Alzheimer’s disease (AD) is a growing concern among communities nowadays. In the United States, there are more than five million Americans who are living with AD, with the majority of people 65 years old and older ^[1]. The Alzheimer’s Association Report estimated that the number of people affected by AD in the United States will be increased up to fourteen million by 2060 ^[1]. The disease, which is the most common cause of dementia, is a progressive and irreversible disorder of the brain that slowly deteriorates the brain function of an individual ^[2]. It progresses from preclinical, early- to moderate-stage, and finally late-stage disease. The early symptoms include mainly cognitive impairment, in particular memory loss. As cognitive function deteriorates, presentation of physical disabilities, such as the inability to walk, sit and eat indicates that the disease has progressed to the later stages ^[3]. Intracellular neurofibrillary tangles and extracellular amyloid β plaques are the hallmark characteristics found in the cortical and limbic areas of the brain that are associated with AD ^[4].

Generally, current treatment for the AD can be classified into two main categories based on the stages of the disease. For mild to moderate cases, galantamine, rivastigmine and donepezil as acetylcholinesterase inhibitors are indicated to provide temporary symptomatic relief among the patients ^[4]. Memantine, an N-methyl D-aspartate (NMDA) antagonist is used as a monotherapy to manage the symptoms in moderate to severe AD ^[4]. These drugs are mostly selective compounds that target individual proteins (“one compound–one target” approach), and are mostly aimed at restoring physiological acetylcholine levels. Nonetheless, various mechanisms of AD pathogenesis have been proposed to date, which are shown to overlap and influence one another ^[5]. This complexity challenges the dominant single-target approach in the treatment of AD. In fact, it has been widely recognized that the conventional single-target approach may not be adequately effective against AD that has a multifactorial origin involving a combination of genetic, metabolic, and environmental factors [5,6,7,8]^[5][6][7][8].

As a result, multi-target strategies have increasingly been considered as alternative options for the management of multifactorial AD in the past decades [9,10]^[9][10]. Amongst them, combination therapies based on a “cocktail drug–multiple targets” approach combining several drugs acting independently on different targets have been adopted to alleviate the symptoms of AD, such as a drug combination consisting of memantine and donepezil that has been used clinically to manage the symptoms in moderate to severe AD ^[4]. These drugs may act on the targets of different or the same pathways that are involved in the pathogenesis of AD ^[9]. Nevertheless, they are often associated with side effects due to drug–drug interactions, for instances bradycardia, atrioventricular block and psychosis ^[11] as well as varying pharmacokinetic profiles of each component drug ^[12].

2. Pathogenesis of Alzheimer’s Disease

As AD is a complex and multifactorial disease, a clear understanding of the underlying pathogenesis of the disease is essential for developing effective therapeutic regimens. Most of the early onset, autosomal dominant AD cases are characterized by the presence of extracellular beta amyloid (Aβ) plaques in various regions of the AD patient’s brain, due to either overproduction or reduced clearance of Aβ peptides or both [14]^[13]. Studies have found that amyloid precursor protein (APP) is associated with the pathogenesis of AD. In the brain, a group of enzymes known as APP secretases, including α-secretases, β-secretases and γ-secretases work together to process the APP [14]^[13]. In the physiological pathway, the α-secretases process the APP and produce soluble APPα (sAPP-α), which can preclude the subsequent β-γ secretase activity [14,15]^[13][14]. Evidence has shown that the soluble APPα is neuroprotective; it allows proper synaptic signaling and maintains neuronal plasticity [16]^[15]. On the other hand, in the amyloidogenic pathway, the β-secretases, or BACE-1, yield soluble APPβ (sAPP-β) and a small carboxy (C)-terminal fragment (CTFβ); both are cut by the γ-secretases into insoluble and neurotoxic Aβ peptides ^[9]. The two dominant forms of Aβ peptides produced in AD are Aβ40 and Aβ42 with 40 and 42 amino acid residues, respectively [14]^[13]. These Aβ peptides tend to aggregate to form oligomers of Aβ (oAβ), which will further aggregate into insoluble amyloid plaques or senile plaques ^[5]. In the case of sporadic or late onset AD, expression of the apolipoprotein E4 (APOE₄) gene has been found to be a factor that contributes to the pathogenesis. Studies have reported that individuals with expression of APOE₄ have an increase in beta-amyloid deposition together with impaired memory ^[5]. Mitochondrial dysfunction is secondary to the primary pathologic event of AD, the production of Aβ [17]^[16]. The Aβ plaques are largely found in the mitochondria of neuronal cells in AD patients; it can modify the structure of mitochondria and block the ion channels, thus interrupting calcium homeostasis [5,16]^[5][15]. This results in decreased mitochondrial respiration and ATP synthesis ^[5]. Events such as elevated mitochondrial fission and diminished mitochondrial fusion are also observed following exposure to Aβ plaques ^[5]. Such senile plaques are also found to cause an increase in oxidative stress due to intracellular formation of reactive oxygen species (ROS) from the mitochondria, which leads to a low energy metabolism rate, and eventually neuronal cell apoptosis with release of cytochrome c [5,16]^[5][15]. As opposed to the Aβ plaques, neurofibrillary tangles (NFTs) are another hallmark of the disease that is detected intracellularly in the brain of AD patients. This phenomenon can be illustrated by the Tau hypothesis. Tau (τ), a microtubule-associated protein, has a role in stabilizing the microtubules [16]^[15]. However, the Aβ42 that accumulates to high levels in the brain, increases the risk of hyperphosphorylation of the tau protein [18]^[17], which is regulated by several kinases, such as glycogen synthase kinase 3 (GSK-3β) and cyclin-dependent kinase 5 (CDK5) [16,19]^[15][18]. As such, the structure of the microtubule is disrupted and becomes unstable as the subunits are dissociated from itself without the support of the tau protein. The phosphorylated tau proteins clump together and form straight, insoluble, and fibrillary tau filaments; they are then aggregated into deposits called NFTs in the cytoplasm, which are neurotoxic. This leads to synaptic loss and affects the signaling process between neurons. As a result, apoptosis of the neuronal cells ensues [16]^[15]. The formation of both neurotoxic Aβ plaques and NFTs can increase oxidative stress and provoke synaptic damage. This attracts microglia to the vicinity of the plaques which act as resident phagocytes for clearance of both Aβ and the NFTs. The activated microglia following the binding of Aβ and the NFTs to its cell-surface receptors induce production of pro-inflammatory cytokines and other mediators for phagocytosis [20,21]^[19][20]. However, elevated cytokine levels under chronic inflammation leads to downregulation of the phagocytic receptor expression on microglia, resulting in an ineffective Aβ clearance [16]^[15]. The Aβ-induced microglia are also found to generate ROS that can cause further oxidative damage to the neuronal cells. Consequently, this has led to a continuous cycle of microglia-mediated neuroinflammation and neuronal cell death [22]^[21]. The pathogenesis of AD is also suggested to be linked to the NMDA receptor. Essentially, binding of the glutamate excitatory neurotransmitter to the NMDA receptors allows regulation of synaptic plasticity and provides normal learning and memory functions through a process called long-term potentiation (LTP) [23]^[22]. In AD, an excitotoxicity hypothesis is proposed, in which the NMDA receptors are overactivated by the Aβ plaques; and this contributes to excessive Ca²⁺ fluxes, leading to excitotoxicity and impairment of the mitochondrial energy metabolism. Thus, free ROS production is favored, subsequently causing an increase in oxidative stress and altered synapse function [24]^[23]. Apart from this, Kochahan and co-workers also reported that Aβ oligomers induced spine loss and have resulted in the reduction of the amount of glutamate receptors available for binding, hence inhibiting the LTP at the hippocampus and other regions of the brain. Without proper excitatory transmission through NMDA receptors and normal synaptic function, it promotes further progression of AD. Therefore, the NMDA receptors have been regarded as potential targets of interest for tackling the cognitive impairment occurring in AD [23]^[22]. Additionally, a cholinergic hypothesis has been postulated for the pathogenesis of end-stage AD [25,26]^[24][25]. Based on the hypothesis, AD may be due to the loss of central cholinergic neurons that leads to deficiency of a neurotransmitter responsible for memory and learning, known as acetylcholine (ACh). Research also showed that the AD brain has notably diminished activity of choline acetyltransferase (ChAT) involved in acetylcholine synthesis and reduced metabolism of the acetylcholinesterase (AChE) [25]^[24]. Nonetheless, cholinergic depletion is not the only factor that causes the decline in cognitive functions [26]^[25]. Instead, aging is also a factor that causes natural loss of ACh and impairs the ability of cholinergic neurons to release ACh for neurotransmission; this increases the susceptibility of hippocampus to damages from other central nervous system complications, such as stress, seizure, or stroke. Ultimately, this has brought about the memory and cognitive deficits in AD [26,27]^[25][26]. Aside from the cholinergic pathways, AChE is also found to be associated with the non-cholinergic function via AChE-induced Aβ aggregation that can eventually lead to neurotoxicity [28]^[27]. In general, multiple hypotheses have been associated with the pathogenesis of AD. Researchers have put in much effort to investigate the mechanisms involved in the AD pathogenesis, which have helped to accelerate the discovery of potential therapeutic agents for the management of AD.

3. Current Drug Therapies for Alzheimer’s Disease

Research conducted on AD thus far has improved the knowledge on the pathophysiology of the disease. Nevertheless, there are only a few medications approved by the Food and Drug Administration (FDA) to manage the disease. These medications are mainly used to improve the symptoms of AD, such as cognitive and global functioning; they are unable to delay the progression or treat the underlying causes of AD [29,30,31]^[28][29][30]. Currently, there are five main pharmacotherapies for AD based on two drug classes, namely AChE inhibitors (rivastigmine, donepezil, galantamine) and NMDA receptor antagonists (memantine), as well as a combination therapy of an acetylcholinesterase inhibitor with memantine. Donepezil (1), considered as the first line treatment for AD, is a second generation of AChE inhibitor along with rivastigmine and galantamine. It is a highly selective, reversible and non-competitive AChE inhibitor, which is slowly absorbed from the gastrointestinal tract and has a relative long half-life (50 to 70 h) [32,33]^[31][32]. Basically, it acts by increasing the concentration of acetylcholine in the synaptic cleft of the hippocampus through inhibition of AChE and causes stimulation of brainstem reticular formation that leads to an increase in hippocampal theta rhythm amplitude [33]^[32]. In China, donepezil was approved for use in mild to moderate AD in 2006 and severe AD in 2017 [33]^[32]. A study that involved 603 patients, by Black et al., concluded that there was significant improvement on the Alzheimer’s Disease Assessment Scale-Cognitive Subscale (ADAS-cog) scores at all time-points [34]^[33]. It was further supported by another 12-week, randomized, multinational study done by Wilkinson et al., which involved 111 patients with mild to moderate AD. The study consistently showed that there was comparable improvement on the ADAS-cog scores. Generally, the use of donepezil was well tolerated in patients with mild to moderate AD due to lesser treatment-emergent adverse effects when compared with the use of rivastigmine [35]^[34]. The recommended dose of donepezil for mild and moderate AD is 5 mg once daily and it may be increased up to 10 mg/day after four to six weeks. In contrast, the recommended dose of donepezil for moderate to severe AD is 10 mg or 23 mg once daily. A randomized, controlled trial comparing the benefits of treating moderate to severe AD using 10 mg and 23 mg daily doses of donepezil concluded that a 23 mg daily dosing of donepezil showed better cognitive benefits in treating moderate to severe AD [36]^[35]. Common adverse effects, such as nausea, diarrhea, agitation and dizziness associated with donepezil are generally of mild to moderate intensity and can be resolved without the need for discontinuation of medication [34]^[33]. Galantamine (2), a medication with dual mechanisms of action, is a rapidly reversible acetylcholinesterase inhibitor and a positive allosteric modulator of nicotinic receptors [37]^[36]. A study by Wallin et al. was previously conducted to evaluate the long-term effect of galantamine treatment in 280 AD patients [38]^[37]. The study showed marked improvement in terms of cognitive assessment (based on the Mini-Mental State Exam (MMSE) and ADAS-cog scores, with a mean change from baseline of 2.6 points and 5.6 points, respectively, upon three years of galantamine treatment). This was significantly better than the predicted annual decline in scores for untreated patients in both parameters (2 to 4 points in MMSE score and 6.7 points in ADAS-cog score) [38]^[37]. It is supported by another study by Thavichachart et al., in which two-thirds of patients (67.8%) reported improvement in ADAS-cog score, especially in those with mild and moderate severity of AD [37]^[36]. Slow dose escalation was well tolerated and fewer adverse effects were reported when patients received 16 mg/day of galantamine versus 24 mg/day of galantamine [37]^[36]. The recommended dose of galantamine for mild AD is 16 mg/day, while 24 mg/day is more beneficial for the patients with moderate AD [39]^[38]. However, several adverse effects are encountered by patients during the galantamine treatment, including nausea (12%), weight loss (11%), dizziness (7%) and vomiting (5%). Fortunately, those were mild- to moderate- intensity adverse events that showed no clinical changes from baseline in all aspects [37]^[36]. Rivastigmine (3) (Figure 2) is another acetylcholinesterase inhibitor that is used for the management of AD. Unlike donepezil and galantamine which selectively inhibit the AChE, rivastigmine acts by inhibiting both the AChE and butyrylcholinesterase (BuChE) in the brain [39,40]^[38][39]. It has low protein binding, and hence shows minimal potential interaction with other drugs; this makes it a more suitable medication for those elderly who take many different medications concurrently [40,41]^[39][40]. A study conducted by Rösler et al. concluded that a high dose of rivastigmine (6–12 mg/day) showed marked improvement in terms of ADAS-cog, global function and progressive deterioration scale (PDS) scores in the AD patients when compared to the placebo group. More patients (24%) in the higher dose group had improved by 4 points or more in ADAS-cog score than those patients in the placebo group (16%). The same was observed with the global function (37% versus 20%) and PDS (29% versus 19%) whereby more patients in the higher dose group had significant improvement than those in the placebo group [42]^[41]. Another study by Karaman et al. stated that long term rivastigmine treatment was well tolerated and improved the cognitive and functional symptoms (such as non-epileptic attacks and weakness) in AD patients [43]^[42]. Based on the ADAS-cog score, only 18.3% of AD patients encountered a reduction of 4 or more points when they were on rivastigmine treatment, in contrast to 45% of placebo-treated patients who experienced a reduction of at least 4 points on the ADAS-cog scale [43]^[42]. The MMSE also showed that the AD patients treated with rivastigmine had a better score than those receiving placebo. There was improvement of 0.20 points from baseline in those patients treated with rivastigmine compared to those receiving placebo with deterioration of 1.2 points from baseline [43]^[42]. The recommended dose of rivastigmine in mild to moderate AD patients ranges from 6 to 12 mg/day orally in two separate doses. The patients should start at 1.5 mg BD and increase the dose in 1.5 mg increments as tolerated [40]^[39]. Generally, the adverse effects are not severe and can be resolved upon slower dose escalation. The most common adverse effects reported are nausea (16.6%), vomiting (12.5%), dizziness, anorexia and headache (8.3%) in those AD patients treated with rivastigmine [43]^[42]. Besides the AChE inhibitors, memantine (4) is another medicine approved to manage the symptoms of AD. It is a voltage-dependent and non-competitive NMDA receptor antagonist, which selectively binds to NMDA receptor-operated calcium channels [44,45,46]^[43][44][45]. Under normal conditions, activation of the synaptic NMDA receptors induces plasticity and enhances the survival of neuronal cells [47,48,49]^[46][47][48]. However, excessive activity of the NMDA receptor is detrimental as it can cause excitotoxicity of the neurons [48,49,50]^[47][48][49]. When this occurs, the neuronal cells will undergo apoptosis, leading to neuronal dysfunction. Memantine inhibits the effects of overactivated NMDA receptors, thus reducing the apoptosis of neuronal cells and preventing neuronal damage. It has been shown to improve the cognitive functions of AD patients in all stages. A study conducted in 2011 by Schulz et al. concluded that a 20 mg once daily regimen of memantine significantly improved the cognition and functional communication in AD patients [51]^[50]. It is usually prescribed in patients with moderate to severe AD, or in mild to moderate AD patients who cannot tolerate acetylcholinesterase inhibitors [52]^[51]. The initial dose of memantine is 5 mg daily, followed by steady weekly increments of 5 mg, and a maximum dose of 20 mg daily. It is better tolerated than the acetylcholinesterase inhibitors, although cases of dizziness, headache, somnolence, constipation, and hypertension have been reported as side effects [52]^[51]. On top of the monotherapy with either AChE inhibitors or NMDA receptor antagonists, a combination therapy consisting of an AChE inhibitor and memantine is another treatment option for AD. In 2014, a fixed dose combination (FDC) of donepezil and memantine was approved as a pharmacological management option for moderate to severe AD [53]^[52]. Concurrent administration of an AChE inhibitor with memantine is believed to have synergistic effects in alleviating the symptoms of AD due to the complementary mechanism of actions. This combination has been shown to be effective in patients with an advanced stage of AD. A study by Tariot et al. suggested that addition of memantine in moderate to severe AD patients receiving stable doses of donepezil is beneficial [54]^[53]. The results favored the group of subjects receiving both donepezil and memantine at the same time. Atri et al. also analyzed the cumulative additive benefits of the memantine–donepezil combination over component monotherapies in moderate to severe Alzheimer’s dementia using area-under-curve (AUC) analysis [55]^[54]. The study found that the AUC of subjects receiving the donepezil–memantine combination had significant improvement compared to those receiving the monotherapy, thus indicating the additive effect of the combination therapy. Currently, such a drug combination is available in fixed doses of donepezil–memantine capsules. The recommended dose of donepezil–memantine FDC capsules is 28 mg memantine and 10 mg donepezil daily for both the moderate and severe AD patients [55]^[54]. A study has demonstrated that these capsules are bioequivalent to the concurrent administration of each individual component [56]^[55]. Furthermore, a fixed-dose capsule of donepezil and memantine may also enhance the adherence of patients to the treatment regimen.