amyloid-beta deposits and inflammatory cytokines in the hippocampus. These results suggest that semaglutide's neuroprotective effects involve modulation of microglial activation, adding an important immunoregulatory mechanism to the known metabolic benefits of GLP-1 receptor agonists (126). In contrast, FornyGermano L. et al. evaluated the effects of semaglutide and tirzepatide, a dual agonist GLP-1/GIP receptors, on murine models of Alzheimer's disease (5XFAD and APP/PS1 mice). However, no significant changes were observed in disease-related pathology, such as amyloid plaques or neurodegeneration. No improvement in behaviour or cognitive functions was found in treated mice compared to controls. It's true that in these specific Alzheimer's models, semaglutide and tirzepatide do not appear to affect disease progression or cognitive and behavioural abilities, suggesting that their therapeutic potential may be limited or dependent on the model and experimental conditions (127).

The preclinical studies performed in type 2 diabetes mellitus using the GLP-1 receptor agonist are summarized in the Table 2.

Regarding clinical evidence of GLP-1 effects on cognitive function, there are less evidence than in vivo and in vitro studies but several trials in humans are ongoing in humans and there is currently an increasing interest on these drugs although the study results are conflicting.

Exenatide: In 2019, the results of a pilot study on exenatide action in AD were published. The study demonstrated that exenatide was safe in patients with early Alzheimer's disease and seems to have a biological effect (reduction of Aβ in extracellular vesicles), but it does not show cognitive or clinical benefits. However, the study was underpowered due to early termination and therefore the authors could not draw any firm conclusions (128). Similarly, in a study where 20 obese patients with schizophrenia were treated with exenatide (once weekly for 3 months), GLP-1 RA had no significant effects on cognitive functions (129).

Liraglutide: Conflicting results have also been obtained with liraglutide. A 26-week double-blinded, randomized clinical trial including 38 subjects with early/moderate AD evaluated the effects of liraglutide compared to placebo on cognitive function. Researchers used Pittsburgh compound B (PIB) positron emission tomography (PET) scans to evaluate the changes in the intracerebral amyloid deposits in the central nervous system. No changes in amyloid deposition or cognition were highlighted, but it may have been due to a short follow-up time (130). Another phase 2 multicenter, randomized, double-blinded study (ELAD study) involved always liraglutide. The ELAD study has enrolled 204 subjects with mild AD treated exactly with liraglutide or matching placebo. It was a robust protocol for a 12-month clinical trial designed to assess whether liraglutide was able to improve metabolic and cognitive decline in mild Alzheimer's disease, using a range of endpoints including neuroimaging, biomarkers, neuroinflammation, and cognitive function. Indeed, the primary outcome was the change in cerebral glucose metabolic rate from baseline to follow-up in two groups. The secondary outcomes were change in clinical and cognitive scores; change in magnetic resonance imaging volume; reduction in microglial activation and in tau formation; change in amyloid levels. The ELAD study is closed, and results highlighted that liraglutide induces changes in the brain but does not improve cognitive measures (131). More encouraging results were obtained in subjects with prediabetic conditions or type 2 diabetes mellitus. In a study by Valdini F. et al. forty (40) subjects with prediabetic conditions or with new onset T2DM received liraglutide (1.8 mg daily) or lifestyle counselling. The aim was to evaluated whether treatment with GLP-1RA could improve cognitive functions in prediabetic or diabetic subjects. The results showed that liraglutide slows down cognitive decline as reported by a psychological test evaluating attention, memory, and executive control (132). Similarly, in a prospective, parallel-group, open- label, phase III study was evaluated whether liraglutide could improve cognitive function in diabetic patients and if such improvement was associated with metabolic changes. Patients,with inadequately control with oral antidiabetic drugs or insulin, were randomized to receive liraglutide or placebo + standard of care group for 12 weeks. Patients assessments were performed by functional near-infrared spectroscopy (fNIRS) to monitor brain activation and neuropsychological cognition tests to evaluate memory, executive function, attention and verbal function. At the end of study, the liraglutide group achieved better scores in all cognitive tests, notably in memory and attention compared to control group. Moreover, liraglutide increased activation of the dorsolateral prefrontal cortex and orbitofrontal cortex. All the cognitive improvement in this group was correlated with morphological changes in brain regions. All these results are metabolism-independent, meaning they are not related to hypoglycemia or weight loss. The results suggest that liraglutide could be a targeted intervention for early cognitive decline, independent of its effects on metabolism (133). In 2018 Wu et al., enrolling 106 patients with T2DM and evaluating the association between serum GLP-1 concentration and mild cognitive function impairment in MCI, demonstrated that lower serum GLP-1 levels are correlated to cognitive dysfunction in patients with diabetes (134). In another randomized, prospective, open-label, parallel-group trial the comparative effects of liraglutide, dapagliflozin and acarbose in patients with T2DMhas been evaluated. The subjects inadequately controlled with metformin, were randomized to receive for 16 weeks liraglutide, dapagliflozin or acarbose. At baseline and post treatment, all participants performed a brain functional magnetic resonance imaging (MRI) scan and subjected to cognitive function assessments tests(such as Mini-Mental State Examination, MMSE, Montreal Cognitive Assessment, MoCA). The results showed that in the group treated with liraglutide, a statistically significant improvement in impaired odor-induced left hippocampal activation and an improvement of cognitive subdomains of delayed memory, attention and executive function was observed (135).

Dulaglutide: Recently, a post-hoc analysis of a phase 3 randomized, double-blinded, placebo-controlled trial (REWIND) assessed the effect of dulaglutide on cognitive impairment (CI) in T2DM. This study involved 9901 participants, 4949 randomized to receive dulaglutide and 4952 placebo. Follow-up lasts 5.4 years and cognitive function was evaluated only in 8828 subjects (4456 received dulaglutide and 4372 received placebo) using the MoCA and Digit Symbol Substitution Test (DSST). CI occurred in 4·05 per 100 patient-years in dulaglutide group and 4·35 per 100 patient-years in placebo one, but after post-hoc adjustment, the risk of CI in people with T2DM aged 50 years or above who had further cardiovascular risk factors was reduced by 14% of subjects in the dulaglutide arm (HR 0.86, 95% confidence interval 0.79–0.95; p = 0.0018). This analysis showed that long-term treatment with dulaglutidemay reduce risk of CI in patients with T2DM (136).

Semaglutide: Similarly, also the effects of semaglutide on cognitive functions are ongoing.Evoke and Evoke+ are two large scale, double- blind, placebo-controlled, phase III trials evaluating efficacy, safety, and tolerability of oral semaglutide in early-stage symptomatic AD. The primary endopoint is to evaluate whether semaglutide modifies biomarkers and improves neuroinflammation. The trials are ongoing and completion is expected in September 2025 (137). Interestingly, Wang W. et al., conducted a target trial emulation using nationwide real-world data of 116 million US patients, evaluating the association of semaglutide with AD diagnosis in patients with T2DM. The study explored whether semaglutide is linked to a lower risk of developing Alzheimer's disease compared to other diabetes treatments. The results suggest that semaglutide may reduce the risk of a first-time Alzheimer's diagnosis, indicating its potential neuroprotective effects beyond blood glucose management, which could help delay or reduce the onset of neurodegenerative diseases like Alzheimer's (138). Current innovations—such as dual and triple agonists, oral formulations, and less frequent dosing—mark a cultural shift, aiming to expand access and enhance effectiveness in tackling obesity, chronic diseases, and metabolic disorders. In particular they underline neuroprotective effects: observational studies suggest a reduction in the risk of dementia by up to 25% compared to metformin in elderly patients with type 2 diabetes, possibly due to the ability to cross the blood-brain barrier and modulate brain inflammation (139). Another 2024 review underlines Neuroprotection effect of GLP-1 RA. Observational studies link the use of GLP−1 receptor agonists to a reduced risk of Alzheimer's disease, dementia, ischemic stroke, and mortality—by as much as 45% in some studies (140). The clinical studies performed in type 2 diabetes mellitus using the GLP-1 receptor agonist are summarized in the Table 3.