The Frequency of Certain Genetic Polymorphisms and Their Influence on the Therapeutic Response of Patients Treated with Olanzapine

Olanzapine is an atypical antipsychotic that serves as the first line of defense against schizophrenia (1). However, patients appear to not react equally to this therapy, most likely due to significant inter-individual variations in plasma drug concentrations (2, 3).

Olanzapine metabolism takes place with the leading participation of two enzymes, CYP1A2 and CYP2D6. The activity of CYP450 enzymes is known to be under the influence of many drugs and substances from the environment that result in potentially significant interactions in which one drug may intensify the toxicity (inhibition of cytochrome) or reduce the therapeutic effect of another drug (induction of cytochrome) (4, 5).

A large number of genetic variations of CYP1A2 are known, but only one allele, CYP1A2*1F, defined by polymorphism −163A (rs76251C>A), has proven to have a significant impact on the concentration of metabolites and olanzapine in white people in terms of activities. For another variation, rs2472297C>T, located within the CYP1A1/1А2 gene cluster, is shown to speed up olanzapine metabolism in the synthesis of n-desmethyl metabolite synthesis. Cytochrome P450 CYP1A2 is very susceptible to induction, which causes great interindividual variations in enzyme activity (6, 7). The expression of CYP1A2 strongly regulates the aryl hydrocarbon receptor (AHR) gene locus (8). It has been shown that one of the polymorphisms in this locus (rs4410790C>T) reduces the synthesis of n-desmethyl olanzapine in humans. However, the quantitative share of the influence of these variations on the systematic exposure to olanzapine in patients is practically negligible; their presence can explain only about 2% of the concentration of this drug in body fluids (9).

The aim of this study was to examine the frequency of certain genetic polymorphisms as well as their influence on the therapeutic response of patients treated with olanzapine.

This research was conducted according to the design of a prospective, interventional, clinical study of phase IV by type of case series, where the stratification of the subjects was carried out according to the obtained types of tested genotypes.

This case-control study was carried out between 2014 and 2016. A total of 120 respondents were recruited in this study. Patients were recruited at the Clinic of Psychiatry, University Clinical Center Kragujevac in Serbia, who, according to research criteria (DSM-V) (10), had a diagnosis of schizophrenia (11). Participation in the study was voluntary; patients were included in the study after obtaining their consent to participate in the study, and the study was approved by the Ethics Committee of the University Clinical Center Kragujevac in Serbia (approval No. 01/5273; date of approval: 8th of July, 2009). The study was conducted in accordance with the Principles of Good Clinical Practice and the Declaration of Helsinki (12). There was no conflict of interest in the study related to the study drug or other possible commercial aspects of the study.

The criteria for inclusion in the study were:

• The patient was capable of understanding the nature of the study and signed an informed patient consent form before any study procedure was performed.

• The patient is of Serbian nationality, male or female, of adult age, 18 years or older.

• Previously diagnosed with schizophrenia (minimum 2 years before inclusion in the study).

• Hospitalized or outpatient patients currently in the phase of exacerbation of psychotic symptoms who, on the BPRS (Brief Psychiatric Rating Scale) (13), have scores of 4 or more on two of the four items (9, 10, 11, and 15).

• Sexually active participants in the study who have committed to using medically acceptable forms of contraception during the duration of the study.

• Stable social environment for patients treated on an outpatient basis, including the presence of a person from the environment in order to obtain reliable data.

The exclusion criteria were:

• On axis 1 of the DSM-V, a diagnosis of a mental disorder other than schizophrenia was made.

• High suicidal risk or risk of heteroaggressive behavior as assessed by the researcher.

• Current diagnosis of comorbid alcoholism or abuse of other psychoactive substances

• Pregnancy or breastfeeding.

• Findings of clinically significant abnormalities on physical examination, electrocardiography (ECG), or laboratory analyses, determined or taken at the initial visit, as well as the presence of significant somatic diseases that, in the opinion of the researcher, prevent the subject's participation in the study.

• Patients who had neuroleptic malignant syndrome.

• Previous allergic reactions to olanzapine.

• Use of other drugs known to affect the metabolism of olanzapine, including but not limited to the following: aprepitant, ciprofloxacin, clarithromycin, diltiazem, erythromycin, fluconazole, imatinib, itraconazole, carbamazepine, ketoconazole, modafinil, nefazodone, rifampicin, verapamil, voriconazole, and phenobarbitone.

• Electroconvulsive therapy was applied in the last year.

A sample for genotyping in the form of 10 ml of whole blood was taken from each subject in a vacuum tube with EDTA and stored at −20ºC until analysis. Samples were taken once, at the beginning or end of the study, as appropriate. The most important scale for assessing therapeutic response in this study was the PANSS, which, next to the BPRS, is the most widely used scale (14).

The PANSS scale was designed by Kay in 1987 and consists of three subscales (positive symptoms, negative symptoms, and general psychopathology). The PANSS has a total of 30 items graded from 1 to 7 (1-absent, 2-minimally present, 3-weakly present, 4-moderately present, 5-moderately very present, 6-strongly present, and 7-extremely present). The scale includes all items from the BPRS scale and selected items from the Psychopathology Rating Scale. The PANSS measurement scale is characterized by high reliability, construct validity, and objectivity (14). Andreasen and others published a paper in the American Journal of Psychiatry in 2005 titled Remission in schizophrenia: proposed criteria and rationale for consensus (15). Consensus determined 8 PANSS items for symptomatic remission criteria: P1-delusions, P2-conceptual disorganization, P3-hallucinations, N1-affective dullness, N4-passivity or apathetic social withdrawal, N6-lack of spontaneity and smooth flow of conversation, G5-mannerisms and unusual body posture, and G9-unusual thought content.

Symptomatic remission criteria are met if the listed items are rated 3 or less and must last at least 6 months. Researchers who used this scale were previously trained and certified in the administration of the PANSS.

A total of 217 people were considered for participation in the study, of which 120 were included in the research. All subjects who were included completed the study. The research examined the frequency of certain genetic polymorphisms as well as their influence on the therapeutic response.

In our study, 6.7% of the subjects were heterozygous carriers. Variant alleles normally occur in Caucasians in about 8% of cases and are very rare or absent in homovariant form (16), which is the case with the patients in our study. In a 2005 study that included the Japanese population, the presence of neither homozygous nor heterozygous variants of this gene were detected (17). The extremely rare presence of this variant, in addition to Caucasians, is also confirmed by Lopez et al. in 2013, where they even state the differences within the Caucasian race, i.e., that Mexicans have a lower presence of this variant compared to Europeans, but still more often than Asian peoples (18).

As for CYP1A2*1C, only the wild type of the enzyme genotype was detected in the studied population. These results differ from those obtained in some other populations. In a study conducted on the Turkish population, it was found that one in ten carriers of the variant allele (as many as 2/3 of them are of the homovariant type). The representation of the variation varies from region to region, so the CYP1A2*1C allelic frequency in East Asia is present in 28% of the population, compared to only 2% in white Europeans. An even higher frequency of variation was found in Africa (31%) and in African Americans (36%). It is slightly lower in South Asia and China (20–24%), significantly lower than in these populations, but still four times more common in Hispanics (8%) (19, 20).

The representation of CYP1A2*1F was significantly more dispersible, and the presence of both wild and variant homozygotes as well as heterozygotes was recorded, where almost 90% were carriers of the variation, which is otherwise a characteristic of the white European population (21, 22). A somewhat smaller representation of the variant allele is among Asian peoples, while it is rarest, but again in half of the cases, found in Africa (19).

CYP1A1/A2 was investigated somewhat less frequently, possibly because the SNP is located between the CYP1A1 and CYP1A2 genes on chromosome 15q24 (23). In the population of patients who participated in our study, a quarter were carriers of the variant allele, among which only a few were carriers of the homovariant type. In other European nations, the prevalence of this polymorphism is even rarer and ranges from 10 to 19%. It is somewhat more common (in a similar percentage as in our study) in Hispanic Americans, while in Asian peoples it exceeds 30%, and in the Chinese population even 50% (24).

In our research, three genotypes were monitored: E158K, V257M, and E308G. All three forms of the E158K genotype were detected in the examined patient population. Almost ¾ of the patients are carriers of the variant allele, and among them, 17% (of the total population) are carriers of the homovariant type. This distribution of variant alleles is characteristic of the white European population. An example is a study from 2014 where the percentage of the variant allele in the population of Russians from Kursk and Belgorod is almost identical to that of our study (25), while the homovariant type is much more common, even up to half in some Asian peoples (26).

The results obtained in our study show that the presence of the V257M variant allele in our population is rare, i.e., below 5%, and only as heterozygous carriers. Such a rare presence of this polymorphism is normally a characteristic of the white European race, although some studies have shown that the percentage exceeds 10% (e.g., in Ireland) with the presence of the homovariant type (27). The presence of the E308G genotype polymorphism in our study population was present in every fourth subject, but only two were carriers of the homovariant form. In other European nations, the variation is slightly more common and ranges from 30 to 35%, with the presence of the homovariant type also being more frequent (3–6%, while in our country it was less than 2%) (27). The presence of variations is significantly more common in Asian peoples and, for example, reaches 46% in the Korean population (28).

CYP2D6*3 was a very rare variation in the studied population, i.e., only two patients (1.7%) were registered as heterozygous carriers. The presence of variant carriers is much more common in some peoples, and in Africa, it ranges up to 9% in some parts of Zimbabwe (29). While in Europe and Asia, the presence of variations is much lower and does not exceed 3% (30).

In the case of the CYP2D6*4 genotype, variations were more frequent and were detected in a third of subjects. The homovariant type was extremely rare and was detected in only 3 patients. It is known that the presence of variant alleles is much more common in European nations and ranges from 18.5 to 26.3%, while it is far less common in Asians (0.42–7.7%) (31). Proof of this is a study from Russia, where this polymorphism is 12.2 times more common in Russians living in the European part compared to smaller ethnic groups (Nanai people) that inhabit the far east of the Asian part of the country (32). As far as other areas are concerned, a study from Brazil showed that the presence of the variant allele is about 13%, which is expected due to the different racial origin or mixing of the inhabitants of this country (33).

The presence of CYP2D6*6 variations was extremely rare in our studied population. Three heterovariants and one homovariant carrier were detected (3.3%). This variation is otherwise rare. Current knowledge indicates that it has reached the highest percentage in the European population and is around 2%, while in other parts of the world (Asia, Africa, and America), it is even rarer and is found only in a few per thousand inhabitants (30).

Polymorphism of metabolic enzymes and olanzapine. In our study, we tested polymorphisms of the olanzapine metabolizing enzymes UGT1A4, CYP1A2, FMO3, and CYP2D6; they did not, on the whole, affect the clinical response to olanzapine therapy. Previous research was mostly focused on the influence of the polymorphism of these enzymes on the concentration of olanzapine in the blood and/or its metabolites and not on the clinical outcomes themselves, where some variations were shown to be able to change the biotransformation of the drug, independently or in the presence of inducers. Acceleration of the metabolism of olazapine or a decrease in concentrations has previously been found in carriers of UGT1A4*3, CYP1A2*1F, and CYP1A2A1/A2 rs2472297C>T variants (9, 34, 35), while a decrease in the metabolic reactions of olanzapine has been shown for carriers of UGT2B10*2 (36), FMO1*6, and FMO3 E308G variations (9).

However, the influence of individual polymorphisms on the metabolism, concentrations, and clinical response of olanzapine is very complex. The main reason for this is the fact that the elimination of olanzapine is mediated by a larger number of enzymes that make up two metabolic pathways, one of which is the main one. Additionally, the metabolism of olanzapine is significantly influenced by other, non-genetic factors, of which the effect of exogenous inducers and inhibitors is one of the most pronounced. Methodological aspects of the design of individual studies are also important. Such circumstances are probably one of the important reasons for different findings when analyzing the same genetic variations.

Presence of gene variations CYP1A2*1C (rs2069514, - 3860G>A), CYP1A2 (rs2472297, 74735539C>T), FMO3 E158K (rs2266782, 15167G>A), FMO3 V257M (rs1736557, 18281G>A), FMO3 E30 8G (rs2266780, 21443A>G), CYP2D6*3 (rs35742686, 2549delA), CYP2D6*4 (rs3892097, 1846G>A), CYP2D6*6 (rs5030655, 1707delT) does not change the clinical response to olanzapine therapy in patients with schizophrenia, compared to patients who are carriers of the wild-type gene.

In our study, the presence of the investigated gene variations (UGT1A4, CYP1A2, FMO3, and CYP2D6) does not affect the clinical response to olanzapine therapy in patients suffering from schizophrenia, compared to patients who are carriers of the wild-type gene.

The results of our study, regarding the overall impact of individual tested polymorphisms in the metabolizing enzymes of olanzapine, are in high agreement with the previous findings. The influence of individual polymorphisms on the metabolism, concentrations, and clinical response of olanzapine is very complex. Additionally, the metabolism of olanzapine is significantly influenced by other, non-genetic factors, of which the effect of exogenous inducers and inhibitors is one of the most pronounced.