- Email: [email protected]
Association analysis of ZNF804A (zinc finger protein 804A) rs1344706 with therapeutic response to atypical antipsychotics in first-episode Chinese patients with schizophrenia
Available online at www.sciencedirect.com
Comprehensive Psychiatry 53 (2012) 1044 – 1048 www.elsevier.com/locate/comppsych
Association analysis of ZNF804A (zinc finger protein 804A) rs1344706 with therapeutic response to atypical antipsychotics in first-episode Chinese patients with schizophrenia Jinbei Zhang 1 , Xiaoli Wu 1 , Feici Diao, Zhaoyu Gan, Zhiyong Zhong, Qingling Wei, Nianhong Guan⁎ Department of Psychiatry, The Third Affiliated Hospital, Sun Yat-sen University, Guangzhou 510630, China
Abstract `The single-nucleotide polymorphism rs1344706 located in the ZNF804A zinc finger protein 804A gene is a well-established genome-wide significant variant for schizophrenia. The aim of this study was to investigate the potential association between this ZNF804A polymorphism and treatment response to atypical antipsychotic. Seventy-one first-episode inpatients with schizophrenia receiving olanzapine, aripiprazole, or quetiapine monotherapy were enrolled. Symptom response to treatment was assessed using the Positive and Negative Syndrome Scale (PANSS) on admission and reassessed after 4 weeks of treatment. Single-nucleotide polymorphism rs1344706 was genotyped by direct sequencing. There was substantial difference in treatment response among patients with 3 different genotypes regarding total PANSS score and positive subscore (for total PANSS score, F = 4.608, df = 2, P = .013; for positive subscore, F = 4.183, df = 2, P = .019). Compared with G homozygotes, T carriers showed significantly less improvement in total PANSS score as well as positive subscore (for total PANSS score, F = 8.724, df = 1, P = .004; for positive subscore, F = 9.392, df = 1, P = .005). Our results suggest that ZNF80A rs1344706 polymorphism may play a role in treatment response to atypical antipsychotic, although replication is required to confirm this finding. © 2012 Elsevier Inc. All rights reserved.
1. Introduction Response to antipsychotic treatment in patients with schizophrenia (SZ) is highly variable. Lines of evidence have suggested that genetic factors may play a large role in the variation of treatment response [1]. Over the past decades, polymorphisms in candidate genes, especially the serotonin 5HT and dopamine D2-like receptors, have been the major focus of most pharmacogenetic studies that tried to identify potential genetic factors influencing the response to antipsychotic treatment [2-5]. Nowadays, genome-wide association studies (GWASs) have successfully identified several susceptibility genes for major psychosis including SZ [6,7]. Yet, so far, whether these new SZ susceptibility genes are implicated in the well-known antipsychotic target system and thereby influence therapeutic response remains largely unknown. Conflict of interest: The authors declare no conflict of interest. ⁎ Corresponding author. Tel.: +86 20 85253128; fax: +86 2087563011. E-mail address: [email protected] (N. Guan). 1 Jinbei Zhang and Xiaoli Wu contributed equally to the work. 0010-440X/$ – see front matter © 2012 Elsevier Inc. All rights reserved. doi:10.1016/j.comppsych.2012.02.002
Because the single-nucleotide polymorphism (SNP) rs1344706 located at gene ZNF804A (zinc finger protein 804A) was identified as the first genetic variant achieving genome-wide significance for SZ [8], the association between SNP rs1344706 and SZ has been confirmed by subsequent other replication studies in a range of patient populations, including several independent SZ GWAS [911] and case-control studies [12,13]. The gene ZNF804A is known to be brain expressed and encodes a protein containing a C2H2-type domain associated with the zincfinger protein family. However, the exact biologic function of the gene ZNF804A remains unknown. In a study on the functional mechanism of SNP rs1344706, Riley et al [12] found that T risk carriers showed increased expression of ZNF804A messenger RNA. This finding was replicated by Williams et al [14] who also reported that the relationship between this SNP and expression is not relevant to disease according to the further examination of expression quantitative trait loci. Despite the poor knowledge on the gene ZNF804A and SNP rs1344706, several studies have characterized the related phenotype of SZ with SNP
J. Zhang et al. / Comprehensive Psychiatry 53 (2012) 1044–1048
rs1344706 using imaging or neuropsychologic approaches [15-18]. For example, one imaging genetics study by Esslinger et al [15] has demonstrated that the ZNF804A rs1344706 T risk allele influenced brain functional disconnectivity between dorsolateral prefrontal cortex and hippocampal formation during an N-back memory task in healthy subjects. Meanwhile, another study by Walters et al [16] on the neuropsychologic effects of rs1344706 found that carriers of the risk allele had significantly “better” cognitive performance than non–risk allele carriers in patients but not healthy controls. A recently published study on the association between symptoms of SZ and SNP rs1344706 showed modest evidence that carriers of the rs1344706 T risk allele score higher on a measure that weights for the number and severity of lifetime overactive episodes [19]. All of these findings suggest that ZNF804A rs1344706 is associated with a broad psychosis phenotype. Given that response to antipsychotic treatment is also a genetically determined phenotype of SZ, we hypothesized that there may be a potential association between ZNF804A rs1344706 and treatment response to atypical antipsychotics. In the present study, 54 first-episode inpatients with SZ receiving normal clinical treatment were enrolled, and the Positive and Negative Syndrome Scale (PNASS) score was used to assess the treatment improvement over 4 weeks.
2. Materials and methods This study was conducted under naturalistic clinical treatment settings from September 2009 to August 2010. The study protocol was previously reviewed and approved by the ethics committee at the Third Affiliated Hospital of Sun-Yat Sen University. All subjects provided written, informed consent before participating in the study. All subjects are of Chinese descent and inpatients from the Psychiatric Department, the Third Affiliated Hospital of Sun-Yat Sen University. Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition, diagnosis of SZ was established using the Structured Clinical Interview for DSMIV (SCID) performed by a single experienced psychiatrist (WXL) and reviewed by a senior psychiatrist (ZHJB). All patients were experiencing their first-episode psychosis and
1045
receiving atypical antipsychotic monotherapy. During their hospitalization period, antipsychotic treatment was provided according to normal clinical practice with olanzapine (OLZ) up to 30 mg/d, aripiprazole up to 30 mg/d, and quetiapine up to 500 mg/d. If needed, low dose of lorazepam (up to 2 mg/d) is added to ameliorate sleep. To confirm the SZ diagnosis, all patients were followed up for at least 6 months after discharge. Exclusive criteria were as follows: (a) severe physical illness; (b) organic mental diseases, mental retardation, or dementia; (c) pregnancy or lactation (for women patients); and (d) substance abuse within the past 3 months. Assessment of psychopathology and response to antipsychotic treatment were performed by 2 training senior psychiatrists (ZJB and GZY) using the Positive and Negative Syndrome Scale (PANSS). All patients were assessed on admission and reassessed after 4 weeks of treatment. All clinical raters were blind to the rs1344706 genotype. Blood samples were collected in EDTA-containing tubes, and leukocytes were isolated using red blood cell lysis buffer containing 150 mmol/L NH4Cl, 10 mmol/L KHCO3, and 0.1 mmol/L EDTA. Genomic DNA was extracted from the leukocytes using SDS-proteinase K treatment followed by phenol/chloroform extraction. Genomic DNA was amplified by polymerase chain reaction to generate a 443 base pair product spanning rs1344706. Primers were as follows: upper GAATCTAGA GTCATGCAGG and lower CAAGTTATTC TCTAGAGTCC. The polymerase chain reaction products were subjected to direct sequencing, conducted by the Beijing Genomics Institute. Genomic DNA samples were randomly selected for replication. The correspondence between the repeated sequencing and the original sequencing was 100%. The data analyses were performed by a biotechnologist. Software program Finetti (http://ihg2.helmholtzmuenchen.de/cgi-bin/hw/hwa1.pl) was used to test the distribution of genotypes for deviations from HardyWeinberg equilibrium. SPSS for Windows (version 13.0; SPSS Inc, Chicago, IL) was used to compare the clinical and demographic variables between patients with 3 different genotypes. For the baseline characteristics, analysis of variance (ANOVA) for continuous variables and χ 2 test for categorical variables were used to examine the difference among 3 groups. The Fisher exact test was used for
Table 1 Baseline characteristics of patients with SZ with 3 genotypes of ZNF804A rs1344706 polymorphism Genotype
GG, n = 19
GT, n = 41
TT, n = 11
F/χ 2, P value
Sex (M/F) Age (y) Age at first onset (y) Total PNASS score on admission Positive subscore on admission Negative subscore on admission Antipsychotic medication (OLZ/Q/A) a
7/12 22.32 ± 5.91 21.63 ± 5.89 101.26 ± 12.97 24.21 ± 4.13 24.32 ± 7.20 11/2/6
22/19 24.56 ± 6.97 23.96 ± 7.01 101.90 ± 12.30 24.95 ± 4.56 24.02 ± 4.36 21/7/13
6/5 22.64 ± 7.54 21.95 ± 7.60 101.45 ± 9.10 23.55 ± 3.80 26.09 ± 5.94 4/5/2
1.62, .45 0.86, .43 0.92, .40 0.02, .98 0.53, .59 0.62, .54 4.95, .29 a
M/F indicates male/female; Q, quetiapine; A, aripiprazole. a Fisher exact test.
1046
J. Zhang et al. / Comprehensive Psychiatry 53 (2012) 1044–1048
Table 2 Comparison of percentage change in PANSS score of first-episode patients with different genotypes after the 4-week treatment Genotype
GG, n = 19
GT, n = 41
TT, n = 11
F, P value⁎
F, P value⁎⁎
Percentage reduction in total PANSS score Percentage reduction in positive PANSS score Percentage reduction in negative PANSS score
44.47 ± 10.44 54.89 ± 13.76 37.68 ± 15.28
34.8 ± 14.85 38.34 ± 23.98 31.95 ± 16.47
31.55 ± 10.52 40.36 ± 16.07 24.18 ± 8.47
4.608, .013 4.183, .019 2.788, .069
8.724, .004 8.392, .005 3.119, .082
⁎ P value among the 3 groups. ⁎⁎ P value between groups GG and GT + GT.
categorical variables if the expected cell size was less than 5. Treatment response over 4 weeks was presented as percentage reduction in PANSS score (mean ± SD). The relationship between age, age of onset, sex, antipsychotic type, baseline PANSS scores, and treatment response was examined using Pearson correlation coefficient. To investigate the effects of genotype on changes in PANSS, comparison of percentage reduction in PANSS score between genotypic subgroups was carried out using univariate ANOVA. Because a higher percentage of subjects in the non–risk carrier group are prescribed OLZ, which might result in significantly improved response, comparison of percentage reduction in PANSS score in the OLZ group was separately performed to further rule out the impact of medication. Variables significantly correlated with the treatment response were selected as covariates to control for confounding factors in univariate ANOVA. For post hoc comparisons, LSD multiple range test was used. For all of the tests, level of significance was set at P b .05.
negative symptoms subscore (P = .002, .025, and .01, respectively). Subsequently, antipsychotic type was included as a covariate in the association of genotype with changes in total PANSS scores and subscores. The association still remained significant for total PANSS scores and positive subscores (P = .013 and .019, respectively). When the comparison was performed between T carriers and GG homozygotes, a more stronger association was identified in total PNASS score and positive subscores (P = .004 and .005, respectively) (Table 2). As demonstrated in Table 3, in the OLZ group, the direction of changes in total PANSS scores and subscores was consistent with that in the whole group, although not significant (all P N .05), which may attribute to the small sample size. As Fig. 1 showed, these results indicated that the patients with the GG genotype showed substantial improvement on positive symptom response compared with the patients carrying the T allele.
4. Discussion 3. Results There were 19 patients with GG genotype, 11 with TT genotype, and 41 with GT genotype. The genotype distribution was in Hardy-Weinberg equilibrium (P = .152). Baseline demographics and clinical characteristics are presented in Table 1. There were no significant differences between the genotypic subgroups about sex, age, age at first onset, total PNASS score, positive PNASS subscore, negative PNASS subscore, and type of antipsychotic medication (all P N .05). Variables including age, age of onset, sex, and baseline PANSS scores showed no significant correlation with the treatment response. However, antipsychotic type closely correlated with the change in PANSS scores after 4 weeks of treatment. Primary ANOVA showed that the rs1344706 polymorphism in ZNF804A gene was significantly associated with the changes in total PNASS score and positive and
Response to antipsychotic treatment is considered to be a genetically determined phenotype of SZ [1]. Over the past decade, most pharmacogenetic studies have focused on the antipsychotic drug targets, such as the serotonin 5-HT and dopamine D2-like receptors [2,5]. Although association between treatment response and polymorphisms in these candidate genes has been reported by quite a few studies, much of the diversity in treatment response remains unexplained. The successful application of GWAS in research for susceptibility gene for major psychosis has come up with several robust variants [6]. These newly identified risk variants now have become the focus of studies related to etiology and pathophysiology of SZ. In the present study, we found that polymorphism rs1344706 in gene ZNF804A, the first genetic variant achieving genome-wide significance, was associated with positive symptom response to atypical antipsychotics in a
Table 3 Comparison of percentage change in PANSS score between OLZ-treated T carriers and noncarriers after the 4-week treatment Genotype
GG, n = 11
GT + TT n = 25
F, P value
Percentage reduction in total PANSS score Percentage reduction in positive PANSS score Percentage reduction in negative PANSS score
44.24 ± 12.88 54.65 ± 15.20 34.13 ± 26.96
40.84 ± 16.35 45.72 ± 26.81 32.16 ± 18.40
0.369, .547 1.058, .311 0.065, .800
Change in positive PNASS score
J. Zhang et al. / Comprehensive Psychiatry 53 (2012) 1044–1048 80 70 60 50 40 30 20 10 0 GG
GT
TT
Fig. 1. Association between rs1344706 genotypes and changes in positive PANSS scores. P = .016.
group of first-episode patients with SZ. In terms of poor positive symptom response to treatment in patients with SZ with T risk allele, our result is consistent with the finding reported by a recent study that the T risk allele was linked to the number and severity of lifetime overactive episodes [19]. In this study, no significant difference in the baseline PNASS scores was found between 3 patient groups with different genotype. Of note, our patients were young firstepisode patients with SZ. To be hospitalized, they had to be severely affected and, in most cases, manifest conspicuous positive symptoms. Hereby, our inpatient group was not representative for the general Chinese first-episode patients with SZ, and the relationship between polymorphism rs1344706 and the baseline PNASS scores could not be inferred from our study. Nevertheless, the exclusive enrollment of first-episode patients was advantageous for it prevented some confounding factors such as duration of illness and therapeutic history. Because the treatment response is also related to the antipsychotic medication, we added type of antipsychotic as a covariate. The association was reduced but remained significant. In addition, because more than 50% of patients in our study were on OLZ treatment, we performed the same statistical analysis in the OLZ group only, which showed the same direction of changes in total PANSS scores and subscores. Yet, to investigate the association between antipsychotic-specific treatments with polymorphism rs1344706, a larger sample receiving the antipsychotic of interest is needed. To our knowledge, this is the first study to investigate the role of ZNF804A rs1344706 in response to atypical antipsychotic treatment. Atypical antipsychotics are considered to be more effective than typical antipsychotics because of their affinity for both 5-HT and dopamine D2 receptors. The risk allele of rs1344706 has been reported to be associated with increased expression of ZNF804A messenger RNA [12]. Although Williams et al [14] also reported that the relationship between this SNP and expression is not relevant to disease, the involvement of the ZNF804A gene in variation of treatment response could not be ruled out. Signal transduction cascade triggered by these receptors plays an important role in the effectiveness of antipsychotic drugs. ZNF804A is predicted to be a transcription regulator that is
1047
most likely to be involved in the downstream of a certain signaling pathway. Therefore, to investigate whether ZNF804A is involved in signaling pathway mediated by 5HT or dopamine D2 receptors may help to clarify the association between polymorphism rs1344706 and treatment response diversity. This study had several limitations. Firstly, our sample size is relatively small for a genetic study, although only firstepisode patients were enrolled. Secondly, selection bias is inevitable because our sample consisted of inpatients only. Such a sample may not be representative for Chinese firstepisode patient with SZ in general. In summary, our study suggests that ZNF804A rs1344706 polymorphism is associated with positive response to atypical antipsychotic treatment in a group of first-episode patients with SZ. More studies with a larger sample and other ethnic population are needed to confirm this finding. Acknowledgment Funding for this study was provided by the Science and Technology Plan Project (2009B080701080) of Guangdong Province and China National Science Fund for Young Scholars (81000180 and 81101028); the Science and Technology Plan Project of Guangdong Province and China National Science Fund for Young Scholars had no further role in study design; in the collection, analysis, and interpretation of data; in the writing of the report; and in the decision to submit the manuscript for publication. The authors would like to gratefully acknowledge the contributions of all of the physicians, nurses, technicians, and patients who participated in this study. References [1] Arranz MJ, Kerwin RW. Neurotransmitter-related genes and antipsychotic response: pharmacogenetics meets psychiatric treatment. Ann Med 2000;32:128-33. [2] Wang L, Fang C, Zhang A, Du J, Yu L, Ma J, et al. The –1019 C/G polymorphism of the 5-HT(1)A receptor gene is associated with negative symptom response to risperidone treatment in schizophrenia patients. J Psychopharmacol 2008;22:904-9. [3] Tsai SJ, Hong CJ, Yu YW, Lin CH, Song HL, Lai HC, et al. Association study of a functional serotonin transporter gene polymorphism with schizophrenia, psychopathology and clozapine response. Schizophr Res 2000;44:177-81. [4] Lencz T, Robinson DG, Xu K, Ekholm J, Sevy S, Gunduz-Bruce H, et al. DRD2 promoter region variation as a predictor of sustained response to antipsychotic medication in first-episode schizophrenia patients. Am J Psychiatry 2006;163:529-31. [5] Lencz T, Robinson DG, Napolitano B, Sevy S, Kane JM, Goldman D, et al. DRD2 promoter region variation predicts antipsychotic-induced weight gain in first episode schizophrenia. Pharmacogenet Genomics 2010;20:569-72. [6] Williams HJ, Craddock N, Russo G, Hamshere ML, Moskvina V, Dwyer S, et al. Most genome-wide significant susceptibility loci for schizophrenia and bipolar disorder reported to date cross-traditional diagnostic boundaries. Hum Mol Genet 2011;20:387-91. [7] Sullivan PF. The psychiatric GWAS consortium: big science comes to psychiatry. Neuron 2010;68:182-6.
1048
J. Zhang et al. / Comprehensive Psychiatry 53 (2012) 1044–1048
[8] O'Donovan MC, Craddock N, Norton N, Williams H, Peirce T, Moskvina V, et al. Identification of loci associated with schizophrenia by genome-wide association and follow-up. Nat Genet 2008;40: 1053-5. [9] Purcell SM, Wray NR, Stone JL, Visscher PM, O'Donovan MC, Sullivan PF, et al. Common polygenic variation contributes to risk of schizophrenia and bipolar disorder. Nature 2009;460:748-52. [10] Shi J, Levinson DF, Duan J, Sanders AR, Zheng Y, Pe'er I, et al. Common variants on chromosome 6p22.1 are associated with schizophrenia. Nature 2009;460:753-7. [11] Stefansson H, Ophoff RA, Steinberg S, Andreassen OA, Cichon S, Rujescu D, et al. Common variants conferring risk of schizophrenia. Nature 2009;460:744-7. [12] Riley B, Thiselton D, Maher BS, Bigdeli T, Wormley B, McMichael GO, et al. Replication of association between schizophrenia and ZNF804A in the Irish case-control study of schizophrenia sample. Mol Psychiatry 2010;15:29-37. [13] Zhang R, Lu SM, Qiu C, Liu XG, Gao CG, Guo TW, et al. Populationbased and family-based association studies of ZNF804A locus and schizophrenia. Mol Psychiatry 2010.
[14] Williams HJ, Norton N, Dwyer S, Moskvina V, Nikolov I, Carroll L, et al. Fine mapping of ZNF804A and genome-wide significant evidence for its involvement in schizophrenia and bipolar disorder. Mol Psychiatry 2010. [15] Esslinger C, Walter H, Kirsch P, Erk S, Schnell K, Arnold C, et al. Neural mechanisms of a genome-wide supported psychosis variant. Science 2009;324:605. [16] Walters JT, Corvin A, Owen MJ, Williams H, Dragovic M, Quinn EM, et al. Psychosis susceptibility gene ZNF804A and cognitive performance in schizophrenia. Arch Gen Psychiatry 2010;67:692-700. [17] Lencz T, Szeszko PR, DeRosse P, Burdick KE, Bromet EJ, Bilder RM, et al. A schizophrenia risk gene, ZNF804A, influences neuroanatomical and neurocognitive phenotypes. Neuropsychopharmacology 2010;35:2284-91. [18] Donohoe G, Rose E, Frodl T, Morris D, Spoletini I, Adriano F, et al. ZNF804A risk allele is associated with relatively intact gray matter volume in patients with schizophrenia. Neuroimage 2011;54:2132-7. [19] Cummings E, Donohoe G, McDonald C, Dinan TG, O'Neill FA, O'Callaghan E, et al. Clinical symptomatology and the psychosis risk gene ZNF804A. Schizophr Res 2010;122:273-5.
Comprehensive Psychiatry 53 (2012) 1044 – 1048 www.elsevier.com/locate/comppsych
Association analysis of ZNF804A (zinc finger protein 804A) rs1344706 with therapeutic response to atypical antipsychotics in first-episode Chinese patients with schizophrenia Jinbei Zhang 1 , Xiaoli Wu 1 , Feici Diao, Zhaoyu Gan, Zhiyong Zhong, Qingling Wei, Nianhong Guan⁎ Department of Psychiatry, The Third Affiliated Hospital, Sun Yat-sen University, Guangzhou 510630, China
Abstract `The single-nucleotide polymorphism rs1344706 located in the ZNF804A zinc finger protein 804A gene is a well-established genome-wide significant variant for schizophrenia. The aim of this study was to investigate the potential association between this ZNF804A polymorphism and treatment response to atypical antipsychotic. Seventy-one first-episode inpatients with schizophrenia receiving olanzapine, aripiprazole, or quetiapine monotherapy were enrolled. Symptom response to treatment was assessed using the Positive and Negative Syndrome Scale (PANSS) on admission and reassessed after 4 weeks of treatment. Single-nucleotide polymorphism rs1344706 was genotyped by direct sequencing. There was substantial difference in treatment response among patients with 3 different genotypes regarding total PANSS score and positive subscore (for total PANSS score, F = 4.608, df = 2, P = .013; for positive subscore, F = 4.183, df = 2, P = .019). Compared with G homozygotes, T carriers showed significantly less improvement in total PANSS score as well as positive subscore (for total PANSS score, F = 8.724, df = 1, P = .004; for positive subscore, F = 9.392, df = 1, P = .005). Our results suggest that ZNF80A rs1344706 polymorphism may play a role in treatment response to atypical antipsychotic, although replication is required to confirm this finding. © 2012 Elsevier Inc. All rights reserved.
1. Introduction Response to antipsychotic treatment in patients with schizophrenia (SZ) is highly variable. Lines of evidence have suggested that genetic factors may play a large role in the variation of treatment response [1]. Over the past decades, polymorphisms in candidate genes, especially the serotonin 5HT and dopamine D2-like receptors, have been the major focus of most pharmacogenetic studies that tried to identify potential genetic factors influencing the response to antipsychotic treatment [2-5]. Nowadays, genome-wide association studies (GWASs) have successfully identified several susceptibility genes for major psychosis including SZ [6,7]. Yet, so far, whether these new SZ susceptibility genes are implicated in the well-known antipsychotic target system and thereby influence therapeutic response remains largely unknown. Conflict of interest: The authors declare no conflict of interest. ⁎ Corresponding author. Tel.: +86 20 85253128; fax: +86 2087563011. E-mail address: [email protected] (N. Guan). 1 Jinbei Zhang and Xiaoli Wu contributed equally to the work. 0010-440X/$ – see front matter © 2012 Elsevier Inc. All rights reserved. doi:10.1016/j.comppsych.2012.02.002
Because the single-nucleotide polymorphism (SNP) rs1344706 located at gene ZNF804A (zinc finger protein 804A) was identified as the first genetic variant achieving genome-wide significance for SZ [8], the association between SNP rs1344706 and SZ has been confirmed by subsequent other replication studies in a range of patient populations, including several independent SZ GWAS [911] and case-control studies [12,13]. The gene ZNF804A is known to be brain expressed and encodes a protein containing a C2H2-type domain associated with the zincfinger protein family. However, the exact biologic function of the gene ZNF804A remains unknown. In a study on the functional mechanism of SNP rs1344706, Riley et al [12] found that T risk carriers showed increased expression of ZNF804A messenger RNA. This finding was replicated by Williams et al [14] who also reported that the relationship between this SNP and expression is not relevant to disease according to the further examination of expression quantitative trait loci. Despite the poor knowledge on the gene ZNF804A and SNP rs1344706, several studies have characterized the related phenotype of SZ with SNP
J. Zhang et al. / Comprehensive Psychiatry 53 (2012) 1044–1048
rs1344706 using imaging or neuropsychologic approaches [15-18]. For example, one imaging genetics study by Esslinger et al [15] has demonstrated that the ZNF804A rs1344706 T risk allele influenced brain functional disconnectivity between dorsolateral prefrontal cortex and hippocampal formation during an N-back memory task in healthy subjects. Meanwhile, another study by Walters et al [16] on the neuropsychologic effects of rs1344706 found that carriers of the risk allele had significantly “better” cognitive performance than non–risk allele carriers in patients but not healthy controls. A recently published study on the association between symptoms of SZ and SNP rs1344706 showed modest evidence that carriers of the rs1344706 T risk allele score higher on a measure that weights for the number and severity of lifetime overactive episodes [19]. All of these findings suggest that ZNF804A rs1344706 is associated with a broad psychosis phenotype. Given that response to antipsychotic treatment is also a genetically determined phenotype of SZ, we hypothesized that there may be a potential association between ZNF804A rs1344706 and treatment response to atypical antipsychotics. In the present study, 54 first-episode inpatients with SZ receiving normal clinical treatment were enrolled, and the Positive and Negative Syndrome Scale (PNASS) score was used to assess the treatment improvement over 4 weeks.
2. Materials and methods This study was conducted under naturalistic clinical treatment settings from September 2009 to August 2010. The study protocol was previously reviewed and approved by the ethics committee at the Third Affiliated Hospital of Sun-Yat Sen University. All subjects provided written, informed consent before participating in the study. All subjects are of Chinese descent and inpatients from the Psychiatric Department, the Third Affiliated Hospital of Sun-Yat Sen University. Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition, diagnosis of SZ was established using the Structured Clinical Interview for DSMIV (SCID) performed by a single experienced psychiatrist (WXL) and reviewed by a senior psychiatrist (ZHJB). All patients were experiencing their first-episode psychosis and
1045
receiving atypical antipsychotic monotherapy. During their hospitalization period, antipsychotic treatment was provided according to normal clinical practice with olanzapine (OLZ) up to 30 mg/d, aripiprazole up to 30 mg/d, and quetiapine up to 500 mg/d. If needed, low dose of lorazepam (up to 2 mg/d) is added to ameliorate sleep. To confirm the SZ diagnosis, all patients were followed up for at least 6 months after discharge. Exclusive criteria were as follows: (a) severe physical illness; (b) organic mental diseases, mental retardation, or dementia; (c) pregnancy or lactation (for women patients); and (d) substance abuse within the past 3 months. Assessment of psychopathology and response to antipsychotic treatment were performed by 2 training senior psychiatrists (ZJB and GZY) using the Positive and Negative Syndrome Scale (PANSS). All patients were assessed on admission and reassessed after 4 weeks of treatment. All clinical raters were blind to the rs1344706 genotype. Blood samples were collected in EDTA-containing tubes, and leukocytes were isolated using red blood cell lysis buffer containing 150 mmol/L NH4Cl, 10 mmol/L KHCO3, and 0.1 mmol/L EDTA. Genomic DNA was extracted from the leukocytes using SDS-proteinase K treatment followed by phenol/chloroform extraction. Genomic DNA was amplified by polymerase chain reaction to generate a 443 base pair product spanning rs1344706. Primers were as follows: upper GAATCTAGA GTCATGCAGG and lower CAAGTTATTC TCTAGAGTCC. The polymerase chain reaction products were subjected to direct sequencing, conducted by the Beijing Genomics Institute. Genomic DNA samples were randomly selected for replication. The correspondence between the repeated sequencing and the original sequencing was 100%. The data analyses were performed by a biotechnologist. Software program Finetti (http://ihg2.helmholtzmuenchen.de/cgi-bin/hw/hwa1.pl) was used to test the distribution of genotypes for deviations from HardyWeinberg equilibrium. SPSS for Windows (version 13.0; SPSS Inc, Chicago, IL) was used to compare the clinical and demographic variables between patients with 3 different genotypes. For the baseline characteristics, analysis of variance (ANOVA) for continuous variables and χ 2 test for categorical variables were used to examine the difference among 3 groups. The Fisher exact test was used for
Table 1 Baseline characteristics of patients with SZ with 3 genotypes of ZNF804A rs1344706 polymorphism Genotype
GG, n = 19
GT, n = 41
TT, n = 11
F/χ 2, P value
Sex (M/F) Age (y) Age at first onset (y) Total PNASS score on admission Positive subscore on admission Negative subscore on admission Antipsychotic medication (OLZ/Q/A) a
7/12 22.32 ± 5.91 21.63 ± 5.89 101.26 ± 12.97 24.21 ± 4.13 24.32 ± 7.20 11/2/6
22/19 24.56 ± 6.97 23.96 ± 7.01 101.90 ± 12.30 24.95 ± 4.56 24.02 ± 4.36 21/7/13
6/5 22.64 ± 7.54 21.95 ± 7.60 101.45 ± 9.10 23.55 ± 3.80 26.09 ± 5.94 4/5/2
1.62, .45 0.86, .43 0.92, .40 0.02, .98 0.53, .59 0.62, .54 4.95, .29 a
M/F indicates male/female; Q, quetiapine; A, aripiprazole. a Fisher exact test.
1046
J. Zhang et al. / Comprehensive Psychiatry 53 (2012) 1044–1048
Table 2 Comparison of percentage change in PANSS score of first-episode patients with different genotypes after the 4-week treatment Genotype
GG, n = 19
GT, n = 41
TT, n = 11
F, P value⁎
F, P value⁎⁎
Percentage reduction in total PANSS score Percentage reduction in positive PANSS score Percentage reduction in negative PANSS score
44.47 ± 10.44 54.89 ± 13.76 37.68 ± 15.28
34.8 ± 14.85 38.34 ± 23.98 31.95 ± 16.47
31.55 ± 10.52 40.36 ± 16.07 24.18 ± 8.47
4.608, .013 4.183, .019 2.788, .069
8.724, .004 8.392, .005 3.119, .082
⁎ P value among the 3 groups. ⁎⁎ P value between groups GG and GT + GT.
categorical variables if the expected cell size was less than 5. Treatment response over 4 weeks was presented as percentage reduction in PANSS score (mean ± SD). The relationship between age, age of onset, sex, antipsychotic type, baseline PANSS scores, and treatment response was examined using Pearson correlation coefficient. To investigate the effects of genotype on changes in PANSS, comparison of percentage reduction in PANSS score between genotypic subgroups was carried out using univariate ANOVA. Because a higher percentage of subjects in the non–risk carrier group are prescribed OLZ, which might result in significantly improved response, comparison of percentage reduction in PANSS score in the OLZ group was separately performed to further rule out the impact of medication. Variables significantly correlated with the treatment response were selected as covariates to control for confounding factors in univariate ANOVA. For post hoc comparisons, LSD multiple range test was used. For all of the tests, level of significance was set at P b .05.
negative symptoms subscore (P = .002, .025, and .01, respectively). Subsequently, antipsychotic type was included as a covariate in the association of genotype with changes in total PANSS scores and subscores. The association still remained significant for total PANSS scores and positive subscores (P = .013 and .019, respectively). When the comparison was performed between T carriers and GG homozygotes, a more stronger association was identified in total PNASS score and positive subscores (P = .004 and .005, respectively) (Table 2). As demonstrated in Table 3, in the OLZ group, the direction of changes in total PANSS scores and subscores was consistent with that in the whole group, although not significant (all P N .05), which may attribute to the small sample size. As Fig. 1 showed, these results indicated that the patients with the GG genotype showed substantial improvement on positive symptom response compared with the patients carrying the T allele.
4. Discussion 3. Results There were 19 patients with GG genotype, 11 with TT genotype, and 41 with GT genotype. The genotype distribution was in Hardy-Weinberg equilibrium (P = .152). Baseline demographics and clinical characteristics are presented in Table 1. There were no significant differences between the genotypic subgroups about sex, age, age at first onset, total PNASS score, positive PNASS subscore, negative PNASS subscore, and type of antipsychotic medication (all P N .05). Variables including age, age of onset, sex, and baseline PANSS scores showed no significant correlation with the treatment response. However, antipsychotic type closely correlated with the change in PANSS scores after 4 weeks of treatment. Primary ANOVA showed that the rs1344706 polymorphism in ZNF804A gene was significantly associated with the changes in total PNASS score and positive and
Response to antipsychotic treatment is considered to be a genetically determined phenotype of SZ [1]. Over the past decade, most pharmacogenetic studies have focused on the antipsychotic drug targets, such as the serotonin 5-HT and dopamine D2-like receptors [2,5]. Although association between treatment response and polymorphisms in these candidate genes has been reported by quite a few studies, much of the diversity in treatment response remains unexplained. The successful application of GWAS in research for susceptibility gene for major psychosis has come up with several robust variants [6]. These newly identified risk variants now have become the focus of studies related to etiology and pathophysiology of SZ. In the present study, we found that polymorphism rs1344706 in gene ZNF804A, the first genetic variant achieving genome-wide significance, was associated with positive symptom response to atypical antipsychotics in a
Table 3 Comparison of percentage change in PANSS score between OLZ-treated T carriers and noncarriers after the 4-week treatment Genotype
GG, n = 11
GT + TT n = 25
F, P value
Percentage reduction in total PANSS score Percentage reduction in positive PANSS score Percentage reduction in negative PANSS score
44.24 ± 12.88 54.65 ± 15.20 34.13 ± 26.96
40.84 ± 16.35 45.72 ± 26.81 32.16 ± 18.40
0.369, .547 1.058, .311 0.065, .800
Change in positive PNASS score
J. Zhang et al. / Comprehensive Psychiatry 53 (2012) 1044–1048 80 70 60 50 40 30 20 10 0 GG
GT
TT
Fig. 1. Association between rs1344706 genotypes and changes in positive PANSS scores. P = .016.
group of first-episode patients with SZ. In terms of poor positive symptom response to treatment in patients with SZ with T risk allele, our result is consistent with the finding reported by a recent study that the T risk allele was linked to the number and severity of lifetime overactive episodes [19]. In this study, no significant difference in the baseline PNASS scores was found between 3 patient groups with different genotype. Of note, our patients were young firstepisode patients with SZ. To be hospitalized, they had to be severely affected and, in most cases, manifest conspicuous positive symptoms. Hereby, our inpatient group was not representative for the general Chinese first-episode patients with SZ, and the relationship between polymorphism rs1344706 and the baseline PNASS scores could not be inferred from our study. Nevertheless, the exclusive enrollment of first-episode patients was advantageous for it prevented some confounding factors such as duration of illness and therapeutic history. Because the treatment response is also related to the antipsychotic medication, we added type of antipsychotic as a covariate. The association was reduced but remained significant. In addition, because more than 50% of patients in our study were on OLZ treatment, we performed the same statistical analysis in the OLZ group only, which showed the same direction of changes in total PANSS scores and subscores. Yet, to investigate the association between antipsychotic-specific treatments with polymorphism rs1344706, a larger sample receiving the antipsychotic of interest is needed. To our knowledge, this is the first study to investigate the role of ZNF804A rs1344706 in response to atypical antipsychotic treatment. Atypical antipsychotics are considered to be more effective than typical antipsychotics because of their affinity for both 5-HT and dopamine D2 receptors. The risk allele of rs1344706 has been reported to be associated with increased expression of ZNF804A messenger RNA [12]. Although Williams et al [14] also reported that the relationship between this SNP and expression is not relevant to disease, the involvement of the ZNF804A gene in variation of treatment response could not be ruled out. Signal transduction cascade triggered by these receptors plays an important role in the effectiveness of antipsychotic drugs. ZNF804A is predicted to be a transcription regulator that is
1047
most likely to be involved in the downstream of a certain signaling pathway. Therefore, to investigate whether ZNF804A is involved in signaling pathway mediated by 5HT or dopamine D2 receptors may help to clarify the association between polymorphism rs1344706 and treatment response diversity. This study had several limitations. Firstly, our sample size is relatively small for a genetic study, although only firstepisode patients were enrolled. Secondly, selection bias is inevitable because our sample consisted of inpatients only. Such a sample may not be representative for Chinese firstepisode patient with SZ in general. In summary, our study suggests that ZNF804A rs1344706 polymorphism is associated with positive response to atypical antipsychotic treatment in a group of first-episode patients with SZ. More studies with a larger sample and other ethnic population are needed to confirm this finding. Acknowledgment Funding for this study was provided by the Science and Technology Plan Project (2009B080701080) of Guangdong Province and China National Science Fund for Young Scholars (81000180 and 81101028); the Science and Technology Plan Project of Guangdong Province and China National Science Fund for Young Scholars had no further role in study design; in the collection, analysis, and interpretation of data; in the writing of the report; and in the decision to submit the manuscript for publication. The authors would like to gratefully acknowledge the contributions of all of the physicians, nurses, technicians, and patients who participated in this study. References [1] Arranz MJ, Kerwin RW. Neurotransmitter-related genes and antipsychotic response: pharmacogenetics meets psychiatric treatment. Ann Med 2000;32:128-33. [2] Wang L, Fang C, Zhang A, Du J, Yu L, Ma J, et al. The –1019 C/G polymorphism of the 5-HT(1)A receptor gene is associated with negative symptom response to risperidone treatment in schizophrenia patients. J Psychopharmacol 2008;22:904-9. [3] Tsai SJ, Hong CJ, Yu YW, Lin CH, Song HL, Lai HC, et al. Association study of a functional serotonin transporter gene polymorphism with schizophrenia, psychopathology and clozapine response. Schizophr Res 2000;44:177-81. [4] Lencz T, Robinson DG, Xu K, Ekholm J, Sevy S, Gunduz-Bruce H, et al. DRD2 promoter region variation as a predictor of sustained response to antipsychotic medication in first-episode schizophrenia patients. Am J Psychiatry 2006;163:529-31. [5] Lencz T, Robinson DG, Napolitano B, Sevy S, Kane JM, Goldman D, et al. DRD2 promoter region variation predicts antipsychotic-induced weight gain in first episode schizophrenia. Pharmacogenet Genomics 2010;20:569-72. [6] Williams HJ, Craddock N, Russo G, Hamshere ML, Moskvina V, Dwyer S, et al. Most genome-wide significant susceptibility loci for schizophrenia and bipolar disorder reported to date cross-traditional diagnostic boundaries. Hum Mol Genet 2011;20:387-91. [7] Sullivan PF. The psychiatric GWAS consortium: big science comes to psychiatry. Neuron 2010;68:182-6.
1048
J. Zhang et al. / Comprehensive Psychiatry 53 (2012) 1044–1048
[8] O'Donovan MC, Craddock N, Norton N, Williams H, Peirce T, Moskvina V, et al. Identification of loci associated with schizophrenia by genome-wide association and follow-up. Nat Genet 2008;40: 1053-5. [9] Purcell SM, Wray NR, Stone JL, Visscher PM, O'Donovan MC, Sullivan PF, et al. Common polygenic variation contributes to risk of schizophrenia and bipolar disorder. Nature 2009;460:748-52. [10] Shi J, Levinson DF, Duan J, Sanders AR, Zheng Y, Pe'er I, et al. Common variants on chromosome 6p22.1 are associated with schizophrenia. Nature 2009;460:753-7. [11] Stefansson H, Ophoff RA, Steinberg S, Andreassen OA, Cichon S, Rujescu D, et al. Common variants conferring risk of schizophrenia. Nature 2009;460:744-7. [12] Riley B, Thiselton D, Maher BS, Bigdeli T, Wormley B, McMichael GO, et al. Replication of association between schizophrenia and ZNF804A in the Irish case-control study of schizophrenia sample. Mol Psychiatry 2010;15:29-37. [13] Zhang R, Lu SM, Qiu C, Liu XG, Gao CG, Guo TW, et al. Populationbased and family-based association studies of ZNF804A locus and schizophrenia. Mol Psychiatry 2010.
[14] Williams HJ, Norton N, Dwyer S, Moskvina V, Nikolov I, Carroll L, et al. Fine mapping of ZNF804A and genome-wide significant evidence for its involvement in schizophrenia and bipolar disorder. Mol Psychiatry 2010. [15] Esslinger C, Walter H, Kirsch P, Erk S, Schnell K, Arnold C, et al. Neural mechanisms of a genome-wide supported psychosis variant. Science 2009;324:605. [16] Walters JT, Corvin A, Owen MJ, Williams H, Dragovic M, Quinn EM, et al. Psychosis susceptibility gene ZNF804A and cognitive performance in schizophrenia. Arch Gen Psychiatry 2010;67:692-700. [17] Lencz T, Szeszko PR, DeRosse P, Burdick KE, Bromet EJ, Bilder RM, et al. A schizophrenia risk gene, ZNF804A, influences neuroanatomical and neurocognitive phenotypes. Neuropsychopharmacology 2010;35:2284-91. [18] Donohoe G, Rose E, Frodl T, Morris D, Spoletini I, Adriano F, et al. ZNF804A risk allele is associated with relatively intact gray matter volume in patients with schizophrenia. Neuroimage 2011;54:2132-7. [19] Cummings E, Donohoe G, McDonald C, Dinan TG, O'Neill FA, O'Callaghan E, et al. Clinical symptomatology and the psychosis risk gene ZNF804A. Schizophr Res 2010;122:273-5.