Personalized Management and Treatment of Alzheimer’s Disease

Alzheimer’s disease (AD) is a priority health problem with a high cost to society and a large consumption of medical and social resources. The management of AD patients is complex and multidisciplinary. Over 90% of patients suffer from concomitant diseases and require personalized therapeutic regimens to reduce adverse drug reactions (ADRs), drug–drug interactions (DDIs), and unnecessary costs. Men and women show substantial differences in their AD-related phenotypes. Genomic, epigenetic, neuroimaging, and biochemical biomarkers are useful for predictive and differential diagnosis. The most frequent concomitant diseases include hypertension (>25%), obesity (>70%), diabetes mellitus type 2 (>25%), hypercholesterolemia (40%), hypertriglyceridemia (20%), metabolic syndrome (20%), hepatobiliary disorder (15%), endocrine/metabolic disorders (>20%), cardiovascular disorder (40%), cerebrovascular disorder (60–90%), neuropsychiatric disorders (60–90%), and cancer (10%). Over 90% of AD patients require multifactorial treatments with risk of ADRs and DDIs. The implementation of pharmacogenetics in clinical practice can help optimize the limited therapeutic resources available to treat AD and personalize the use of anti-dementia drugs, in combination with other medications, for the treatment of concomitant disorders.


Introduction
Alzheimer's disease (AD) is a health priority in developed societies and in developing countries, along with cardiovascular disease, cancer, stroke, and major neuropsychiatric pathologies. Direct and indirect costs for the management of AD represent an overload on the economy of families, nations, and health resources. The worldwide cost of dementia exceeds US$800 billion dollars (>1% of GDP). The average cost per patient/year ranges from $30,000 to $60,000, depending on stage of the disease, quality of medical care, social
In terms of anthropometric parameters, substantial differences are observed between females and males in terms of height, weight, and body-mass index (BMI) ( Table 1). Only 28% of patients show normal weight (29% F; 20% M), with obesity in >25% of the cases and only 1% with underweight (Table 1).
Cognitive markers (MMSE, ADAS) indicate than females show worse cognitive performance than males (Table 1). Cognitive impairment and depression are the first symptoms, which appear in over 90% of LOAD, and 80% and 9%, respectively, in EOAD cases [25].
Late-life depression is associated with cognitive impairment, and depression is linked with an increased risk for AD. Some overlapping pathogenic substrates (i.e., stress, cortisol levels, brain hypoperfusion, neuroinflammation, neurotrophic dysfunction, Aβ accumulation, tauopathic connections, epigenetic factors, and gut microbiota-brain axis) may explain the comorbidity of both clinical entities [26]. Mood disorders are more frequent in women than in men. Over 60% of AD patients show depressive symptoms, which are more severe in women than in men. Likewise, anxiety is also more frequent in females than in males. About 50% of men do not show anxiety, whereas only 30% of women with dementia are free of symptoms in early stages of the disease. Both anxiety and depression fluctuate with the clinical course of the disease [24,[27][28][29][30]. Behavioral disorders and psychotic symptoms are also frequent (20-90%) in patients with AD along the clinical course of the disease [24,31,32].

Biomarkers
In addition to conventional clinical markers, which allow us to make a differential diagnosis and assess the possibility of concomitant diseases, the most useful biomarkers for a predictive diagnosis or diagnostic confirmation of antemortem AD are genomic markers, epigenetic biomarkers, neurotransmitters, and levels of Aβ/tau in the brain (PET Scan) and/or in body fluids [37][38][39][40][41][42][43][44][45][46].

Biomarkers
In addition to conventional clinical markers, which allow us to make a differential diagnosis and assess the possibility of concomitant diseases, the most useful biomarkers for a predictive diagnosis or diagnostic confirmation of antemortem AD are genomic markers, epigenetic biomarkers, neurotransmitters, and levels of Aβ/tau in the brain (PET Scan) and/or in body fluids [37][38][39][40][41][42][43][44][45][46].
APP mutations can cause EOAD, with increased Aβ levels or Aβ fibrillogenesis, while some coding variants (APP A673T) may be protective with reduced Aβ levels [51]. EOAD dominant mutations tend to occur in the APP coding region or in the presenilin-related catalytic site of γ-secretase whose protease dysfunction is responsible for the abnormal process of APP cleavage and consequent accumulation of Aβ in senile plaques and vessels. In >40% AD cases, the presence of the apolipoprotein E4 (APOE-4) allele is the most important risk factor, involved in the impairment of Aβ clearance from brain tissue, in atherosclerosis and in hypoperfusion. Most immunotherapeutic interventions with Aβ antibodies (aducanumab, solanezumab, and crenezumab) attempt to halt the amyloidogenic process and slow-down cognitive deterioration in mild-AD or in presymptomatic cases with demonstrated genomic dysfunction [52]. In addition to these primary pathogenic genes, many other genes have been associated with AD in next-generation sequencing (NGS) and genome-wide association studies (GWAS) in different populations [47,48,[54][55][56][57][58].

Epigenetic Markers
Various epigenetic aberrations are associated with AD pathogenesis, including hypomethylation/hypermethylation in the promoters of pathogenic genes, alterations in histones, and changes in the linear and three-dimensional structure of nuclear chromatin, as well as profound alterations in microRNAs (miRNAs) that regulate gene expression in the cellular cytoplasm. Some of these epigenetic alterations have been proposed as potential biomarkers of AD [72][73][74][75][76][77][78][79].
The main problems observed with the use of epigenetic biomarkers in AD are their variability and lack of specificity. Changes in global DNA methylation are very sensitive and appear diminished in a multitude of central nervous system (CNS) diseases, such as AD (p < 0.001), Parkinson's disease (p < 0.004), cerebrovascular disorders and stroke (p < 0.01), major depression (p < 0.05), migraine (p < 0.03), epilepsy (p < 0.05), and intellectual organic disability (OID) (p < 0.001), and to a lesser extent in schizophrenia (Figure 4, upper  panel). These values are very sensitive to therapeutic interventions but unreliable as predictive or diagnostic values. The low diagnostic value of DNA methylation is compensated for by the exquisite sensitivity of this biomarker that responds in a highly sensitive The pathogenic load of APOE-4 affects 35-40% of cases, with significant phenotypic consequences (Tables 2-4). APOE-4/4 carriers tend to show an earlier age-at-onset in >80% of the cases; lower peripheral ApoE, nitric oxide, histamine, Aβ, HDL-cholesterol, and triglyceride levels; higher levels of total cholesterol and LDL-cholesterol; more pronounced brain atrophy and slower brain bioelectrical activity; more severe brain hemodynamic dysfunction represented by hypoperfusion, reduced brain blood flow velocity and increased pulsatility and resistance indices; increased lymphocyte apoptosis; faster cognitive deterioration; more frequent metabolic disorders, cardiovascular disorders, hypertension, atherosclerosis, liver metabolism dysfunction, behavioral disturbances, and alterations in circadian rhythm patterns; and a poor response to conventional treatments [5,[11][12][13]22,48,[59][60][61][62][63][64][65][66][67][68][69][70][71].

Epigenetic Markers
Various epigenetic aberrations are associated with AD pathogenesis, including hypomethylation/hypermethylation in the promoters of pathogenic genes, alterations in histones, and changes in the linear and three-dimensional structure of nuclear chromatin, as well as profound alterations in microRNAs (miRNAs) that regulate gene expression in the cellular cytoplasm. Some of these epigenetic alterations have been proposed as potential biomarkers of AD [72][73][74][75][76][77][78][79].
The main problems observed with the use of epigenetic biomarkers in AD are their variability and lack of specificity. Changes in global DNA methylation are very sensitive and appear diminished in a multitude of central nervous system (CNS) diseases, such as AD (p < 0.001), Parkinson's disease (p < 0.004), cerebrovascular disorders and stroke (p < 0.01), major depression (p < 0.05), migraine (p < 0.03), epilepsy (p < 0.05), and intellectual organic disability (OID) (p < 0.001), and to a lesser extent in schizophrenia (Figure 4, upper panel). These values are very sensitive to therapeutic interventions but unreliable as predictive or diagnostic values. The low diagnostic value of DNA methylation is compensated for by the exquisite sensitivity of this biomarker that responds in a highly sensitive manner to the therapeutic response of each patient. In addition, what appears to be important in AD is that global DNA methylation shows an APOE-dependent pattern. APOE-4 carriers tend to show a more severe DNA hypomethylation pattern than patients carrying the APOE-3 allele, which is aggravated in parallel with the degree of cognitive impairment (Figure 4, lower panel).
Life 2022, 12, x FOR PEER REVIEW 18 of 41 manner to the therapeutic response of each patient. In addition, what appears to be important in AD is that global DNA methylation shows an APOE-dependent pattern. APOE-4 carriers tend to show a more severe DNA hypomethylation pattern than patients carrying the APOE-3 allele, which is aggravated in parallel with the degree of cognitive impairment ( Figure 4, lower panel).  Epigenetic biomarkers can also help in the personalization of anti-AD treatments (Pharmacoepigenetics) [80][81][82][83] and serve as a guide in the search for epigenetic drugs with prophylactic and/or therapeutic action in the treatment of AD [72,[81][82][83].

Neurotransmitters
Conventionally, the main neurotransmitter affected in AD is acetylcholine. However, premature neuronal death alters the levels of many other essential neurotransmitters for the normal functioning of the CNS [84][85][86].
Deficits in noradrenaline, dopamine, serotonin, histamine, GABA, glutamate, and various neuropeptides (GRF, CRF, somatostatin, and vasopressin) are particularly relevant, the alteration of which can lead to AD-related neuropsychiatric disorders. However, none of these biomarkers are sufficiently sensitive or specific to AD ( Figure 5), although their quantification in CSF or blood is useful for monitoring brain damage and/or the efficacy or ineffectiveness of the treatments the patient receives.
Epigenetic biomarkers can also help in the personalization of anti-AD treatments (Pharmacoepigenetics) [80][81][82][83] and serve as a guide in the search for epigenetic drugs with prophylactic and/or therapeutic action in the treatment of AD [72,[81][82][83].

Neurotransmitters
Conventionally, the main neurotransmitter affected in AD is acetylcholine. However, premature neuronal death alters the levels of many other essential neurotransmitters for the normal functioning of the CNS [84][85][86].
Deficits in noradrenaline, dopamine, serotonin, histamine, GABA, glutamate, and various neuropeptides (GRF, CRF, somatostatin, and vasopressin) are particularly relevant, the alteration of which can lead to AD-related neuropsychiatric disorders. However, none of these biomarkers are sufficiently sensitive or specific to AD ( Figure 5), although their quantification in CSF or blood is useful for monitoring brain damage and/or the efficacy or ineffectiveness of the treatments the patient receives.  Figure 5, upper panel). In contrast, serotonin levels tend to decrease in anxiety (p < 0.02), AD (p < 0.05), depression (p < 0.05), and ALS (p < 0.05), and show high levels in ataxic syndromes (p < 0.002), OID (p < 0.05), and in xenoestrogenic syndrome (p < 0.001), a novel clinical entity present in women with chronic use of contraceptives or hormone replacement therapy (HRT) ( Figure 5, lower panel).

Aβ/Tau Levels
The most popular biomarkers for AD in body fluids (CSF, plasma) are the quantification of amyloid-β (Aβ42), total tau (T-tau), and phosphorylated tau (P-tau) in the CSF (AD CSF profile: decreased Aβ42 levels together with increased T-tau and P-tau levels). Other CSF markers (synaptotagmin, rab3a, SNAP-25, and neurogranin) are also altered [40,41]. However, the heterogeneity of AD and the inconvenience of having to perform a lumbar puncture to obtain CSF do not allow these biomarkers to reach high quotas of sensitivity and specificity, nor generalized use.

Concomitant Disorders and Phenotype-Modifying Treatments
Most patients with dementia (>90%) require multifactorial treatments for the management of concomitant disorders and/or neuropsychiatric symptoms associated with dementia. The chronic administration of drugs from different categories increases the risk of ADRs and DDIs [17,[87][88][89].
Cardiovascular risk factors (hypertension, hypercholesterolemia, and dyslipidemia) and ECG abnormalities are more frequent in males than in females. Hypertension is present in 21% of the cases. Systolic blood pressure is similar in females and males, but diastolic blood pressure is significantly higher in males than in females (p < 0.001) ( Table 1). Cholesterol levels (Total, LDL) are higher in men and HDL-cholesterol and triglyceride levels are more elevated in females (Table 1). Brain damage and increased cognitive deterioration are currently associated with cardiovascular disorders and blood pressure changes in AD [90,91]. APOE-4 carriers with dementia also exhibit cardiovascular disorders, atherosclerosis, and cerebrovascular damage [5,12,[43][44][45][60][61][62][63][64][65][66]92]. Lipid metabolism disorders contribute to the cerebrovascular component of AD. Abnormalities in cholesterol metabolism and dysfunction of lipid rafts in cell membranes and arteriosclerosis are pathogenically relevant for cerebral ischemia and hypoperfusion, and accelerate premature neuronal death in patients who are predisposed to AD [44,[93][94][95][96]. In contrast, the epidemiological link between diabetes and AD appears to be circumstantial, with no apparent pathogenic implications beyond the deleterious effects of hyperglycemia on brain function [97][98][99].
As a consequence of all these concomitant pathologies, patients with dementia consume a wide variety of drugs whose side-effects contribute to accelerating the degenerative process and cognitive decline. Of special importance are cardiovascular agents, statins, antidiabetics, antihypertensive drugs, analgesics, diuretics, bronchodilators, antirheumatics, and various categories of psychotropic drugs (neuroleptics, antidepressants, anxiolytics, hypnotics, and sedatives). The correct administration of these drugs requires a personalized therapeutic intervention, together with conventional anti-dementia treatments [15,17,100].
Combination treatments applied under pharmacogenetic guidance indicate that alterations in biochemical, hematological, and metabolic parameters affect drug efficacy and safety. Concerning cognitive function and neuropsychiatric disorders treated with multifactorial regimes, females and males respond differentially to treatment, showing a moderate improvement in cognition during the first year of treatment and significant improvements in anxiety and depression.

Alzheimer's Disease Therapeutics and Drug Development
For the past 50 years, the major focus of pharmacological development in AD has been cognitive enhancers. The introduction of acetylcholinesterase inhibitors (AChEIs) in the early 1990s was the first option to restore cholinergic neurotransmission after the identification of a selective cholinergic deficit in the basal forebrain and neuronal loss in neocortex and hippocampus. Tacrine (9-amino-1,2,3,4-tetrahydroacridine) was the first AChEI introduced in 1993 for the treatment of AD. The Chinese product Huperzine A was approved in 1994.
A new generation of AChEIs (donepezil, galantamine, and rivastigmine) was introduced years later. In 2003, the FDA approved memantine, an N-methyl-D-aspartate (NMDA) glutamate receptor partial inhibitor. Since then, no new FDA-approved drugs for AD were reported until the recent approval of the antibody aducanumab in 2021 [100,101].

Immunotherapy
Since the early 2000s, several categories of vaccines and immunotherapeutic procedures have been developed for the treatment of AD, following the pioneering studies of Schenk and coworkers in 1999 [105]. For two decades, millions of dollars have been invested in passive and active immunotherapy in experimental AD models and in clinical trials. About 1000 papers on AD immunotherapy were published (85% anti-Aβ and 15% anti-tau) prior to the FDA approval of aducanumab as an immunotherapeutic strategy in mild-AD [103,106].
Active immunization studies demonstrate that presymptomatic immunization of PDAPP transgenic mice, overexpressing human mutant APP (Phe717Val), prevents Aβplaque formation, astrogliosis, and neuritic dystrophy. Immunization with Aβ42 reduces the extent and progression of AD-related neurodegeneration and improves cognition in the TgCRND8 murine model of AD [105][106][107][108]. Over 400 experimental studies with different immunization procedures were able to replicate these results. However, clinical studies did not show the same efficacy and initial clinical trials with the Aβ vaccine AN1792 had to be discontinued. This was because of severe undesirable effects, such as acute meningoencephalitis or microhemorrhagic lesions in the brain of some immunized patients. Some of these ADRs were attributed to a T-cell-mediated pro-inflammatory response; other mechanistic causes of brain damage in patients treated with active vaccines remain elusive. Improvements in immunization processes and vaccine preparations in recent years have made it possible to obviate many of these drawbacks and avoid adverse effects with the second-generation Aβ-active immunotherapies, anti-tau immunotherapies, and anti-Aβ monoclonal antibodies targeting Aβ epitopes [109][110][111][112][113][114]. Other modalities of immunization, such as dual vaccines (EB101) [111], Aβ3-10-KLH vaccine [112], or active full-length DNA-Aβ42 trimer immunization [113], also reduce amyloidogenic and tauopathic markers in transgenic animals [114].

The Pharmacogenomic Machinery in Alzheimer's Disease
The pharmacogenetic apparatus is made up of a genomic-epigenetic machinery composed of networks of genes that encode proteins, enzymes, and messengers involved in the therapeutic and/or toxic effects of each drug under the regulatory control of the epigenome [81].
The main pharmagenes encoding byproducts of the pharmacogenetic machinery involved in drug efficacy and safety can be classified into five major categories: pathogenic (Tables 2-4), mechanistic, metabolic, transporter, and pleiotropic genes [11,12,81]. Over 400 genes may influence drug pharmacogenetics; 240 genes have been associated with ADRs, and various rare variants have been identified in nearly 150 pharmagenes with clinical relevance [122,123].
Defective pharmagenes accumulate in approximately 85% of AD patients ( Figure 6). The most frequent pharmagenes with dysfunctional effects in AD are the follows: CES1 Pharmagenes and their products are redundantly and promiscuously regulated by the epigenome (DNA methylation/demethylation, histone/chromatin remodeling, and miRNA regulation), thereby configuring the pharmacoepigenetic apparatus [81,124]. The same drug can affect the function of many different genes, and the same gene can influence the absorption, distribution, metabolism, and elimination (ADME) of various drugs [124]. Pharmagenes and their products are redundantly and promiscuously regulated by the epigenome (DNA methylation/demethylation, histone/chromatin remodeling, and miRNA regulation), thereby configuring the pharmacoepigenetic apparatus [81,124]. The same drug can affect the function of many different genes, and the same gene can influence the absorption, distribution, metabolism, and elimination (ADME) of various drugs [124].

Mechanistic Genes Involved in Cholinergic Neurotransmission
Mechanistic genes are those that encode proteins, enzymes and receptor subunits related to the mechanism of drug action. Key elements in cholinergic neurotransmission include ACh precursors (choline, acetyl-CoA), ACh synthesis (choline acetyltransferase), degradation enzymes (acetylcholinesterase, butyrylcholinesterase), choline transporter, vesicular ACh transporter, and cholinergic receptors (nicotinic, muscarinic).
Cholinergic neurons of the basal forebrain (basocortical cholinergic pathway, septohippocampal cholinergic pathway), where the nucleus basalis of Meynert is located, and cortical cholinergic projections, are the brain territories primarily affected in AD, with 60-80% depletion of cholinergic markers in severe cases [100].
Choline acetyltransferase (ChAT; EC 2.3.1.6), which catalyzes the biosynthesis of ACh from choline and acetyl-CoA, is encoded in the ChAT gene at 10q11.23. The first intron of the ChAT gene encodes the vesicular acetylcholine transporter (VAChT). VAChT transports ACh from the cytoplasm into the synaptic cleft. Mutations in ChAT and/or VAChT (SLC18A3) may represent potential susceptibility to AD [100].
Acetylcholinesterase (AChE; EC 3.1.1.7) is a serine hydrolase that hydrolyzes ACh to yield choline and acetate in the synaptic cleft; residual choline is recycled by the choline transporter and made available at the presynaptic level for de novo synthesis of ACh. AChE variants with catalytic activity include synaptic AChE (AChE-S) or tailed AChE (AChE-T), the most frequent variant in the brain; erythrocyte AChE (AChE-E) or

Mechanistic Genes Involved in Cholinergic Neurotransmission
Mechanistic genes are those that encode proteins, enzymes and receptor subunits related to the mechanism of drug action. Key elements in cholinergic neurotransmission include ACh precursors (choline, acetyl-CoA), ACh synthesis (choline acetyltransferase), degradation enzymes (acetylcholinesterase, butyrylcholinesterase), choline transporter, vesicular ACh transporter, and cholinergic receptors (nicotinic, muscarinic).
Cholinergic neurons of the basal forebrain (basocortical cholinergic pathway, septohippocampal cholinergic pathway), where the nucleus basalis of Meynert is located, and cortical cholinergic projections, are the brain territories primarily affected in AD, with 60-80% depletion of cholinergic markers in severe cases [100].
Choline acetyltransferase (ChAT; EC 2.3.1.6), which catalyzes the biosynthesis of ACh from choline and acetyl-CoA, is encoded in the ChAT gene at 10q11.23. The first intron of the ChAT gene encodes the vesicular acetylcholine transporter (VAChT). VAChT transports ACh from the cytoplasm into the synaptic cleft. Mutations in ChAT and/or VAChT (SLC18A3) may represent potential susceptibility to AD [100].
Acetylcholinesterase (AChE; EC 3.1.1.7) is a serine hydrolase that hydrolyzes ACh to yield choline and acetate in the synaptic cleft; residual choline is recycled by the choline transporter and made available at the presynaptic level for de novo synthesis of ACh. AChE variants with catalytic activity include synaptic AChE (AChE-S) or tailed AChE (AChE-T), the most frequent variant in the brain; erythrocyte AChE (AChE-E) or hydrophobic AChE (AChE-H); and read-through AChE (AChE-R). The Yt erythrocyte blood group antigen system is inserted into the AChE molecule (AChE, His322Asn). A 4-bp deletion located 17 kb upstream of the transcription start site that abolishes 1 of 2 adjacent hepatocyte nuclear factor-3 (HNF3) binding sites causes hypersensitivity to AChEIs and severe CNS symptoms under low-dose exposure to pyridostigmine. AChE activity is decreased in AD brains, and APP is involved in the regulation of AChE [100,125].
Human serum cholinesterase (acylcholine acylhydrolase) or butyrylcholinesterase (BuChE) (EC 3.1.1.8) is a serine hydrolase that catalyzes the hydrolysis of ACh and choline esters such as the muscle relaxants succinylcholine and mivacurium. BuChE is a 574-amino acid protein encoded by the butyrylcholinesterase (BCHE) gene (4 exons, 64 kb) at 3q26.1. Over 30 genetic variants of BCHE have been described. In AD cases, the allelic frequency of the K-variant is >0.2 (vs. 0.09 in controls), and the risk for AD in carriers of the K-variant increases in the presence of the APOE-4 allele in most studies. In mild cognitive impairment (MCI) cases, BCHE-K and APOE-4 accelerate cognitive decline, hippocampal volumetric loss, and progression to AD [126].
The choline transporter (CHT, CHT1) is encoded by the SLC5A7 (Solute Carrier Family 5 (Choline Transporter), Member 7) gene (2q12.3), with nine exons spanning 25 kb. CHT is a Na + -and Cl − -dependent high-affinity 580-amino acid protein (63.2 kD) with 12 transmembrane (TM) domains responsible for the uptake of choline for ACh synthesis in cholinergic neurons. An A-to-G transition at nucleotide 265, resulting in an Ile89-to-Val substitution (I89V) within the third TM domain, reduces the maximum rate of choline uptake by about 40%. The high-affinity choline transporter (CHT1, SLC5A7) expressed in cholinergic neurons represents a rate-limiting step for ACh synthesis. Alterations in CHT function decrease choline uptake and ACh synthesis, with a consequent impairment in cholinergic neurotransmission. CHT1 dysfunction may contribute to AD pathology. Aβ decreases choline uptake activity and cell surface CHT protein levels. CHT trafficking is different in wild-type APP (APPwt) and in Swedish mutant APP (APPSwe) SH-SY5Y human neuroblastoma cells. APP-CHT interaction is decreased in APPSwe transgenic mice [127,128].
The vesicular acetylcholine transporter (VAChT) is encoded by the SLC18A3 (Solute Carrier Family 18 (Vesicular Actylcholine), Member 3) gene (10q11.23). This gene encodes a transmembrane protein that transports ACh into presynaptic secretory vesicles to be released at cholinergic terminals in the CNS and peripheral nervous system. The SLC18A3 gene is located within the first intron of the ChAT gene. Mutations in the SLC18A3 gene cause presynaptic congenital myasthenic syndrome-21 (CMS21). Nitrosylation of VAChT is increased in the frontal cortex and hippocampus of APP/PS1 mice [129]. B6.eGFPChAT congenic mice with multiple gene copies of VAChT exhibit high VAChT protein expression in the hippocampal formation, accompanied by enhanced ACh release [130]. Mice with a targeted mutation in the SLC18A3 gene show a 40% reduction in transporter expression, with memory deficits that can be reversed with AChEIs. Diminished expression of the splicing regulator hnRNPA2/B1BACE1 causes abnormal splicing in BACE1; increased APP processing; accumulation of Aβ; and increase in GSK3, tau hyperphosphorylation, caspase-3 and neuronal death. In human brain, there is a correlation between decreased levels of VAChT and hnRNPA2/B1 levels and increased tau hyperphosphorylation [131]. There is a selective loss of cholinergic terminals in the neocortex and hippocampus of double transgenic (APP-K670N/M671L + PS1-M146L) mice, with relevant alterations in VAChT. The levels of ChAT, AChE, and BuChE are similar in the hippocampus of young apoE4 and apoE3 mice. ChAT levels decrease more in apoE4 than in apoE3 mice. The levels of muscarinic receptors are also higher in apoE4 mice. ACh release from hippocampal slices is reduced in old apoE4 mice in parallel with reduced VAChT levels [132].
The distribution of the vesicular acetylcholine transporter in early AD shows a decrease of about 47-62% in the cingulate cortex and parahippocampal-amygdaloid complex. The numbers of ChAT and VAChT neurons correlate to the severity of dementia and show no relationship with APOE status. Cholinergic baso-cortical and septo-hippocampal pathways are particularly damaged in AD, as reflected by PET studies of the VAChT [133].
Nicotinic and muscarinic ACh receptors are the final effectors of cholinergic neurotransmission. Nicotinic receptors (nAChRs) play an important role in the prefrontal cortex, where cortical and subcortical inputs are integrated to execute higher activities of the CNS (learning, attention, working memory planning, decision-making, and perception of reality). Mutations in the CHRNB2 or CHRNA7 genes that encode the nicotinic receptor β2 and α7 subunits can lead to brain disorders, including AD [134]. α7nAChR, encoded by the CHRNA7 gene, is involved in AD pathogenesis connected to hypocholinergic neurotransmission and Aβ deposition. Carriers of the CHRNA7 rs7179008 variant showed decreased risk of dementia. Single-nucleotide polymorphisms (SNPs) in the CHRNA7 or CHRFAM7A genes may affect susceptibility to AD. CHRFAM7A-2-bp deletion or CHRNA7 SNPs (rs1514246, rs2337506, and rs8027814) might be protective for AD [100,135].
Loss of basal forebrain cholinergic neurons correlates with cognitive decline in AD. Exposure to Aβ up-regulates neuronal α7nAChRs and increases neuronal excitability, and α7-nAChRs mediate, in part, Aβ-induced neurotoxicity, which is prevented by either the α7-nAChR antagonist methyllycaconitine or by α7 subunit gene deletion. In contrast, it appears that α7nAChR selective agonists (e.g., PHA-543613) and galantamine may ameliorate Aβimpaired working and reference memory, suggesting that α7 nAChR activation reduces Aβinduced cognitive deficits, whereas receptor blockage increases Aβ toxicity and cognitive impairment [136].

Transporter Genes
There are over 200 transporter proteins responsible for the transfer of endogenous and exogenous substances through the blood-brain barrier. The main categories of transporters, with some relevance in AD, include ATPase (P-type, V-type, and F-type subfamilies), ATPbinding cassette transporters (Subfamilies ABC1, MDR/TAP, CFTR/MRP, ALD, OABP, GCN20, and WHITE), and Solute carriers (high-affinity glutamate and neutral amino acid transporter family) (SLC) [81]. ABC and SLC variants are involved in the pathogenesis of dementia, and a large number of ABC transporters influence the efficacy and safety of more than 1000 different drugs, including AChEIs and memantine [11,12,81].
Many other transporters are associated with AD pathogenesis [100]. Defective transporters are attractive candidates for therapeutic intervention in AD, and their mutations should be considered in PGx strategies [100,137].
APOE-3 carriers are the best responders, and APOE-4 carriers are the worst responders to donepezil in either monotherapy or in drug combination regimes; CYP2D6-EMs tend to be good responders, and CYP2D6-PMs exhibit a poorer response to donepezil, with probably more ADRs [11,12,66,67,138,139,143,144].
ABCA1 regulates cholesterol transport and APOE metabolism. AD patients with the ABCA1 rs2230806 G/G genotype respond better to donepezil than carriers of the A/A and A/G genotypes, and ABCA1 rs2230806 G/G-APOE3 non-carriers show a better clinical response to donepezil. Patients homozygous for the T/T/T genotype in the ABCB1 haplotypes 1236C/2677G/3435C and 1236T/2677T/3435T show lower plasma donepezil concentration-to-dose ratios and better clinical response to donepezil. Donepezil may inhibit ABCB1 [100].
Patients harboring the APOE-4/BCHE-K* genotype show an earlier age of onset, an accelerated cognitive decline, and an irregular response to donepezil. In patients with MCI, donepezil accelerates cognitive decline in homozygous BCHE-K and APOE-4 carriers. The BCHE-K variant is associated with lower AChE hydrolyzing activity, and BuChE activity increases in parallel with disease progression. Donepezil treatment is not recommended in BCHE-K and APOE-4 carriers with MCI or dementia [145].
Donepezil treatment tends to increase APP forms in APOE-4 non-carriers and interacts with many drugs, causing cardiotoxicity [100].

Galantamine
Galantamine is a reversible AChEI with allosteric modulatory effects on nicotinic ACh receptors. This drug is rapidly absorbed (C max = 1 hr), with low protein binding (28.3-33.8%), a steady-state volume of distribution (V ss ) of 193 L, and an elimination half-life of 7-8 h (20-25% is excreted unchanged in urine) [147]. Median clearance in female/male AD patients is 12.4/14.8 L/h, probably due to body weight differences rather than a real gender effect. In patients with liver dysfunction, the metabolic clearance is reduced by 60%. Galantamine increases the levels of brain VAChT. Galantamine acts as a major substrate of CYP2D6, CYP3A4, ABCB1 and UGT1A1, and as an inhibitor of AChE and BCHE. Galantamine efficacy and safety can be modified by APOE, APP, AChE, BCHE, CHRNA4, CHRNA7, and CHRNB2 variants [100,141,147] (Table 4). CYP2D6 and CYP3A4 enzymes are responsible for galantamine metabolism. Glucuronidation, Ndemethylation, O-demethylation, epimerization, and N-oxidation are its major metabolic pathways. Galantamine pharmacokinetic parameters mainly depend on CYP2D6 variants. CYP2D6-PMs exhibit 45-61% higher dose-adjusted galantamine plasma concentrations than CYP2D6-NMs and IMs [141,147], without substantial effects on pharmacodynamics. There is no linear correlation between galantamine concentration and cognitive response in AD patients. Interaction with foods may modify the effects and bioavailability of galantamine [100].
Several studies indicate that galantamine in AD may show better results in APOE-4 non-carriers. Patients with MCI treated with galantamine for one year also showed a lower rate of whole brain atrophy, preferentially among APOE 4 carriers. Others suggest no Life 2022, 12, 460 25 of 38 major influence of APOE variants in the effects of galantamine in AD. CHRNA7 rs8024987 variants may also affect galantamine in females [100].
Saturable first-pass metabolism leads to 35% bioavailability of the administered dose and nonlinear short half-life pharmacokinetics, with renal elimination. Rivastigmine is a pseudo-irreversible dual inhibitor of AChE and BCHE with a very short t 1/2 (1-2 h) and longer duration of action due to blockade of AChE and BCHE for around 8.5 and 3.5 h, respectively.
Rivastigmine increases VAChT and ChAT expression in the frontal cortex, hippocampus, striatum, and cerebellum, providing additional effects on cholinergic neurotransmission. Liver and intestine esterases metabolize rivastigmine [141,143,148], whereas ChAT, AChE, BCHE, CHRNA4, CHRNB2, APOE, APP, and MAPT variants may affect rivastigmine pharmacokinetics and pharmacodynamics (Table 4). Rivastigmine is more effective in APOE-4 non-carriers in different ethnic groups. CYP enzymes are not involved in the metabolism of rivastigmine. UGT2B7-PMs show a deficient response to treatment and higher rivastigmine levels. CYP2D6*3, UGT2B7, and UGT1A9*5 carriers, in combination treatments with memantine, show differential responses to treatment. CHRNA7 variants and two intronic SNPs of ChAT (rs2177370 and rs3793790) influence the response to AChEIs. Lower response to rivastigmine and reduced enzyme activity are observed in patients harboring the BCHE-K-variant (rs1803274). SNPs outside the coding sequence of the BCHE gene (rs1126680, rs55781031) and the K-variant (p.A539T) may reduce enzyme activity. Carriers of these deleterious SNPs should receive lower doses of rivastigmine or start treatment with a different AChEI. BCHE-wt/wt females show a better response to rivastigmine than males. Males with the BCHE-K* variant tend to show faster cognitive decline than females. Rivastigmine may attenuate the progression of cognitive decline in male BuChE-K and in female BuChE wt/wt. Patients with the bigenic BCHE-K-APOE-4 genotype are poor responders to rivastigmine patch or memantine add-on therapy [100].

Huperzine A
Huperzine A is a natural Lycopodium sesquiterpene alkaloid extracted from the Chinese medicinal plant Huperzia serrata. This compound is a reversible and highly selective second-generation AChEI for AD and was approved in China in 1994 [149]. Huperzine A pharmacokinetics show age-dependent differential features. Age is a covariate with influence on huperzine A clearance. The plasma concentration-time profile of huperzine A in elderly patients reveals a one-compartment model with first-order absorption and elimination.
CYP1A2 is the major metabolizing enzyme of huperzine A in rat liver microsomes; secondary metabolizing enzymes are CYP3A1/2, CYP2C11, and CYP2E1. In humans, huperzine A is excreted unchanged by kidney, with no apparent involvement of CYP1A2, CYP2A6, CYP2C9, CYP2C19, CYP2D6, CYP2E1, or CYP3A4 enzymes [150] (Table 4). At a toxicological dose in rats, huperzine A may induce CYP1A2 by enhancement of transcription. Carboxylesterases (CESs) (CES1 and CES2) are enzymes catalyzing the hydrolysis of ester, amide, and carbamate chemicals. CESs might be tangentially involved in huperzine A metabolism.
Huperzine A penetrates the brain and interacts with ABCB1 and ABCG2 efflux transporters. Huperzine A is a substrate of ABCB1. In Abcb1a−/− mice, the brain to plasma concentration ratio of huperzine A is higher than in wild-type animals [100].

Pharmacogenetics of Memantine
Memantine is a non-competitive low-affinity NMDA receptor antagonist. Its long t 1/2 is about 70 h, and it is eliminated unchanged via the kidneys; however, several genes can influence its efficacy and safety [141]. Memantine displays its therapeutic effect by inhibition of glutamate via NMDA receptors, with partial antagonistic activity on GRIN2A, GRIN2B, GRIN3A, HTR3A, and CHRFAM7A. Memantine efficacy and safety may be modulated by pathogenic genes (APOE, PSEN1, and MAPT) and mechanistic genes (GRIN2A, GRIN2B, GRIN3A, HTR3A, CHRFAM7A, c-Fos, Homer1b, and PSD-95). Memantine is a strong inhibitor of CYP2B6 and CYP2D6 and a weak inhibitor of CYP1A2, CYP2A6, CYP2C9, CYP2C19, CYP2E1, and CYP3A4 [141,143,151]. In human liver microsomes, memantine inhibits CYP2B6 and CYP2D6; decreases CYP2A6 and CYP2C19; and has no effect on CYP1A2, CYP2E1, CYP2C9, or CYP3A4. When co-administered with CYP2B6 substrates, the metabolism of memantine is decreased by 65%. Memantine clearance is highly dependent on the NR1I2 rs1523130 variant. NR1I2 rs1523130 CC carriers show a faster elimination rate than CT or TT carriers [151].
Memantine transport across the blood-brain barrier (BBB) might be facilitated by proton-coupled organic cation antiporters. In AD patients, memantine can be used as a monotherapy or in combination with AChEIs. Memantine induces proteomic changes in the hippocampus and the cerebral cortex of transgenic mice (3 × Tg-AD), with modifications in the expression of 233 and 342 proteins, respectively. Memantine also reduces cerebrovascular Aβ and hemosiderin deposits in APP23 transgenic mice with cerebral amyloid angiopathy, by enhancing Aβ-cleaving insulin degrading enzyme (IDE) expression. Memantine increases histamine neuron activity, as reflected by a 60% increase in brain tele-methylhistamine levels and an increase in hypothalamic H3 autoreceptors, where histamine neurons are located [100].

Pharmacogenetics of Aducanumab
Up to now, no reliable information has been available on the pharmacogenetics of aducanumab and no specific studies have been reported in this regard. Based on its potential mechanism of action, as a scavenger of Aβ, it can be inferred that different mutations in the APP, PSEN1, and PSEN2 genes, related to the intensity of the Aβ load in neuritic plaques, affect the efficacy of this monoclonal antibody. It also seems clear that the APOE genotype affects the safety and efficacy of aducanumab. APOE-4/4 carriers not only respond poorly to aducanumab but worsen and develop severe side-effects, neuroinflammatory reactions, perivascular microedema, and white matter lesions [118][119][120].
Chronic administration of aducanumab to tgAPPPS1-21 mice modifies the expression of a series of proteins associated with AβPP trafficking/processing, neuronal cytoskeleton, stress response, metabolism, and mitochondria. If some of these pathways and their respective proteins, potentially implicated in AD pathogenesis, are affected in humans, it is likely that mutations in the corresponding genes might affect aducanumab efficacy and safety [152].
By analogy with other monoclonal antibodies, such as Abciximab, a Fab antibody fragment of the chimeric human-murine monoclonal antibody 7E3 that inhibits platelet aggregation by binding to IIb/IIIa receptors, or Trastuzumab, an immunoglobulin G1 (human-mouse monoclonal rhuMab HER2γ1-chain antihuman p185c-erbB2 receptor), which binds to the extracellular domain of human epidermal growth factor receptor 2 protein (HER-2) in breast neoplastic cells overexpressing HER-2, it cannot be excluded that aducanumab might be processed via phase-I enzymes of the CYP family, phase II enzymes, and CNS transporters (ABC and SLC families), or that mutagenic variants that affect genes responsible for neuroimmune cascades (IL1B, IL6, TNF) might modify the therapeutic response and toxicity of aducanumab [143].

Pharmacogenetics of Multifactorial Treatments
Multifactorial treatments should be designed in a personalized fashion with neuroprotective drugs, anti-dementia drugs, drugs to control concomitant pathologies, psychotropic drugs for the treatment of neuropsychiatric disorders, and other medications especially important for specific metabolic deficits. In these cases, the use of pharmacogenetic procedures is particularly useful, as illustrated in the following sections.
Most studies with multifactorial combination therapies have used the APOE gene and CYP gene variants as reference genes in pharmacogenetics. The main conclusions obtained from these studies agree that the best responders to conventional treatments with AChEIs, memantine, and various neuroprotective agents are patients with the APOE-3/3 genotype, while the worst responders are patients with the APOE-4/4 genotype. Intermediate responses are observed in patients with APOE-2/4 and APOE-3/4 genotypes, where the presence of the APOE-4 allele gives them the character of not good responders.
Since the APOE and TOMM40 genes occupy adjacent loci, it has been shown that both genes influence the pathogenesis of AD and the therapeutic response to anti-dementia treatments. In the first pharmacogenetic study in AD with the bigenic genotype APOE-TOMM40, carriers of the TOMM40 poly T-S/S genotype were good responders, similar to APOE-3/3 carriers, while patients harboring the L/L variant were the worst responders. Carriers of VL/VL and S/VL variants showed an intermediate response to multifactorial treatments. When analyzing the effect of both genes on the therapeutic response in AD patients, carriers of the APOE-4/4 genotype in haplotypes with TOMM40-L/L and S/L variants responded the worst to treatment. In contrast, carriers of the APOE-3/3 genotype in haplotypes with the TOMM40-S/S variant, and to a lesser extent TOMM40-S/VL and TOMM40-VL/VL variants, were the best responders to anti-dementia treatments. The APOE-4/4 genotype is exclusively associated with the TOMM40-L/L genotype in 100% of AD cases. It has been suggested that this haplotype (APOE-4/4-TOMM40-L/L) is responsible for premature neuronal death, early onset of the disease, accelerated cognitive decline, and a poor response to conventional treatments [22].

Pharmacoepigenetics
Various epigenetic aberrations, present in AD patients, contribute to enhancing brain neurodegeneration through the abnormal expression of pathogenic genes. Similar epigenetic alterations can alter the expression and function of the mechanistic, metabolic, transporter, and pleiotropic genes that constitute the pharmacoepigenetic apparatus, conditioning the response to drugs [81,82,100].
Very few pharmacoepigenetic studies have been carried out in AD, and there are no epigenetic drugs capable of protecting neurons against the process of premature death to which they are subjected in AD. Among the many epigenetic factors that influence the pathogenesis of AD and the response to drugs, different variants in the genes that encode sirtuins could alter the epigenetic machinery with deleterious effects on the brain. For example, the SIRT2-C/T variant (rs10410544) increases vulnerability to AD in APOEε4negative patients. In the first pharmacoepigenetic study with the bigenic cluster SIRT2-APOE, 18 haplotypes were identified with potential influence on the therapeutic response to multifactorial regimens with anti-dementia drugs. The APOE-4 allele accumulates in SIRT2-T/T and SIRT2-C/T carriers. SIRT2-C/C carriers are the worst responders, SIRT2-C/T carriers are the best responders, and SIRT2-T/T carriers are intermediate responders to combination treatments. When studying the therapeutic response by bigenic clusters of APOE-SIRT2, it was found that patients with the 33CC genotype were the ones who responded best to treatment; 33TT and 34CT carriers showed an intermediate response, and 24CC and 44CC carriers showed the worst response to multifactorial treatment. A similar analysis of the bigenic cluster SIRT2-CYP2D6 showed that the best responders were patients carrying the geno-phenotype SIRT2-C/T-CYP2D6-EM [13].

Pharmacogenomics of Mood Disorders and Anxiety
The treatment of depression and anxiety in cases of dementia is always inserted into multifactorial therapeutic regimens with anti-dementia agents and complementary treatments for the control of concomitant pathologies. The addition of antidepressants and anxiolytics in small doses to these therapeutic regimens is usually well tolerated. In personalized protocols, a similar favorable response is observed in men and women. APOE-4 carriers tend to respond worse to antidepressants and anxiolytics than APOE-3 carriers. Patients with the geno-phenotypes CYP2D6-NM, CYP2C9-NM, and CYP2C19-NM usually respond better than PMs or UMs; IM patients show an intermediate response to antidepressants. In contrast, patients with mutations in the CYP3A4 and CYP3A5 genes show an abnormal response to benzodiazepines, when compared to CYP3A4/5-NMs. CYP3A4/5-PMs show a more pronounced effect with an excess of side-effects to benzodiazepines, and CYP3A4/5-UMs respond poorly to benzodiazepines due to ultra-rapid metabolization of the drugs and the consequent lack of therapeutic effect [24,67,[153][154][155].
In patients with depression treated with antidepressants chosen by trial and error, according to the pharmacological experience of the prescribing physician, the rate of error and therapeutic failure is over 50%. When these patients receive the right drug, in correct doses, adapted to their pharmacogenetic profile, an improved therapeutic response is obtained in over 60% of cases between one and three months of personalized treatment [156].

Future Trends
Apart from the standard care needed by every patient with dementia, in the overall management of AD it is important to improve diagnostic accuracy (>30% error), correctly identify and treat concomitant pathologies, implement presymptomatic diagnostic protocols, personalize pharmacological treatments, and initiate prevention programs in the population at risk. A major problem in the management of AD is the lack of curative treatments; polypharmacy related to concomitant pathologies; and unnecessary abuse of psychotropic drugs, especially in nursing homes.
Once the differential diagnosis of AD has been established, with diverse procedures [40][41][42]46,[157][158][159][160][161], it is necessary to optimize the therapeutic resources adapted to each form of dementia, depending on the predominance of cognitive and behavioral impairment over the motor component or vice versa [162].
In people over 80 years of age, practically >95% of patients consume more than ten different medications per day for the treatment of ailments that contribute to the aggravation of the clinical picture of dementia (cardiovascular diseases and associated pathologies, cerebrovascular insufficiency and cerebral microinfarctions, metabolic disorders, arthropathies, etc.) [163][164][165][166][167][168][169].
When polypharmacy is unavoidable, pharmacogenetic procedures should be implemented for the personalization of indispensable pharmacological treatments to avoid ADRs and DDIs [17]. In the Caucasian population, only 20% of the cases are extensive metabolizers for drugs metabolized via CYP2D6-2C9-2C19-3A4/5 enzymes. This implies that by trial and error, the treatments administered to these patients, without knowledge of their pharmacogenetic profile, in more than 60% of cases will harm or will not provide any benefit, from a therapeutic point of view [12,14,15,17,23,100].
In addition to the conventional treatments available for different forms of dementia, assuming their limitations in terms of the cost-benefit ratio, and without forgetting that the search for etiopathogenic treatments is an unavoidable priority, the search for new forms of symptomatic treatments today should not be ruled out [169], as well as the use of alternative therapies that contribute to alleviate dysfunction, disconnection from the socio-family environment, and aberrant behaviors present in a high number of patients with dementia [170][171][172][173][174]. Furthermore, a deeper understanding of AD pathophysiology may lead to the identification of neuronal signaling pathways as potential novel targets for therapeutic intervention (i.e., tyrosine kinases) [175][176][177][178][179].
Once the necessary treatments have been established, therapeutic follow-up programs with reliable biomarkers should be implemented to guarantee the efficacy of the treatments administered and to eliminate those ineffective drugs that could also contribute to aggravating the clinical picture of dementia.
In terms of prevention, genomic screening programs should be implemented to identify the population at risk. These programs should include genomic clusters with the most informative pathogenic genes at a reasonable price. New NGS and GWAS techniques must be validated across different ethnicities to effectively cover genomic risk.
There are currently over 4000 human genes for which no functional or structural assignment has been established nor an alteration (either structural -SNP-or functional -epigenetic-related abnormal gene expression) associated with any pathology. It is expected that in the coming years, some of these genes will show subtle variants potentially associated with rare diseases and/or degenerative processes. On the other hand, Figures 3 and 6 clearly illustrate the accumulation of various genetic variants in the same patient, supporting a possible universal rule in genomics. A common rule to all complex, polygenic, and multifactorial diseases, such as AD is that the greater the number of genes affected, the earlier the onset of the disease, the more accelerated its clinical course and the worse the response to conventional treatments; the fewer genes affected, the later the onset of the disease, the slower the clinical course, and the more favorable therapeutic response.
An important issue is to validate the weight that each polymorphic variant has in the pathogenesis of AD. APOE is a paradigmatic example. Everyone accepts that the APOE-4 allele is a major risk factor for AD. However, technical difficulties, lack of interest (or both) mean that the pathogenic role of APOE-4 has been negligently considered until now in terms of therapeutics [101,180].
It is highly unlikely that current treatments, including aducanumab, will contribute to the efficient management of therapeutic deficiencies of AD. Under optimal conditions (and only in presymptomatic cases or in very mild cases), aducanumab may be useful in less than 20% of patients, without forgetting that in some cases (i.e., APOE-4 carriers), its effects could be deleterious [115][116][117][118][119][120][121].
Another drawback is the high cost of treatment, which will hardly be borne by the public health system in many countries and will establish greater inequality between rich and poor.
The only way to effectively fight AD is through prevention. AD is destroying the brains of the at-risk population since the brain stops maturing at around 30-35 years of age. From then, it can take more than 30 years before symptoms appear. That is the time available to the patient and the medical community to intercept the course of the disease and slow or stop the neurodegenerative process in genomically-vulnerable individuals. For this, it is necessary to implement preventive programs with prophylactic interventions and treatments aimed at protecting the brain against the destructive process of neurodegeneration. Among the preventive strategies, all the intercurrent diseases, which anyone throughout life can suffer from and whose inadequate treatment can contribute to neuronal damage, should not be ignored.
At present, no preventive strategy is specific enough to effectively combat premature neurodegeneration associated with AD because, among other things, the etiopathogenesis of the disease itself is in question as it focuses primarily on the pathogenic binomial of amyloidosis-tauopathy with little therapeutic success over two decades [181][182][183][184]. However, any future preventive strategy should be based on the following methodological steps: (i) identification of the population at risk by global genomic screening; (ii) identification of potential concomitant pathologies that increase brain risk; (iii) initiation of multifactorial prophylactic intervention once the brain stops maturing; (iv) monitoring of the efficacy of the preventive strategy with epigenetic, proteomic, and metabolomic biomarkers; and (v) prophylactic intervention of concomitant pathologies and/or treatment of symptomatic ailments.
The practical application of pharmacogenetics based on the individual genomic profile of each patient requires the following: (i) adjusting the dose of each treatment according to the condition of NM (regular dose), IM (50% reduction of the usual dose), PM (avoid drugs whose only metabolization route is associated with the mutant enzyme), or UM (increase the dose by 25%; if there is no response or side effects appear, change the drug regime); (ii) avoid co-administration of drugs that are substrates and inhibitors of the same metabolic enzyme; (iii) in IMs, use the help of inducers that enhance the metabolic capacity of the enzyme with decreased activity; (iv) in some cases, it would be possible to modulate the enzymatic activity of mutant genes with epigenetic drugs; and (v) avoid any treatment for which there is evidence of toxicity from the patient's mutant condition (12-15, 17, 22-24).
When pharmacogenetics become a mature discipline, the most effective manner to optimize the available therapeutic resources and reduce ADRs and DDIs is the implementation of pharmacogenetic procedures as a daily routine in the clinic. The routine use of pharmacogenetics is limited by a series of educational factors (information deficit among health professionals), technical factors (poor characterization of the pharmacogenetic properties of more than 50% of commonly used drugs), biomedical factors (shortage of biomarkers of drug efficacy and toxicity), economic factors (high cost of pharmacogenetic screening), administrative factors (lack of organization in hospitals and health centers for personalized medicine), and regulatory factors (poor definition of pharmacogenetics parameters by regulatory agencies for the use of drugs and for the development of new pharmaceutical products) [185,186].
Cognitive markers indicate than females show worse cognitive performance than males. Likewise, depression and anxiety are also more prevalent in women than in men. The ECG is abnormal in 40% of the patients (38% F; 43% M). No sex-related differences are found in MRI anomalies, which are present in over 70% of the cases.
In addition to conventional clinical markers, the most useful biomarkers for a predictive diagnosis or diagnostic confirmation of antemortem AD are genomic markers, epigenetic biomarkers, neurotransmitters, and levels of Aβ/tau in the brain (PET Scan) and/or in body fluids.
The genomic screening of pathogenic genes in AD patients revealed that most patients (>60%) are carriers of over 10 pathogenic genes. Genes that accumulate pathogenic variants more frequently in the same patient with AD are PRNP, PSEN1, ACE, A2M, BIN1, CLU, CPZ, LHFPL6, MS4A6A, MS4A4E, and PICALM. Over 80% of AD patients in the Caucasian population are deficient metabolizers for the most common drugs, which are metabolized via CYP2D6, CYP2C9, CYP2C19, and CYP3A4/5 enzymes. From 60% to >90% of the patients, depending on the clinical stage of the disease, require multifactorial treatments with risk of ADRs and DDIs. The implementation of pharmacogenetics in clinical practice can help optimize the limited therapeutic resources available to treat AD and personalize the use of anti-dementia drugs in combination with other medications for the treatment of concomitant disorders.