Influence of Pathogenic and Metabolic Genes on the Pharmacogenetics of Mood Disorders in Alzheimer’s Disease

Background: Mood disorders represent a risk factor for dementia and are present in over 60% of cases with Alzheimer’s disease (AD). More than 80% variability in drug pharmacokinetics and pharmacodynamics is associated with pharmacogenetics. Methods: Anxiety and depression symptoms were assessed in 1006 patients with dementia (591 females, 415 males) and the influence of pathogenic (APOE) and metabolic (CYP2D6, CYP2C19, and CYP2C9) gene variants on the therapeutic outcome were analyzed after treatment with a multifactorial regime in a natural setting. Results and Conclusions: (i) Biochemical, hematological, and metabolic differences may contribute to changes in drug efficacy and safety; (ii) anxiety and depression are more frequent and severe in females than males; (iii) both females and males respond similarly to treatment, showing significant improvements in anxiety and depression; (iv) APOE-3 carriers are the best responders and APOE-4 carriers tend to be the worst responders to conventional treatments; and (v) among CYP2D6, CYP2C19, and CYP2C9 genophenotypes, normal metabolizers (NMs) and intermediate metabolizers (IMs) are significantly better responders than poor metabolizers (PMs) and ultra-rapid metabolizers (UMs) to therapeutic interventions that modify anxiety and depression phenotypes in dementia. APOE-4 carriers and CYP-related PMs and UMs deserve special attention for their vulnerability and poor response to current treatments.


Introduction
Over 50 million people suffer dementia and it is expected that 75 million people will be affected in 2030 and 145 million in 2050, at an increasing rate of 7.7 million new cases per year. The global economic cost for dementia is over US$604 billion, equivalent to 1% of the global gross domestic product. Alzheimer's disease (AD) is the most frequent form of dementia (>50%), followed by vascular (VD) and mixed dementia (MXD) (30-40%) and other phenotypic presentations of neurocognitive disorders (NCDs) [1]. In patients over 70-75 years of age, MXD is the most prevalent form of dementia (>70%) [2].
Genomic defects, epigenetic aberrations, cerebrovascular damage, and environmental inducers are the major risk factors that precipitate pathogenic cascades leading to the clinical phenotype of dementia. Dementia is characterized by progressive cognitive deterioration, behavioral changes, functional decline, and classical neuropathological hallmarks represented by extracellular Aβ deposition in senile plaques, intracellular neurofibrillary tangles, aberrant dendritic morphology, and neuronal loss in critical regions of the central nervous system (CNS) [3][4][5][6][7]. dementia. Early-onset AD patients exhibit greater prevalence of all BDs, especially anxiety, irritability, and sleep disorders [48].
In this retrospective study, we investigate the presence of anxiety and depression in a well-characterized sample of AD patients and analyzed the influence of APOE variants and CYP2D6, CYP2C19, and CYP2C9 geno-phenotypes on the frequency and severity of anxiety and depression as well as the effect that different APOE single nucleotide polymorphisms (SNPs) and CYP phenotypes exert on the evolution of mood disorders after treatment with multifactorial therapy.

Sex-Related Biochemical, Hematological, Metabolic and Clinical Phenotypes
Age was almost identical in females and males ranging from 50-96 years. However, globally, biochemical, hematological, metabolic, psychometric, and anthropometric parameters were significantly different between females and males ( Table 1). The electrocardiogram (EKG) was abnormal in 47.82% of cases (43.36% in females and 54.35% in males, p < 0.001). Systolic and diastolic pressure values were higher in males than females (p < 0.001); however, heart rate was faster in females than males (p < 0.001). Total cholesterol, HDL-cholesterol, and LDL-cholesterol levels were also higher in females than males (p < 0.001), and triglycerides were higher in males than females (p < 0.005). Glucose levels were much higher in males than females (p < 0.001). Other important differences were found in kidney function, liver function, creatine phosphokinase (CK), alkaline phosphatase (ALP), and electrolytes (K + ). Red blood cell, hematocrit, and hemoglobin values were higher in males than females whereas platelets were more elevated in females than males. White blood cell number was also higher in males than females, especially leukocytes, monocytes, basophils, and eosinophils, while lymphocytes were slightly more abundant in females. Iron and ferritin values were higher in males; in contrast, folic acid and vitamin B12 values were higher in females (Table 1).

Parametric Differences
Mood disorders in dementia are the result of a number of factors, either endogenous or exogenous. Our study clearly shows substantial sex-related differences in the frequency and intensity of anxiety and depression in AD patients (Tables 2 and 3; Figures 1 and 3), as well as striking differences in anthropometric, biochemical, hematological, metabolic, and psychometric parameters between both sexes (Tables 1-3) that directly and/or indirectly may affect cognition and emotional status. Although mood disorders appear to be an intrinsic component of the dementia behavioral phenotype, in our sample of well-characterized patients, with a clinical follow-up for two decades, the improvement observed in both anxiety and depression cannot be exclusively attributed to psychotropic treatment, but also to the multifactorial intervention, covering concomitant problems, which were present in over 60% of the patients. Mood disorders in our sample appear to be unrelated to cardiovascular factors; and, apparently, both anxiety and depression present with greater frequency, with higher intensity, in females with lower red blood cell parameters and poorer metabolic conditions. Mood disorders are also unrelated to hyperglycemia in dementia (Table 1).
Our study corroborates the assertion that APOE is a major, but not exclusive, pathogenic determinant of the pharmacogenetic outcome in AD and that APOE-4 carriers are at biological disadvantage with respect to patients harboring the APOE-3 allele, as previously documented [2,4,6,18,[34][35][36][37][49][50][51]. Most studies, where AD patients were treated with multifactorial combinations, revealed that APOE-3/3 carriers are the best responders and APOE-4/4 carriers are the worst responders. Concerning CYP-related pharmacogenomic outcomes, CYP2D6-EMs are the best responders, CYP2D6-PMs are the worst responders, and CYP2D6-IMs and UMs show an intermediate response. Many other pathogenic genes [50] and genes encoding components of the epigenetic machinery [51] also influence the pharmacogenetic outcome. Genome-wide association studies (GWAS) identified 31 genes located in 19 risk loci for major depressive disorder (MDD). Common and rare variants of L3MBTL2 are associated with AD. Transcript (mRNA) expression levels of SORCS3 and OAT are differentially expressed in AD brain tissues, and 13 MDD risk genes may interact with core AD genes such as HACE1, NEGR1, and SLC6A15 [52].
About 25% of our patients with moderate-to-severe depression were treated with fluoxetine, paroxetine, or sertraline. Over 80% of these patients showed a net improvement in their depressive condition. However, these positive results cannot only be attributed to the effect of antidepressants, but also to the complementary treatments for their concomitant ailments. Meta-analyses of double-blind randomized controlled trials comparing antidepressants vs. placebo for depression in AD revealed inefficacy in most cases with different drugs (sertraline, mirtazapine, imipramine, fluoxetine, and clomipramine) [55].
The mechanisms underlying depression in dementia remain unclarified [56]. Genetic and environmental factors are potentially involved [44,[57][58][59][60][61][62]. For instance, an increase in depression and anxiety symptoms have been reported in nursing homes during the coronavirus pandemic [63]. About 60% of depressive patients are receiving an inappropriate medication according to their pharmacogenetic background [64,65], and community psychiatrists and pharmacists are more accurate in their psychotropic prescriptions when they know the CYP profile of their patients [65][66][67].
Fluoxetine is a major substrate of CYP2D6 and CYP2C9; a minor substrate of CYP1A2, CYP2B6, CYP2E1, and CYP3A4; a strong inhibitor of CYP2D6, a moderate inhibitor of CYP1A2 and CYP2C19, and a weak inhibitor of CYP2B6, CYP2C9, and CYP3A4 [25,54]. Paroxetine is a major substrate of CYP2D6; a strong inhibitor of CYP2D6; a moderate inhibitor of CYP2B6 and a weak inhibitor of CYP1A2, CYP2C8, CYP2C9, CYP2C19, and CYP3A4 [25,54]. Sertraline is a major substrate of CYP2D6 and CYP2C19; a minor substrate of CYP2A6, CYP2B6, CYP2C9, and CYP3A4; a moderate inhibitor of CYP2C19, CYP2D6, and CYP3A4; and a weak inhibitor of CYP1A2, CYP2C8, and CYP2C9 [25,54]. CYP2C19 and CYP2D6 variants affect the occurrence of ADRs in patients treated with selective serotonin reuptake inhibitors (SSRIs)(citalopram, escitalopram, sertraline, fluvoxamine, fluoxetine, paroxetine), including anxiety associated with CYP2D6, and ECG prolonged QT intervals associated with CYP2C19 [68]. The serotonin receptor 1A (HTR1A) rs878567 and CYP2C19 rs12248560 gene variants are associated with depression severity [69]. Angiotensin-converting enzyme (ACE) variants influence mood in AD [49]. The co-administration of ACE inhibitors and statins with antidepressants may affect therapeutic outcomes [70]. In our case, we did not observe any ADRs in the patients treated with small doses of enalapril, atorvastatin, and SSRIs.
Our study demonstrates that a multifactorial regime, covering biochemical and metabolic deficiencies, together with neuroprotectants and adjusted psychotropic medication, can be beneficial for emotional stability in AD patients, and that APOE-4 carriers and CYP-related PMs and UMs, representing over 40% of the AD population, deserve special attention for their vulnerability and poor response to current treatments.

Statistical Analysis
Data were analyzed by using IBM SPSS Statistics 20 (IBM Corp., Armonk, NY, USA) and SigmaPlot 10.0 (SYSTAT Software Inc., San Jose, CA, USA). Comparisons between groups were studied by t-Test, Mann-Whitney Rank Sum Test, Chi Square without Yates correction and Fisher exact, and Pearson Correlation Analysis (Nonlinear Regression, Durbin-Watson Statistic, Normality Test, Constant Variance Test, 95% Confidence), when appropriate. In studies of correlation analysis, all cases have been ordinated from the lowest to the highest values for maximum differentiation in the figures. All values are expressed as mean ± SD, and the degree of significance is considered when p < 0.05.

Conclusions
Major conclusions from this study include the following: (i) Biochemical, hematological, and metabolic differences may contribute to changes in drug efficacy and safety (tentative contribution: 5-20% that depends on the number of drugs involved); (ii) anxiety and depression are more frequent and severe in females than males with dementia; (iii) cardiovascular disorders and associated risk factors are more frequent in males than females; however, vascular problems appear to be unrelated to mood disorders in dementia; (iv) both females and males respond similarly to a multifactorial regime, showing significant improvements in anxiety and depression; (v) APOE-3 carriers are the best responders and APOE-4 carriers tend to be the worst responders to conventional treatments in combination regimes; (vi) among CYP2D6, CYP2C19, and CYP2C9 geno-phenotypes, NMs and IMs are significantly better responders than PMs and UMs to therapeutic interventions that modify anxiety and depression phenotypes; and (vii) APOE-4 carriers and CYP-related PMs and UMs deserve special attention for their vulnerability and poor response to current treatments.
The main conclusions related to pharmacogenetics should be interpreted with caution taking into account the polypharmacy received by dementia patients and assuming that the pharmacogenetic response depends on many more genes than those discussed in this study.  Informed Consent Statement: All patients and/or their legal representatives provided informed consent for genotyping, clinical assessment, and treatment before they participated in the study.
Data Availability Statement: Data supporting reported results can be found in the CIBE Database at EuroEspes International Center of Neuroscience and Genomic Medicine. www.euroespes.com.

Conflicts of Interest:
The authors declare no conflict of interest.