T2DM has also been associated with an increased risk of dementia
[8][9], in particular, AD by 45–90%
[10][11]. The Rotterdam study was among the first to show an elevated risk of dementia with T2DM
[12]. Moreover, it has been shown that patients with T2DM have a higher risk of amnestic mild cognitive impairment (aMCI)
[13]. People who experience cognitive impairment combined with AD, when compared to people who experience only cognitive impairment, appear to be affected by the DM type and related complications as well as the antidiabetic treatment they receive
[14]. Insulin resistance and hyperglycemia as features of T2DM have a detrimental effect on cognitive abilities
[15], since insulin and insulin-like growth factor, also called somatomedin C (IGF-1), play an important role in cognitive ability, neural function, and development
[16].
2. Intranasal Insulin
Insulin performs many important functions in the brain related to food intake regulation, body weight, eating habits, and energy homeostasis
[17][18].
It was proposed that AD might be a metabolic disease of the brain, driven by insulin resistance and insulin-like growth factor (IGF-1) resistance
[19][20].
Several studies have shown that insulin administration in AD patients reduces the action of kinases that promote tau protein hyperphosphorylation and enhances β-amyloid clearance and synaptic plasticity
[21][22]. In fact, an earlier study by Craft et al.
[23] showed that in the case of elevated insulin without hyperglycemia, memory was enhanced in AD patients, thus supporting the important role of insulin in memory improvement. Consequently, it has been hypothesized that increasing insulin function in the brain might counterbalance AD pathology. However, peripheral administration of insulin in order to reach the brain carries the risk of hypoglycemic events and the difficulty of passing the blood–brain barrier. On the other hand, intranasal insulin avoids the risk of hypoglycemia as it bypasses the blood–brain barrier
[24] and through the nasal passages reaches the cortex and hippocampus within 15–30 min
[25].
In a small (n = 24) pilot study
[26] that examined a 3-week intervention in patients with MCI or early AD and compared intranasal insulin with placebo, improvements in working memory and cognitive skills were found due to intranasal insulin. Moreover, in a study by Craft et al.
[27] chronic administration of intranasal insulin for 4 months in 104 patients with MCI or mild to moderate AD improved cognitive and functional ability, with these changes being associated with alterations in levels of β-amyloid but also in the CSF β-amyloid/tau protein ratio. Insulin has been shown to inhibit the deterioration of the cerebral glucose metabolism rate in specific areas of the brain
[27]. It should be noted that in this study, intranasal insulin appeared to be an effective therapeutic approach for patients with AD, with no side effects due to prolonged treatment
[27].
Some of the clinical trials evaluated fast-acting forms of insulin while others tested longer-acting insulin analogs. In a more recent study
[28] in which researchers compared NPH insulin to insulin detemir and placebo in adults with MCI or AD, NPH insulin appeared to improve memory after 2 and 4 months compared to the placebo, while no significant effects of long-acting insulin were observed compared to the placebo. In addition, NPH insulin administration was associated with a decrease in tau-P181/β-amyloid ratio; however, various genetic factors such as APOE4 status affected insulin levels and insulin resistance
[28].
APOE4 is the strongest genetic risk factor for AD
[3], and about 25% of the population carries at least one ε4 allele
[5]. There has been an improvement following insulin administration in the cognitive function of AD patients who were not ApoE4 carriers, while no improvement was found in patients with APOE4; in some cases, the symptoms of the disease worsened
[26][29]. A recent study by Claxton et al.
[30] examined responses to intranasal administration of insulin detemir to MCI and AD patients, who showed improvements in cognitive, verbal, and audiovisual memory. In this study, APOE4 played an important role in the results, and, in contrast to the aforementioned study
[26], it seemed that the responses were regulated differently. Significant improvements in verbal memory and peripheral insulin resistance levels in APOE4 carriers were observed after three weeks of treatment, while no improvements were observed in ApoE4 non-carriers
[30].
In an ongoing Phase II/III clinical trial with the acronym SNIFF (Study of Nasal Insulin in the Fight Against Forgetfulness)
[31], two different insulin delivery devices were used in order to deliver 20 IU of insulin or placebo after breakfast and dinner to 240 patients with either MCI or early AD. After one year of treatment, no statistically significant effect of intranasal insulin on cognitive abilities was found in the main cohort of 240 patients who used one of the two devices. Nonetheless, a group of 49 patients who used another device exhibited a slowing of worsening in the subscale of ADAS-COG-12 and daily life activities at one year
[31]. It should be noted that in this study, the change in the insulin delivery device in the middle of the experiment may have played an important role and may have affected the results.
3. Metformin
Metformin is a biguanide that increases peripheral glucose uptake, suppresses gluconeogenesis in the liver, and increases insulin sensitivity in peripheral tissues. Metformin is the first drug prescribed in patients with T2DM, mainly due to the beneficial effects observed on hemoglobin A1c levels, weight, and cardiovascular mortality, as well as due to its safe action profile
[32]. Currently, clinical research data on the use of metformin in AD are limited, and the results are inconclusive.
Currently, clinical research data on the use of metformin in AD are limited and the results are inconclusive. Several studies in the last decade have shown that metformin may significantly improve cognitive dysfunction in patients with T2DM
[33][34]. Moore et al. (2013)
[35] observed an increased risk of cognitive impairment in patients with T2DM after long-term metformin treatment. On the contrary, Ng et al.
[36] reported that metformin reduced the risk of cognitive impairment in T2DM patients, aged 55 years and older, who were monitored for more than 4 years. In the first study
[35], it is possible that the negative results were due to vitamin B12 deficiency. According to the authors of this study
[35], vitamin B12 and calcium supplements alleviated the aforementioned vitamin B12 deficiency and had beneficial effect on cognitive function. In a study from Taiwan’s National Health Insurance that contains a large database of structured data about people aged 50 years and over, some of whom (n = 25,393) were diagnosed with T2DM and others were undiagnosed (n = 101,816), it was found that dementia prevalence increased by 2.6 times in patients with T2DM
[37]. In particular, it was found that metformin reduced dementia risk by 24% compared to patients who had not used any antidiabetic medication. In a small randomized control trial, a significant positive effect of metformin on executive function was found as well as some improvements in memory and attention, while there was no effect of metformin on CSF AD biomarkers
[38]. In contrast to the above evidence, in a case–control study of diabetic individuals (n = 7086), which assessed the risk of AD in relation to the type of antidiabetic drugs, it was found that long-term use of metformin caused a slight increase in AD risk, while no such effect was observed following long-term use of sulfonylureas, thiazolidinedione, or insulin
[39]. A possible explanation for this increased risk of AD and cognitive impairment may be a vitamin B12 deficiency, often seen after metformin use.
Based on the above evidence, there is high need to further investigate the role of vitamin B12 deficiency. Another important issue is the route of administration, since drugs such as metformin have been administered only via systemic routes and, as a consequence, their action depends on their ability to cross the blood–brain barrier (but also from the peripheral insulin levels). Given the widespread use of metformin and its effect on cognitive functions, additional research is needed, in particular, a long-term study with adequate sample or a meta-analysis of smaller studies in order to further elucidate its action.
4. Incretins
Incretins, including glucagon-1 peptide (GLP-1) and glucose-dependent insulin-releasing polypeptide (GIP), are important metabolic hormones responsible for the expression of the insulin gene, proliferation of ng β-cells, and lowering glucose levels by stimulating insulin secretion mechanisms
[40].
GLP-1 is secreted by the gut in response to food intake, and its receptors (GLP-1Rs), expressed in pancreatic β-cells, enhance insulin release in response to high glucose levels. Following the secretion of the GLP-1, the enzyme dipeptidyl-peptidase 4 (DPP4) degrades the GLP-1 within minutes. Therefore, GLP-1 analogs, which are resistant to the enzyme DPP4, have been developed for clinical use, and GLP1-R receptor agonists (liraglutide, exentin-4) have been approved for use in patients with DM
[41]. GLP-1 and its receptors are not found exclusively in the pancreas and vascular endothelium but are also expressed in the brain and specifically in the hippocampus, hypothalamus, cerebral cortex, and olfactory bulbs
[42]. The role of incretins and incretin analogues in the brain is neuroprotective
[43], as they enhance cell proliferation, memory, and synaptic plasticity, while reducing β-amyloid plaques, oxidative stress, and inflammation
[44][45][46].
Long-acting liraglutide has been shown to normalize the distribution of cell membrane insulin receptors in a rat model with AD (APPSWE/PS1dE9), thus improving insulin signaling disorders
[47]. In addition, systematic administration of liraglutide in transgenic mice with AD for 8 weeks prevented the underlying neurodegenerative effects observed in AD, such as neuronal loss, memory impairment, and a decrease in synaptic plasticity in the hippocampal region
[48]. In particular, liraglutide reduced the deposition of β-amyloid plaques by 40–50%, while a decrease was also observed in the inflammatory response based on activated glial cells
[48]. In mice that received intrahippocampal injections of β-amyloid, it was observed that pretreatment with liraglutide before injection was a protective factor against impairments in spatial memory and long-term potentiation (LTP) induced by β-amyloid
[49]. Additional experiments in transgenic mice have shown that liraglutide promotes neurogenesis, has a positive effect on the cerebral microvascular system, and also reduces tau protein hyperphosphorylation in AD
[50][51][52][53]. It also appears that liraglutide has not only preventive properties but also the ability to reverse several of the key pathological features that appear in the final phase of AD in mice models
[54]. Positive results have also been observed in the rat model APPswe/PS1ΔE9 with AD, where the long-term administration of the analogue hormone liraglutide GIP (D-Ala2GIP) protects synaptic plasticity and memory formation and reduces β-amyloid plaques and neuroinflammation, while normalizing stem cell proliferation
[45].
Dipeptidyl-peptidase 4 (DPP4) enzyme inhibitors are also used as an alternative treatment. They can extend the action time of GLP-1 and GIP, thus regulating glucose in T2DM
[55]. A study by Kornelius et al.
[56], found that linagliptin (a DPP4 inhibitor) can restore the impaired insulin signaling induced by β-amyloid in neuronal cells, indicating the important therapeutic role that DPP4 inhibitors may play in the neurotoxicity of AD. Two other DPP4 inhibitors, saxagliptin and vildagliptin, showed similar efficacy when given orally to AD transgenic mice, resulting in reduced β-amyloid deposition, improved memory, and increased levels of hippocampal GLP-1, as well as reduced tau protein phosphorylation and markers of inflammation
[57][58]. An alternative substance is exentin-4, a long-acting incretin GLP-1 receptor agonist, which has a neuroprotective effect in neurodegenerative diseases such as AD and Parkinson’s disease and is fully approved for use in patients with T2DM
[44][59]. In an in vitro study by Bomfim et al.
[60], the property of β-amyloid oligomers to attenuate axial transport was inhibited by the administration of exentin-4 (GLP-1R agonist), which appeared to improve cognitive ability by reducing the serine phosphorylation of the insulin receptor substrate (IRS-1) in the hippocampus. The only human study of liraglutide in AD patients
[61], showed that a 6-month treatment had moderate neuroprotective effects, mainly expressed by improvements in cerebral glucose metabolism. In the same study, liraglutide administration had no effect on the β-amyloid deposition of AD patients when compared to placebo patients.
Additional research is needed to clarify the role of incretins in the treatment of AD in humans. Despite promising evidence from animal experiments, existing studies have failed to demonstrate reversal of AD pathology in humans. More studies are necessary to determine the exact action of incretins at each individual stage of AD, in order to define the therapeutic window for these drugs.