Prolactin concentrations during aripiprazole treatment in relation to sex, plasma drugs concentrations and genetic polymorphisms of dopamine D2 receptor and cytochrome P450 2D6 in Japanese patients with schizophrenia
Correspondence: Kazuo Mihara, MD, PhD, Department of Neuropsychiatry, Faculty of Medicine, University of the Ryukyus, 207 Uehara, Nishihara-cho, Okinawa 903-0215, Japan. Email: firstname.lastname@example.org
The authors investigated the correlation between prolactin concentrations during aripiprazole treatment and various factors, including age, sex, plasma concentrations of aripiprazole and its active metabolite, dehydroaripiprazole, and genetic polymorphisms of dopamine D2 receptor (DRD2) and cytochrome P450(CYP)2D6.
The subjects were 70 inpatients with schizophrenia (36 men and 34 women), receiving fixed doses of aripiprazole (24 mg in 45 cases and 12 mg in 25 cases) for periods of between 2 and 30 weeks. Prolactin concentrations were measured by chemiluminescence immunoassay. Plasma concentrations of aripiprazole and dehydroaripiprazole were measured using liquid chromatography with mass spectrometric detection. The genotypes of Taq1A, −141C Ins/Del DRD2 and CYP2D6 were detected by polymerase chain reaction methods.
Prolactin concentrations were significantly higher in women than in men (8.9 ± 7.5 vs 3.4 ± 3.0 ng/mL, P < 0.001). No correlations were found between prolactin concentrations and plasma concentrations of aripiprazole, dehydroaripiprazole or the sum of the two compounds. Prolactin concentrations were not affected by any polymorphism.
The present study suggests that only sex plays a significant role in prolactin concentrations during aripiprazole treatment.
ARIPIPRAZOLE IS A novel antipsychotic drug with a pharmacological profile of dopamine and 5-HT1A partial agonisms, and 5-HT2A antagonism. The efficacy of aripiprazole has been established for the treatment of psychiatric disorders, especially schizophrenia. The advantages of this drug are characterized by the favorable profile of its fewer adverse effects, i.e., a low potential for extrapyramidal adverse effects, weight gain, hyperlipidemia and sedation compared with other second-generation antipsychotic drugs. Also, aripiprazole rather decreases prolactin concentrations in the treatment of schizophrenia through the intrinsic activity on dopamine receptor in the pituitary. Furthermore, the addition of aripiprazole can resolve hyperprolactinemia induced by other antipsychotic drugs.
Several factors influence prolactin responses to antipsychotics in the treatment of schizophrenia. For instance, female patients with schizophrenia have a higher prolactin response to many antipsychotics than male patients. Plasma concentrations of some antipsychotic drugs are positively related to prolactin concentrations.[8, 9] The dopamine D2 receptor (DRD2) gene has Taq1A and −141C Ins/Del polymorphisms, and the subjects with the A1 allele or without the Del allele have lower DRD2 density than the others. The former is suggested to show higher prolactin response to antipsychotics than the latter.[11-16] Several mutated alleles of cytochromeP450(CYP)2D that affect the CYP2D6 enzyme activity have been reported. It has been reported that patients with two CYP2D6*10 alleles decreasing the enzyme activity have lower prolactin response to perphenazine due to blunted pharmacodynamic tissue sensitivity in the pituitary.[14, 18] However, these factors which possibly influence prolactin concentrations during aripiprazole treatment have not been studied systematically.
Therefore, in the present study, the authors aimed to study the correlation between prolactin concentrations during aripiprazole treatment and various factors, including sex, plasma drug concentrations and the genetic polymorphisms of Taq1A and −141C Ins/Del DRD2 and CYP2D6.
The study included 36 male (age: 38.2 ± 14.1 years as mean ± SD) and 34 female (39.2 ± 16.4 years) Japanese inpatients with schizophrenia who fulfilled the DSM-IV-TR. They were physically healthy without any history of substance abuse, neurological disorder, delirium or dementia and without any significant abnormalities, including clinical laboratory examinations, electrocardiography and electroencephalography. The mean ± SD of bodyweight and duration of illness in total subjects were 59.5 ± 13.7 kg and 138.8 ± 148.4 months, respectively. Among the patients there were no smokers or heavy drinkers. Eleven women were postmenopausal, and 16 were regularly menstruating.
The study was approved by the Ethics Committee of the University of the Ryukyus, and all of the patients gave written informed consent to participate in this study. All the subjects had received aripiprazole for periods of between 2 and 30 weeks. It has been shown that plasma concentrations of aripiprazole and its active metabolite, dehydroaripiprazole, reach steady state by 2 weeks of repeated oral administrations. The daily dose was fixed at 24 mg in 45 cases and 12 mg in 25 cases during the study period, and was given once a day at 12.30 hours. Except for 1–8 mg/day of biperiden (n = 20) and 1–6 mg/day of flunitrazepam (n = 42) the patients received no drugs. The doses of aripiprazole for patients co-administered biperiden and flunitrazepam were 24 mg in 12 cases and 12 mg in eight cases, and 24 mg in 31 cases and 12 mg in 11 cases, respectively. Female patients did not receive oral contraceptives. Blood samples were taken at 08.00 hours.
Plasma prolactin concentrations were measured by chemiluminescence immunoassay (normal range for male or postmenopausal female subjects, 3.6–12.8 ng/mL and fertile female subjects, 6.1–30.5 ng/mL: Abbott, Tokyo, Japan). Plasma concentrations of aripiprazole and dehydroaripiprazole were measured using the liquid chromatography with mass spectrometric detection described by Kubo et al. The lower limit of qualification for both compounds was 0.1 ng/mL using 0.4 mL of plasma, and the interassay coefficient of variation was less than 8% at 0.2 ng/mL for both aripiprazole and dehydroaripiprazole.
DNA was isolated from peripheral leucocytes using QIA-amp DNA Blood Maxi (QIAGEN, Tokyo, Japan). The Taq1A and −141C Ins/Del in the DRD2 gene were determined by the polymerase chain reaction (PCR) methods by Grandy et al. and Arinami et al., respectively. Among the CYP2D locus, CYP2D6*5 and CYP2D6*14, both of which cause absent enzyme activity, and CYP2D6*10 were identified using PCR analyses as described by Steen et al., Kubota et al. and Johansson et al., respectively. Alleles other than the CYP2D6*5, CYP2D6*10 and CYP2D6*14 alleles were defined as wild type (wt). Three mutated alleles (mut) and the wt allele, which were regarded as predominant over other CYP2D6 alleles in the population, were selected based on a review of the frequency of CYP2D6 alleles in the Japanese population.
The unpaired t-test was used to analyze sex difference in prolactin concentrations. Because dehydroaripiprazole has similar pharmacological properties to the parent compound, and the plasma concentration of the metabolite amounts to 40% of that of aripiprazole at steady state, the sum of the two compounds is considered to contribute to the overall antipsychotic efficacy. The Pearson correlation test was performed to examine the correlation of prolactin concentrations with drug concentration and age. Genotype differences were analyzed using either the unpaired t-test or one-way anova. The steady-state plasma concentrations of aripiprazole and dehydroaripiprazole were adjusted by the doses of aripiprazole, and anova, followed by the Scheffé test, was used for comparison of the concentration/dose (C/D) ratios among the CYP2D6 genotype groups. Multiple regression analyses were used to examine the correlations between prolactin concentrations and age, sex, drug concentrations, and the three polymorphisms. The following dummy variables were used for analyses of male = 0, female = 1 for sex, A2/A2 = 0, A1/A1+A1/A2 = 1 for Taq1A, Ins/Del+Del/Del = 0, Ins/Ins = 1 for −141C Ins/Del, and wt/wt+mut/wt = 0, mut/mut = 1 for CYP2D6. A two-tailed P-value of less than 0.05 was regarded as statistically significant. spss 11.0J for Windows (spss, Japan, Tokyo) was used for these statistical analyses. Power analyses were performed with G*Power 3.1.4 (University of Kiel, Kiel, Germany).
The mean ± SD (and ranges) plasma concentrations of prolactin were 3.4 ± 3.0 (0.6–12.8) ng/mL in men and 8.9 ± 7.5 (1.0–28.9) ng/mL in women, respectively. The mean plasma prolactin concentrations were significantly higher in women than in men (P < 0.001).
The mean ± SD (and ranges) plasma concentrations of aripiprazole and dehydroaripiprazole were 231 ± 147 (20–719) ng/mL and 108 ± 68 (10–481) ng/mL, respectively. The values for patients coadministered biperiden and flunitrazepam were 265 ± 147 (75–719) ng/mL and 110 ± 53 (30–194) ng/mL, and 251 ± 140 (20–719) ng/mL and 113 ± 48 (10–192) ng/mL, respectively. There were no significant differences in these concentrations between men and women (P = 0.105 and P = 0.066). No correlations were found between plasma concentrations of aripiprazole (r = −0.048, P = 0.690), dehydroaripiprazole (r = −0.066, P = 0.588), the sum of aripiprazole plus dehydroaripiprazole (r = −0.056, P = 0.648) and plasma prolactin concentrations. No correlations were found between prolactin concentrations and age in total (r = −0.015, P = 0.904), male (r = −0.323, P = 0.054) or female (r = 0.072, P = 0.684) subjects.
The patients had the following genotypes: A1/A1 (n = 14), A1/A2 (n = 20), A2/A2 (n = 36) for Taq1A, Ins/Ins (n = 51), Ins/del (n = 19) for −141C Ins/Del, and wt/wt (n = 25), mut/wt (n = 38), and mut/mut (n = 7) for CYP2D6, respectively. Table 1 shows the mean ± SD plasma prolactin concentrations in the three genetic polymorphisms. No differences were found in prolactin concentration between Taq1A polymorphism (F = 0.521, d.f. = 2, P = 0.596 for total patients, F = 0.506, d.f. = 2, P = 0.608 for men, F = 0.275, d.f. = 2, P = 0.761 for women), between −141C Ins/Del polymorphism (F = 0.001, d.f. = 1, P = 0.971 for total patients, F = 0.615, d.f. = 1, P = 0.438 for men, F = 1.062, d.f. = 1, P = 0.310 for women) and between CYP2D6 polymorphism (F = 0.105, d.f. = 2, P = 0.900 for total patients, F = 1.061, d.f. = 2, P = 0.358 for men, F = 0.032, d.f. = 2, P = 0.969 for women).
Table 1. Prolactin concentrations (ng/mL) among three genetic polymorphisms
|Taq1 A|| || || || || || |
| A1/A1||14||4.6 ± 3.9||8||3.0 ± 2.0||6||6.8 ± 4.9|
| A1/A2||20||6.1 ± 6.4||10||2.8 ± 2.8||10||9.4 ± 7.5|
| A2/A2||36||6.6 ± 6.9||18||3.9 ± 3.5||18||9.3 ± 8.3|
|Power (%)|| ||14|| ||14|| ||9|
|−141C Ins/Del|| || || || || || |
| Ins/Ins||51||6.1 ± 5.9||24||3.7 ± 3.4||27||8.2 ± 6.8|
| Del/Ins||19||6.0 ± 7.3||12||2.8 ± 2.2||7||11.5 ± 9.9|
| Del/Del||0|| ||0|| ||0|| |
|Power (%)|| ||5|| ||14|| ||14|
|CYP2D6|| || || || || || |
| wt/wt||25||5.6 ± 5.6||12||2.5 ± 2.2||13||8.5 ± 6.2|
| mut/wt||38||6.2 ± 6.8||21||4.0 ± 3.5||17||9.0 ± 8.7|
| mut/mut||7||6.7 ± 6.2||3||2.8 ± 1.2||4||9.6 ± 7.1|
|Power (%)|| ||7|| ||24|| ||5|
The mean ± SD for the C/D ratio of aripiprazole in the patients with wt/wt, mut/wt, mut/mut were 9.6 ± 4.3, 13.4 ± 5.3, and 17.7 ± 6.8 ng/mL/mg, respectively, and the values for dehydroaripiprazole were 5.0 ± 1.7, 6.4 ± 2.9, and 5.3 ± 2.0 ng/mL/mg, respectively. The respective values for the sum of aripiprazole plus dehydroaripiprazole were 14.6 ± 5.8, 19.8 ± 7.9, and 23.0 ± 8.6 ng/mL/mg. The mean C/D ratios of aripiprazole were significantly (P = 0.002) higher in the patients with mut/mut than those with wt/wt. The value of the sum of aripiprazole plus dehydroaripiprazole was also significantly (P = 0.028) higher in the patients with mut/mut than those with wt/wt. The values of aripiprazole and the sum of aripiprazole plus dehydroaripiprazole were significantly (P = 0.022) higher in the patients with mut/wt than in those with wt/wt.
Plasma concentrations of aripiprazole and dehydroaripiprazole were excluded from the independent variables in multiple regression analyses, because these have strong multicollinearities (|rs| = 0.933) with plasma concentration of the sum of aripiprazole plus dehydroaripiprazole. Multiple regression analyses, including age, sex, plasma concentration of the sum and the three polymorphisms, showed that plasma prolactin concentrations were significantly correlated with sex in total (P < 0.001) and age in men (P < 0.05) (Table 2). A total of 23 men and 14 women showing lower prolactin concentrations than the reference ranges did not have any specific genotypes (data not shown).
Table 2. Standardized partial regression coefficients and multiple correlation coefficient in multiple regression analyses in total, male and female patients between prolactin concentration and various factors
|Plasma concentrations of aripiprazole plus dehydroaripiprazole (ng/mL)||−0.156||0.163||−0.294|
|DRD2|| || || |
|Taq1 A (the presence of A1 allele)||−0.092||−0.186||−0.008|
|−141C Ins/Del (the absence of Del allele)||0.072||−0.108||0.174|
|CYP2D6 (the absence of wt allele)||0.021||−0.194||−0.028|
|Sex||0.478***|| || |
|Multiple correlation coefficient||0.482**||0.433||0.342|
In this study, women showed significantly higher prolactin concentrations than men during the treatment with aripiprazole. This result is consistent with previous studies regarding not only first-generation antipsychotics, such as, bromperidol and haloperidol, but also second-generation antipsychotics, such as, risperidone and olanzapine. Because estrogen exerts a facilitatory effect on prolactin secretion from the pituitary, high estrogen concentrations in women may be associated with relatively high prolactin concentrations.
No correlations were observed between plasma concentrations of aripiprazole, dehydroaripiprazole, the sum of aripiprazole plus dehydroaripiprazole and prolactin concentrations, suggesting that plasma drug concentrations are not determinants for prolactin concentrations. A study using positron emission tomography has shown that DRD2 occupancy by aripiprazole reaches a plateau of almost complete DRD2 saturation at serum concentrations of approximately 100–150 ng/mL in patients with schizophrenia or schizoaffective disorder. Also, it has been shown that the mean serum aripiprazole concentration in a large clinical sample was above 200 ng/mL. Plasma aripiprazole concentrations exceeded 150 ng/mL in more than half of the patients (n = 44) in this study. This may explain the negative correlation.
Also, no difference in prolactin concentrations was found between CYP2D6 genotypes. This result suggests that prolactin concentrations are not related to the polymorphism. The CYP2D6 is involved in the metabolism of aripiprazole, and the mean C/D ratio of aripiprazole and the sum of aripiprazole plus dehydroaripiprazole increased significantly in accordance with the number of the mutated alleles of CYP2D6, which is consistent with the authors’ previous study. Therefore, it is possible that higher drug concentrations in patients with mutated allele of CYP2D6 may conceal the diminished pharmacodynamic sensitivity in the pituitary caused by the mutated alleles.[14, 18]
Only one study has reported that Taq1A polymorphism was not associated with prolactin concentrations during aripiprazole treatment. In the present study, prolactin concentrations were not affected by TaqA1 or −141C Ins/Del genotypes. These findings, together with the previous report, suggest that these polymorphisms do not influence prolactin concentrations during aripiprazole treatment, which is not in accordance with previous studies using other antipsychotics.[11-16, 18] In this study, the DRD2 occupancy by aripiprazole may be saturated as mentioned above. In addition, aripiprazole has an intrinsic activity of roughly 30% at the DRD2, because of the partial agonism. These pharmacokinetic and pharmacodynamic characteristics of aripiprazole might nullify the difference in the DRD2 density affected by the polymorphisms.
Multiple regression analyses showed that prolactin concentrations were negatively correlated with age in men. However, no statistical correlation was found between prolactin concentration and the age of men in simple regression. The authors have no clear explanation for the inconsistency. The significance may be a chance finding, or it is possible that the finding is reflected by the aging process, because prolactin concentrations tend to fall with age.
There are several limitations in this study. The possibility of committing type II errors cannot be ruled out because of the low power as presented in Table 1. The dose of aripiprazole and the duration of treatment were not unified. The study did not eliminate the influence of hormone levels associated with menstrual cycle phase in premenopausal females on prolactin concentrations. The correlation between prolactin concentrations and clinical response was lacking, although it was reported that hypoprolactinemia was one of the causes of male infertility. Rare mutated alleles for CYP2D6 were not identified. Other genetic polymorphisms which might affect prolactin concentrations, including ankyrin repeat and kinase domain containing 1 gene, were not investigated. Finally, the impact of the inhibitory effect of aripiprazole on prolactin secretion might have been too small to detect significant findings.
In conclusion, the present study suggests that only sex plays a significant role in prolactin concentrations during aripiprazole treatment.
All authors indicate that there is no duality of interest that could be perceived to bias this work. This study was supported by a Grant from the Research Group for Schizophrenia, Japan and by Grants-in-Aid from the Japanese Ministry of Education, Culture, Sports, Science and Technology (#16790696 and #19591366). Authentic aripiprazole and dehydroaripiprazole were kindly provided by Otsuka Pharmaceutical Co, Ltd, Tokyo, Japan. We thank Mr David Webb for his helpful advice.