Background

Schizophrenia (SCZ) is a chronic, hereditary and disabling neuropsychiatric disorder with a worldwide prevalence of approximately 1% [1, 2]. In the etiology of SCZ, genetic factors are thought to play an important role, the heritability currently ranges from 64 to 81%; although, genetic mechanisms remain unclear [2,3,4]. The mainstay of schizophrenia treatment has been antipsychotic drugs over the past 60 years; however, clinical response differs significantly between patients, with an overall response rate of 50–70% [1, 5,6,7]. Many of patients with schizophrenia discontinue or switch drug regimens due to lack of treatment efficacy and/or drugs adverse side effects. In this view, there is a great need for the identification of predictive clinical and biological markers of treatment consequence [8, 9]. Pharmacogenetic biomarkers focus to predict which patients could improve with specified drugs according to genetic variants. Thus, genotype-based customized drug treatments may allow optimizing the antipsychotic treatment, while hel** to minimize drugs side effects [10, 11].

To date, pharmacogenomics studies of response to treatment in schizophrenia, have typically focused on genes encoding for drug targets, called pharmacodynamics related genes. Many of the research investigating the association of pharmacodynamic genes with antipsychotic treatment response have concentrated on dopaminergic pathways, one of the primary mechanisms of function of antipsychotics, especially the gene coding for the dopamine D2 receptor(DRD2 gene), which is a binding object for all available antipsychotic drugs. Dopaminergic gene SNPs are strongly related to drug sensitivity of antipsychotics; for example, several studies have indicated positive associations between DRD2 gene and antipsychotic response [10,11,12,13,14].

Lingyue Ma and et al. by a systematic review and meta-analysis indicated that for Asian patients, at rs1799978(A241G) in DRD2 gene AA genotype had a significantly greater improvement after risperidone therapy [33]). The ARMS-PCRs were performed in two 15ul reactions for each patient, which contained an initial denaturation at 95 °C for 2 min followed by 35–36 cycles of denaturing at 95 °C for 30 s, annealing at 65 °C for 30s and extending at 72 °C for 45s. After that, a final extension at 72 °C for 5 min was applied.

COMT rs4680 was analyzed by polymerase chain reaction-restriction fragment length polymorphism (PCR- RFLP). The PCR was conducted in a 15ul reaction system which included an initial denaturation at 95 °C for 2 min followed by 35 cycles of denaturing at 95 °C for 30 s, annealing at 61 °C for 30 s and extending at 72 °C for 30s. After 35 cycles, it experienced a final extension at 72 °C for 5 min. NIaIII restriction enzyme was used for digestion; as explained by Qianqian He and et al [34]. Genoty** of 120-bp duplication polymorphism in DRD4 was performed using the Gap-PCR (primers by [35]). The PCR was carried out in a15ul reaction system which contained a first denaturation at 95 °C for 2 min followed by 30 cycles of denaturing at 95 °C for 30 s, annealing at 66 °C for 30 s and extending at 72 °C for 55s. After 30 cycles, a final extension at 72 °C for 5 min was done.

Statistical analysis

All data analysis was conducted using Statistical Package for the Social Sciences SPSS, Version 26. Continuous variables were expressed in the form of mean ± standard deviation. Kolmogorov–Smirnov test was applied to check whether the data were normally distributed; which normally distributed and abnormally distributed data between two groups were calculated respectively by t-test and Mann-Whitney U test. In order to assess the association between categorical variables of clinical parameters and genotype associations chi-square test was performed. Associations between polymorphisms and antipsychotic treatment response were analyzed under five genetic models including homozygous model, heterozygous model, recessive model, dominant model and co-dominant model. Logistic regression analysis was carried out in order to analyze these five genetic models. Adjusted odds ratios, adjusted P-values and 95% confident interval (95%CI) were calculated. Furthermore, logistic regression analysis was used to examine the effect of interaction among polymorphisms on antipsychotic treatment response. P < 0.05 was considered as statistical significance in all tests.

Gpower software version 3.1.9.7 was applied in order to analyze the power of logistic regression. According to Chen et al., 2010 [36] effect size in logistic regression can be classified as follows:

  • Odds Ratio < 1.68 - Very small.

  • 1.68 ≤ Odds Ratio < 3.47 - Small.

  • 3.47 ≤ Odds Ratio < 6.71 - Medium.

  • Odds Ratio ≥ 6.71 – Large.

The power analysis demonstrated that a sample size of 101 is sufficient to reveal a medium to large effect size with a minimum power of 80% at a significance level of 5%.

Results

Clinical and demographic characteristics of patients

One hundred and one patients with schizophrenia were included in this study (75 men and 26 women), which among them 51 patients were classified in the treatment-responder group and 50 patients met the criteria for treatment-resistance. There were significant differences between two groups in marital status, smoking, duration of hospitalization, chlorpromazine-equivalent daily dose and total PANSS score(Table 1). With regarding to the clinical characteristics, significantly longer duration of hospitalization, higher PANSS score and also higher chlorpromazine-equivalent daily dose were observed in treatment-resistant group(Table 1). Furthermore, married and smoker patients were significantly more in the treatment-responder group comparing to the treatment-resistant group.

Table 1 clinical and demographic characteristics of patients
Full size table

The frequently prescribed antipsychotics were perphenazine(47%), olanzapine(37%), risperidone(32%), quetiapine(26%) and haloperidol (24%).

Genetic analysis

Genotypic and allelic associations

The population was in Hardy-Weinberg equilibrium for the four polymorphisms genotyped in this study (p > 0.05). The G allele of DRD2 A-241G was associated with increased risk of resistant to treatment when compared to A allele (OR(95%CI): 3.661,P = 0.02,Table 2). As for DRD4 120-bp duplication, there existed a considerable difference in allele distribution(P = 0.064,Table 2), indicating a higher frequency of 240-bp allele in treatment-resistant patients; although, the P value was not significant.

Table 2 Genotype and allele frequencies of polymorphisms within DRD2, COMT and DRD4
Full size table

The logistic regression analysis revealed that regarding DRD4 120-bp duplication, patients with 120/120 and 240/120 had a lower risk of develo** resistant to treatment as compared to patients with 240/240 genotype(AOR(95%CI): 0.196, P value: 0.033, Table 3). Moreover, regarding DRD4 120-bp duplication, homozygous and heterozygous genetic models indicated relations with antipsychotic treatment response, however it did not reach the significance level(P = 0.055 and P = 0.053 Respectively.

Table 3 Associations between DRD2, COMT and DRD4 polymorphisms and antipsychotic treatment response under five genetic models by using chi-square and logistic regression for the AORs
Full size table

Gene-gene interaction analysis

Whether the presence of three polymorphism’s genotypes could influence the risk for treatment resistant to antipsychotic drugs was determined between DRD4 120-bp duplication, COMT rs4680 and DRD2 A-241G. We carried out all possible subgroup analyses; which the significant interactions are indicated in Table 4. In the COMT Val/Val subset, we found significant association of the DRD4 genotype with antipsychotics treatment response; where the combination of COMT Val/Val genotype and DRD4 240/240 genotype had a high risk for develo** treatment- resistance(OR(95%CI) = 3.232(1.056–9.892), P = 0.04). Also, among patients with COMT Val/Met - Met/Met genotypes(Met allele carriers) those whose genotypes where AA for DRD2 A-241G were significantly more likely to respond to antipsychotic drugs as compared to other genotype combinations(OR(95%CI) = 2.540(1.138–5.668), P = 0.023).

Table 4 Gene-gene Interaction analysis for 2 and 3 locus models by using logistic regression
Full size table

Furthermore, analyzing the interactions of DRD2 A-241G and DRD4 120-bp duplication polymorphisms, revealed a significant association between DRD2 AA genotype and DRD4 120 bp allele carriers(DRD4 120/240 − 120/120), patients with this genotype combination had a significantly better respond to antipsychotics(OR(95%CI) = 3.000(1.279–7.035), P = 0.012).

Additionally, logistic regression analysis indicated a significant interaction among DRD2 A-241G, DRD4 120-bp duplication and COMT rs4680 polymorphism’s; where patients with AA − 120/240 or 120/120 - Val/Met or Met/Met showed a significantly better respond to antipsychotics when comparing to patients with GA or GG-240/240-Val/Val genotype(OR(95%CI) = 2.363(1.057–5.281), P = 0.036).

Discussion

The key findings of the present study were as follows. First, our genetic analysis for DRD2 A-241G(rs1799978) polymorphism detected a significantly higher frequency of G allele in resistant to treatment patients in comparison with responders. A possible explanation for this association could be that since DRD2 binds to dopamine and is a G-protein coupled receptor, A-241G polymorphism is considered to be related to DRD2 density and affinity [37]. Furthermore, regarding DRD2, it is recorded that this receptor lonely could adjust effects of atypical antipsychotics; suggesting that DRD2 plays a substantial role in patients response to atypical antipsychotics [44].

Our sample size was relatively small especially the subgroups in gene-gene interactions were small and a few genotypes had limited carriers(GG genotype of DRD2 A-241G polymorphism and GC of DRD2 rs1801028); also CC genotype of DRD2 rs1801028 was not observed in our sample. Furthermore, no multiple testing correction was used for P-values in gene-gene interaction analyses and we should mention the type I error possibility as a limitation. Besides, the studied population was only from one single ethnicity. Consequently, further investigations with larger sample sizes and meta-analyses, from various ethnicities analyzing several polymorphisms involved in pathways related to antipsychotics actions are warranted; in order to move toward personalized medicine in schizophrenia.

Conclusion

In summary, our results suggest that COMT, DRD2 and DRD4 genes together and DRD2 and DRD4 genes separately, may effect and predict antipsychotic treatment response in Iranian population. This kind of study may provide the possibility of genetic screening before starting a new antipsychotic trial, resulting in a better chance to achieve the most effective treatment for each patient in a shorter period of time, decreasing costs and minimizing adverse side effects of drugs.