Hyperprolactinemia in Children and Adolescents with Use of Risperidone: Clinical and Molecular Genetics Aspects

Abstract

Objective:

In children and adolescents treated with risperidone, hyperprolactinemia is a frequent complication that may have clinical repercussions. Several genes have been associated with this occurrence. The aim of this study was to evaluate the frequency of hyperprolactinemia in children and adolescents treated with risperidone, and its associations with clinical and pharmacological data and certain polymorphisms of the following genes: Dopamine receptor D2 (DRD2), 5-hydroxytryptamine (serotonin) receptor 2C (HTR2C), cytochrome P450, family 2, subfamily D, polypeptide 6 (CYP2D6), leptin (LEP), leptin receptor (LEPR), melanocortin 4 receptor (MC4R), and scavenger receptor class B, member 2 (SCARB2).

Methods:

The study included patients using risperidone (8–20 years old) and healthy subjects not exposed to the medication. Psychopathological symptoms, doses, and duration of treatment with risperidone, sex, skin color, body mass index (BMI), use of other psychotropic drugs, and polymorphisms of DRD2, HTR2C, CYP2D6, LEP, LEPR, MC4R, and SCARB2 genes were evaluated.

Results:

There were 120 patients and 197 individuals not exposed to risperidone who were evaluated. Among patients, hyperprolactinemia was found in 79 (65.8%) cases, with no differences regarding sex, skin color, or being in monotherapy with risperidone (26.7% of total patients) or not. The level of prolactin was not correlated, either in case or control groups, with chronological age, bone age, prescribed dose of risperidone, weight-adjusted dose of risperidone, or BMI (p > 0.05), but was negatively correlated with the treatment duration (r = –0.352, p = 0.001 among cases; and r = –0.324, p = 0.039 among controls). There were significant differences in use of risperidone between patients and healthy subjects without the medication in the frequency of the polymorphisms of the DRD2, HTR2C, and LEP genes. Considering both sexes together and also specifically among females, the occurrence of hyperprolactinemia was higher in the presence of the C allele of the rs6318 single nucleotide polymorphisms (SNP) of the HTR2C gene.

Conclusions:

This group of children and adolescents with or without isolated use of risperidone presented with a high frequency of hyperprolactinemia, although asymptomatic, and associated, when considering only females or both sexes together, with being a carrier of the C allele of the rs6318 SNP of the HTR2C gene.

Introduction

Risperidone is increasingly being used in children and adolescents because of its well-established clinical efficacy and lower risk of neurological adverse effects compared with conventional antipsychotics (Troost et al. 2007). Its use is not limited to schizophrenia and other psychotic states, whose onset is very rare before 18 years of age (Correll 2008; Freeman et al. 2009; Ercan et al. 2011; Farmer et al. 2011; Roke et al. 2013). It is also prescribed for the treatment of behavioral problems associated with intellectual disabilities and the autism spectrum, disruptive behavior, and tic disorders, among other conditions (Correll 2008; Freeman et al. 2009; Ercan et al. 2011; Farmer et al. 2011; Roke et al. 2013).

Despite its therapeutic potential, the use of risperidone may cause adverse effects such as hyperprolactinemia (Troost et al. 2007; Zito et al. 2008), which can manifest as galactorrhea, gynecomastia, amenorrhea, sexual dysfunction, infertility, hirsutism, osteopenia, and other conditions (Troost et al. 2007; Roke et al. 2013). More than 70% of children and adolescents of both sexes develop hyperprolactinemia at the baseline and >30% mantain it after 1 year of continuous medication use (Yasui-Furukori et al. 2008; Calarge et al. 2009a; Roke et al. 2009; Cabaleiro et al. 2014). The effects of hyperprolactinemia on sexual and reproductive functions are particularly important in postpubertal patients because they have symptoms, whereas children and prepubertal adolescents with hyperprolactinemia are more often asymptomatic. Delays in growth and sexual maturation, gynecomastia in boys, or hirsutism in girls may have an impact on self-esteem and, therefore, lead to interruption of medication (Troost et al. 2007; Roke et al. 2009). Further, over time, although there are conflicting data, the hyperprolactinemia induced by risperidone can be associated with cardiovascular complications, immune supression, the development of mammary tumors, prolactinomas, and even psychiatric symptoms such as depression, anxiety, hostility, memory problems, and psychotic disorders (La Torre and Falorni 2007; Roke et al. 2009; Ho et al. 2011; Calarge et al. 2014).

The hyperprolactinemia induced by risperidone is mainly derived from the blocking of D2 dopamine receptors (DRD2) in the anterior pituitary gland, which abolishes the inhibitory effect of dopamine on prolactin secretion (Stahl 2009; Correia et al 2010). The allele A-241 G of the DRD2 gene has been associated with a faster response to treatment with antipsychotic drugs in adults with schizophrenia (Lencz and Malhotra 2009). This is consistent with the findings that a lower density of DRD2 may be associated with increased receptor occupancy, resulting in improved response to treatment, but also in higher prolactin concentrations (Mihara et al. 2003; Lencz and Malhotra 2009).

In contrast to dopamine, serotonin (5-hydroxytryptamine) stimulates the secretion of prolactin, and its antagonists reduce prolactin concentrations (La Torre and Falorni 2007). Both 2A and 2C serotonin receptors (5HT2A and 5HT2C, respectively) are involved in the release of prolactin, and agonists of these receptors increase the secretion of this hormone (Van de Kar et al. 2001). The antagonistic effect exerted by risperidone on these receptors inhibits prolactin secretion, thus counteracting the prolactin increase induced by risperidone binding to DRD2 receptors (La Torre and Falorni 2007; Correia et al. 2010).

The hepatic P450 2D6 enzyme cytochrome complex is involved in the metabolism of risperidone and cytochrome P450, family 2, subfamily D, polypeptide 6 (CYP2D6) gene polymorphisms may be associated with adverse effects (Lett et al. 2012). The CYP2D6 gene is highly polymorphic (Lett et al. 2012). The CYP2D6 gene is highly polymorphic. More than 90 allelic variants have been described (Lett et al. 2012). Interindividual variations derived from genetic differences in the risperidone metabolism may be considered in explaining its induced prolactin increase. The active metabolite 9-hydroxy-risperidone, originating from the enzymatic action of the P450 2D6 cytochrome complex on risperidone, has an important role on prolactinemia elevation—even more so than risperidone itself (Knegtering et al. 2005; Troost et al. 2007).

The aims of this study were to assess the frequency of hyperprolactinemia in a sample of children and adolescents taking risperidone to evaluate associations with clinical and pharmacological data and with certain polymorphisms of the DRD2, 5-hydroxytryptamine (serotonin) receptor 2C (HTR2C), CYP2D6, leptin (LEP), leptin receptor (LEPR), melanocortin 4 receptor (MC4R), and scavenger receptor class B, member 2 (SCARB2) genes. The hypothesis of the authors is that certain polymorphisms of the HTR2C, DRD2, and CYP2D6 genes are associated with a higher frequency of hyperprolactinemia caused by risperidone use. The polymorphisms of LEP, LEPR, MC4R, and SCARB2, although not described as being associated with the occurrence of hyperprolactinemia, were also studied because they are described as being possibly associated with obesity and its metabolic complications—aspects that were also studied by the authors of the present study and whose results will be described in a separate article.

Methods

This was a cross-sectional study with a sample composed of patients whose inclusion criteria were: 1) Being between 8 and 20 years old; 2) attending, as patients, the Psychiatric Outpatient Clinic of Children and Adolescents, Department of Medical Psychology and Psychiatry, Clinical Hospital and School of Medical Sciences, State University of Campinas (Unicamp); 3) being in treatment for mental and behavioral disorders in which risperidone was the drug of choice; and 4) having understood, agreed to, and signed (either themselves or their caregivers) an informed consent form approved by the Unicamp Ethics Committee of Research (Form 44199; Certificate of Presentation for Ethics Assessment 04369612.8.0000.5404; June 26, 2012). Exclusion criteria for the same group were: 1) Being overweight or obese (body mass index [BMI] above +1 SD for age and sex) before the beginning of risperidone use; 2) having any diseases or using drugs that are known to predispose to obesity, metabolic syndrome, or hyperprolactinemia; 3) having mental and behavioral disorders caused by other medical conditions; 4) having comorbid mental and behavioral disorders resulting from the use of psychoactive substances; 5) having a diagnosis of severe/profound intellectual disability; 6) having a diagnosis of eating disorder; or 7) in the case of women, being pregnant, lactating, or using hormonal contraception.

A group composed of healthy young adults was also included. It was added with the pupose of evaluating the distribution of the studied polymorphisms in the Brazilian population, which, in contrast to some other studied populations, is more homogeneous and has very mixed ethnic origins. Inclusion criteria for this group were: 1) Being <25 years old; and 2) having understood, agreed to, and signed the informed consent form, approved by the Unicamp Ethics Committee of Research. Exclusion criteria of participants from this group were: 1) having a personal history of serious mental or behavioral disorders or recent use of antipsycotics; 2) currently taking risperidone or other antipsychotics; or 3) in the case of women, being pregnant, lactating, or using hormonal contraception.

From the individuals of the group taking risperidone, at the time of initial evaluation, the following data were evaluated: Gender (male or female), age, skin color (white, black, brown, yellow, indigenous), BMI, psychiatric diagnoses according to International Statistical Classification of Diseases and Related Health Problems, 10th Revision (ICD 10) criteria (World Health Organization 2010), period of time of treatment with risperidone, current dose of risperidone, and use of other psychoactive drugs. In the group not exposed to risperidone, sex, age, and skin color were evaluated. In the group taking risperidone, the BMI (kg/m²) was calculated and transformed into a z score (World Health Organization 2007). Obesity was defined as z scores of at least +2 SD, and overweight was defined as a z score at least +1 SD but <+2 SD for age and sex (World Health Organization 2007). In the group not exposed to risperidone, as all indivuals were >18 years old, overweight was defined as BMI between 25 and 29.99 kg/m² and obesity was denifed as BMI of at least 30 kg/m² (World Health Organization 1989).

Determinations of serum prolactin, thyroid-stimulating hormone (TSH), and free thyroxine (T4) (chemiluminescence) were taken among patients using risperidone after 8 hours of fasting in the Physiology Laboratory, Department of Clinical Pathology, Unicamp (Rdsystems kit; intra- and interessay errors up to 10%). Hyperprolactinemia was defined as values >20 mg/dL in males and 25 mg/dL in females, in the absence of hypothyroidism (Casanueva et al. 2006). The individuals who developed hyperprolactinemia while using risperidone were labeled as “cases,” and those who did not develop the condition when exposed to the medication were “controls.” Dosages of thyroid hormones were performed to exclude patients with hypo- or hyperthyroidism, and the measured values were considered according to the normality range adopted by the kits used in the Human Physiology Laboratory, Clinical Hospital, University of Campinas.

Genomic DNA was extracted from all participants, cases, controls, and healthy individuals not exposed to risperidone. The extraction was taken from total venous blood by the proteinase K technique (Boehringer Mannheim, Germany), lyzed with ethylenediaminetetraacetic (EDTA). The following single nucleotide polymorphisms (SNPs) were determined in the Human Genetics Laboratory of the Molecular Biology and Genetic Engineering Center, Unicamp, by real-time polymerase chain reaction (PCR) using the TaqMan® (Applied Biosystems, Foster City, CA) allelic discrimination assay: HTR2C gene, SNP rs6318 and rs3813929; LEP gene, SNP rs7799039; LEPR gene, SNP rs1137101; DRD2 gene, SNP rs1799978 and rs6277; SCARB2 gene, rs3853188; MC4R gene, SNP rs17782313; and CYP2D6 gene, SNP rs1065852 (allele CYP2D6*10). For the two HTR2C gene SNPs with X-linked inheritance, the allelic determination was also separated by sex in both case and control groups.

All data were stored and analyzed using the Statistical Package for Social Sciences (SPSS) version 22.0 (IBM SPSS Statistics 22.0 INC, Somers, NY). Descriptive data analyses were initially performed. Chi-square tests or Fisher's exact tests, Mann–Whitney, and Kruskal–Wallis tests were then used to: 1) Evaluate the presence or absence of hyperprolactinemia in individuals who were exposed to risperidone, regarding sex, chronological and bone ages, skin color, and nutritional status; 2) assess the prolactin levels among individuals of the case and control groups and the possible association of hyperprolactinemia with concomitant use of other psychiatric drugs; 3) evaluate gender differences regarding any of the studied parameters; 4) determine the effect of the found allele in each studied SNP, regarding the presence or absence of hyperprolactinemia, among patients from the group exposed to risperidone; and 5) evaluate the distribution of each allele of the studied SNP in individuals of the three groups, those who were cases taking risperidone and had hyperprolactinemia, those who were controls taking risperidone and did not have the condition, and those who were not taking risperidone. To evaluate the correlation between prolactin levels with age, total and weight-adjusted doses of risperidone, and period of time of treatment with risperidone, the Spearman correlation test was used. The significance level was 5%. All tests were two tailed.

Results

This study comprised 120 patients taking risperidone, with a mean age of 13.0 ± 3.1 years (95% CI: 12.4–13.5 years) and median of 13 years. Ninety-eight patients (81.7%) were males. Hyperprolactinemia was found in 79 (65.8%) cases, whereas the other 41 (34.2%) were controls without prolactin elevation after exposure to risperidone. For pairing the SNP distribution with those of the patients, another 197 individuals not exposed to the medication had blood samples collected for DNA analysis.

Among patients taking risperidone, comparisons between cases and controls, according to prolactin levels and clinical, pharmacological, and radiological parameters, as well as clinical features of individuals not exposed to risperidone, are shown in Table 1.

Table 1.

Comparison between Groups Exposed and Not Exposed to Risperidone, According to Clinical, Pharmacological, and Radiological Parameters, and Prolactin Concentrations

	Patients taking risperidone
	Cases - with hyperprolactinemia n = 79	Controls - without hyperprolactinemia n = 41	p value	Not taking risperidone n = 197
Sex
Female	14–17.7%	8–19.5%	0.810^#	119–60.4%
Male	65–82.3%	33–80.5%		78–39.6%
Skin color
Yellow	1–1.3%	–		0.214^###	17–8.6%
White	47–59.5%	32–78.0%	174–88.3%
Black	9–11.4%	2–4.9%	1–0.5%
Mixed	22–27.8%	7–17.1%	5–2.5%
Pubertal status^a
Prepubescent	14–17.7%	12–29.3%	0.145^#		–
Pubescent/postpubescent	65–82.3%	29–70.7%		197–100%
Nutritional status
Eutrophic	60–75.9%	23–56.1%		0.029^##	154–78.2%
Overweight	15–19.0%	17–41.5%	37–18.8%
Obese	4–5.1%	1–2.4%	6–3.0%
Chronological age (years)
Mean ± SD	13.2 ± 3.2	12.5 ± 3.0	0.216^§		23.6 ± 1.9
Median	13.0	12.1		23.5
CI 95%	12.5–13.9	11.5–13.4		23.3–23.8
Minimum–Maximum	8.0–20.5	8.1–20.9		18.5–24.9
Bone age (years)
Mean ± SD	13.2 ± 3.1	12.7 ± 3.1		0.349^§	–
Median	13.2	12.8
CI 95%	12.5–13.9	11.7–13.7
Minimum–Maximum	6.6–18.3	7.1–18
Period of time since the beginning of treatment with risperidone (months)
Mean ± SD	23.4 ± 28.6	30.9 ± 23.9	0.026^§		–
Median	11.9	27.7
CI 95%	16.9–29.8	23.4–38.5
Minimum–Maximum	0.1–143.0	0.1–85.9
Prescribed dose of risperidone (mg/day)
Mean ± SD	2.2 ± 1.3	1.9 ± 1.2		0.248^§	–
Median	2.0	2.0
CI 95%	1.9–2.5	1.5–2.3
Minimum–Maximum	0.25–6.0	0.5–5.0
Weight-adjusted dose of risperidone (mg/kg/day)
Mean ± SD	0.04 ± 0.03	0.04 ± 0.03	0.283^§		–
Median	0.04	0.03
CI 95%	0.04–0.05	0.03–0.05
Minimum–Maximum	0.01–0.12	0.01–0.10
Prolactin concentration (mg/dL)
Mean + SD	42.90 ± 17.9	12.0 ± 6.3
Median	38.3	11.1	0.0001^§	–
CI 95%	38.9–47.0	10.0–13.9
Minimum–Maximum	20.60–99.50	1.5–24.6

Pubertal stage estimated from the bone age, considering as cutoff the age of 10 years.

Chi-square test (Degree of freedom = 1); ^##Chi-square test (Degrees of freedom = 2); ^###Chi-square test (Degrees of freedom = 3); ^§Mann–Whitney test.

In Table 2, comparisons of prolactin levels are shown, analyzed as continuous variables according to clinical, pharmacological, and radiological profiles between cases and controls. There was a high correlation between the chronological age and the bone age (r = 0.921; p = 0.0001) of the participants.

Table 2.

Distributions of Prolactin Levels of Patients Separated According to Clinical, Radiological, and Pharmacological Profiles of Cases and Controls

	Cases			Controls
	n	Prolactin levels in mg/dL (mean ± SD)	p value	n	Prolactin levels in mg/dL (mean ± SD)	p value
Sex
Female	14	50.6 ± 21.0	0.079^#	8	17.0 ± 7.6	0.022^#
Male	65	41.3 ± 16.9		33	10.8 ± 5.4
Skin color
Yellow	1	–		0.184^§	–		–	0.624^§
White	47	41.9 ± 17.7	32		11.8 ± 6.8
Black	9	39.2 ± 22.4	2		9.5 ± 2.3
Mixed	22	47.4 ± 16.8	7		13.4 ± 4.8
Pubertal status^a
Prepubescent	14	43.1 ± 17.5	0.980^#		12	11.6 ± 5.6	0.854^#
Pubescent/postpubescent	65	42.9 ± 18.2		29	12.1 ± 6.7
Nutritional status
Eutrophic	60	42.2 ± 17.9		0.369^§	23	12.7 ± 5.8	0.205^§
Overweight	15	44.2 ± 18.4	17		11.6 ± 6.7
Obese	4	33.5 ± 11.9	1		1.5
Monotherapy
Yes	22	44.2 ± 18.4	0.710^#		10	11.6 ± 5.8	0.777^#
No	57	42.5 ± 17.9		31	12.1 ± 6.5
Number of different associated psychotropics, when not in monotherapy
One	30	42.3 ± 13.1		0.123^§	19	11.7 ± 5.5	0.611^§
Two	21	38.9 ± 20.6	9		11.9 ± 7.3
Three	5	59.7 ± 26.0	1		20.3
Four	1	37.5	2		12.2 ± 15
Period of time since the beginning of treatment with risperidone
≤12 months	41	48.3 ± 20.4	0.010^#		11	14.9 ± 4.6	0.043^#
>12 months	38	37.2 ± 12.8		30	10.9 ± 6.5
Prescribed dose of risperidone
≤2 mg/day	52	43.0 ± 17.4		0.804^#	30	11.6 ± 5.9	0.571^#
>2 mg/day	27	42.7 ± 19.3	11		12.9 ± 7.6
Weight-adjusted dose of risperidone
≤0.04 mg/kg/day	43	45.8 ± 20.3	0.254^#		23	12.8 ± 6.0	0.423^#
>0.04 mg/kg/day	36	39.6 ± 14.2		18	10.9 ± 6.7

Pubertal stage estimated from the bone age, considering as cutoff the age of 10 years.

Mann–Whitney test.

Kruskal–Wallis test.

The main groups of psychopathological diagnoses by the time of sample collections were: Conduct disorders (F91) and mixed disorders of emotions and conduct (F92) in 58 (48.3%) individuals; hyperkinetic disorders (F90) in 47 (39.2%); mild (F70) or moderate (F71) mental retardation, with significant impairment of behavior requiring attention or treatment (F7x.1), or with other impairments of behavior (F7x.8) in 36 (30%); depressive disorders (F32, F33, F34, F38, F39) in 33 (27.5%); pervasive developmental disorders (F84) in 26 (21.7%); emotional disorders with onset specific to childhood (F93) or neurotic, stress-related, and somatoform disorders (F40–F48) in 21 (17.5%); specific developmental disorders of speech and language (F80) or of scholastic skills (F81) in 15 (12.5%); and schizophrenia (F20) and schizotypal disorder (F21) in 11 (9.2%).

None of the patients with hyperprolactinemia had the clinical symptoms of the condition, such as galactorrhea or gynecomastia. There were no patients diagnosed with hypo- or hyperthyroidism.

In 32 samples (26.7%), risperidone was being used as monotherapy by the time of the evaluation. In 49 samples (40.8%), there was a psychotropic drug used in combination with risperidone; in 30 samples (25%), there were two psychiatric drugs; in 6 (5%), three psychiatric drugs; and in 3 (2.5%), four other psychotropics, without any statistical differences between case and control groups of patients (χ² ₍₃₎ = 4.610; p = 0.203). Among those who were not on monotherapy by the time of the assessment, psychiatric medications used in combination with risperidone were the following: Antidepressants in 53 (60.2%) samples; psychostimulants in 27 (30.7%); clonidine in 19 (21.6%); anticonvulsants in 11 (12.5%); benzodiazepines in 8 (9.1%); nonbenzodiazepine sedative drugs (promethazine, levomepromazine, or periciazine) in 7 (8%); other atypical antipsychotics in 6 (6.8%); lithium carbonate in 6 (6.8%); and biperiden in 1 (1.1%). There was no statistically significant difference in the incidence of hyperprolactinemia between the groups that were or were not taking risperidone as monotherapy (χ² = 0.165, p = 0.685) either regarding the different classes of psychotropic drugs used in combination with risperidone, when present (antidepressants [χ² = 0.002; p = 0.967]; psychostimulants [χ² = 0.669; p = 0.413]; clonidine [χ² = 0.072; p = 0.789]; anticonvulsivants [p = 0.747, Fisher's exact test]; lithium carbonate [p = 0.663, Fisher's exact test]; benzodiazepines [p = 1.000, Fisher's exact test]; nonbenzodiazepine sedative drugs [p = 0.689, Fisher's exact test]; other antipsychotics [p = 0.663, Fisher's exact test]; and biperiden [p = 0.342, Fisher's exact test]).

In both case and control groups, the prolactin levels did not show significant correlations with chronological age (p = 0.166 and p = 0.246, respectively), with bone age (p = 0.703 and p = 0.178, respectively), with prescribed dose of risperidone (p = 0.794 and p = 0.765, respectively), with weight-adjusted dose of risperidone (p = 0.531 and p = 0.427, respectively), or with BMI (p = 0.477 and p = 0.292, respectively). There was negative correlation in both case and control groups of patients with the period of time of treatment with risperidone (r = –0.352; p = 0.001 and r = –0.324; p = 0.039, respectively). Hyperprolactinemia occurred more often (χ² = 6.908; p = 0.009) among cases who were within 12 months of risperidone use (41–78.8%, samples) than in those who were on risperidone therapy for >12 months (38–55.9%). When, in the association tests, other cutoff points for period of time of treatment were considered (1, 2, 3, 6, 9, and 24 months), there were also statistically significant associations between patients who had had up to 9 months of risperidone use (34–79.1%) and those who had used the drug for >9 months (45–58.4%) (χ² = 5.220; p = 0.022); and between patients who were within 24 months of risperidone use (52–75.4%) and those who had used the medication for >24 months (27–52.9%) (χ² = 6.554; p = 0.010).

All clinical and laboratory parameters considered in this study were compared in both sexes. Differences were present on the following parameters: Frequency of diagnosis of depressive disorders, present in 11 (50%) females and in 22 (22.7%) males (χ² = 6.678; p = 0.010); and prescription of other psychiatric drugs in association with risperidone, which occurred in 68 (69.4%) male patients and in 20 (90.9%) female patients (χ² = 4.355; p = 0.039). Among females, the coadministration of antidepressants was more frequent, prescribed for 16 (72.7%) patients, whereas in males these were prescribed for 37 (37.8%) (χ² = 8.911; p = 0.003). Benzodiazepines were prescribed for 4 (4.1%) male patients and for 4 (18.2%) female patients (p = 0.037, Fisher's exact test). Regarding the sex distribution of the studied SNP, in an analysis of all patients taking risperidone, both cases and controls, differences were found only in both SNP of the HTR2C gene (p = 0.0001), which is linked to the X chromosome.

The genotypic distributions of the studied SNP between the groups taking risperidone, both cases and controls, and those not taking risperidone are shown in Table 3. Although, when comparing both sexes together, the absence of the G allele of the rs6318 SNP of the HTR2C gene was significantly more frequent in patients taking risperidone than in individuals not exposed to the medication, and the C/G genotype was less frequent, and as this gene is X-linked and heterozygous subjects are exclusively females, it was necessary to separate the analysis by sex. In females, the genotype C/C (i.e., not carrying the G allele) of the rs6318 SNP was significantly more common in those taking risperidone than in those not exposed to the medication (p = 9.671; Fisher's exact test = 0.009), but the C/G and G/G genotypes were not less frequent. In males, in which individuals are hemizygous for the HTR2C gene, there was no statistically significant difference between cases and controls.

Table 3.

Comparison of the Groups that Used (n = 120 – both Cases and Controls) and that Did Not Use (n = 197) Risperidone According to the Distribution of the SNP Alleles of the HTR2C, LEP, LEPR, DRD2, SCARB2, MC4R, and CYP2D6 Genes

Genotype	Taking risperidone (%)	Not taking risperidone (%)	χ²	p
HTR2C rs6318
C/– or C/C	26 (22)	19 (10)	11.693	0.003
C/G	7 (6)	26 (13)
G/– or G/G	87 (72)	152 (77)
HTR2C rs3813929
C/– or C/C	101 (84)	150 (76)			11.264	0.004
C/T	5 (4)	32 (16)
T/– or T/T	14 (12)	15 (8)
LEP rs7799039
A/A	16 (13)	48 (24)	7.295	0.026
A/G	58 (49)	71 (36)
G/G	46 (38)	78 (40)
LEPR rs1137101
A/A	34 (28)	51 (26)			0.235	0.889
A/G	59 (49)	101 (52)
G/G	27 (23)	45 (23)
DRD2 rs1799978
A/A	92 (77)	174 (88)	7.508	0.006
A/G	28 (23)	23 (12)
G/G	0 (0)	0 (0)
DRD2 rs6277
C/C	65 (54)	58 (29)			25.490	0.0001
C/T	48 (40)	95 (48)
T/T	7 (6)	44 (22)
SCARB2 rs3853188
A/A	103 (86)	170 (86)	2.805	0.246
A/C	17 (14)	23 (12)
C/C	0 (0)	4 (2)
MC4R rs17782313
C/C	7 (6)	8 (4)			1.037	0.595
C/T	32 (27)	61 (31)
T/T	81 (67)	128 (65)
CYP2D6 rs1065852
C/C	76 (63)	119 (60)	1.575	0.455
C/T	37 (31)	71 (36)
T/T	7 (6)	7 (4)

p value is two-tailed.

rs = reference SNP identification number.

Regarding the rs3813939 SNP of the HTR2C gene, when comparing both sexes together, the C/T genotype was significantly less frequent among the group taking risperidone than among those not exposed to the medication. However, as for the rs6318 SNP, it was necessary to separate the analysis by sex. There were no differences found in the allelic distributions between the groups, either among females or among males.

The A/A genotype of the rs7799039 SNP of the gene LEP was significantly less frequent among those taking risperidone than among those not exposed to the medication and the A/G genotype was more frequent. The A/G genotype of the rs1799978 SNP of the DRD2 gene was significantly more frequent among those taking risperidone than among those not exposed to the medication and the A/A genotype was less frequent. The T/T genotype of the rs6277 SNP of the DRD2 gene was significantly less frequent among those taking risperidone than among those not taking the drug, and the C/C genotype was more frequent.

Differences were observed regarding the distribution of the rs6318 SNP of the HTR2C gene between patients from case and control groups (χ² ₍₂₎ = 7.798; p = 0.020), with this higher frequency in the presence of the C allele (Table 4). As this is a gene located in the X chromosome, the analyses of associations were also separated by sex. The association remained among female patients (χ² ₍₂₎ = 7.019; p = 0.030) but not among male patients (χ² = 1.522; p = 0.217). There were no statistically significant association between cases and controls regarding the allelic distribution of the rs3813929 SNP of the HTR2C gene or of any other of the studied genes, either when cases were considered as a whole (Table 4) or when they were separated according to sex, age, skin color, or whether they were or were not taking risperidone as monotherapy.

Table 4.

Distribution of the SNP Alleles, among Patients Exposed to Risperidone, in Case and Control Groups, According to the Presence or Absence of Hyperprolactinemia (n = 120)

	Genotype			χ²	p
HTR2C rs6318
	C/– or C/C	C/G	G/– or G/G
Cases	22	1	56	7.798	0.020
Controls	5	4	32
HTR2C rs3813929
	C/– or C/C	C/T	T/– or T/T
Cases	66	3	10	0.281	0.869
Controls	35	2	4
LEP rs7799039
	A/A	A/G	G/G
Cases	12	38	29	0.759^*	0.684
Controls	4	20	17
LEPR rs1137101
	A/A	A/G	G/G
Cases	24	37	18	0.606^*	0.739
Controls	10	22	9
DRD2 rs1799978
	A/A	A/G	G/G
Cases	61	18	–	0.241^**	0.624
Controls	30	11	–
DRD2 rs6277
	C/C	C/T	T/T
Cases	45	29	5	1.057^*	0.589
Controls	20	19	2
SCARB2 rs3853188
	A/A	A/C	C/C
Cases	66	13	–	0.996^**	0.318
Controls	37	4	–
MC4R rs17782313
	C/C	C/T	T/T
Cases	5	20	54	0.281^*	0.869
Controls	2	12	27
CYP2D6 rs1065852
	C/C	C/T	T/T
Cases	51	24	4	0.252^*	0.882
Controls	26	12	3

p value is two-tailed.

rs = reference SNP identification number.

Discussion

As the authors had hypothesized, carrying the C allele of the rs6318 SNP of the HTR2C gene was associated with the occurrence of hyperprolactinemia upon exposure to risperidone. Contrary to the assumptions of the authors, however, none of the other surveyed SNPs was associated with higher incidence of hyperprolactinemia.

The results of this study show a high frequency of hyperprolactinemia among children and adolescents taking risperidone, whether as monotherapy or not. Doses of 0.5–2 mg/day of risperidone were recommended for the age group in the studied sample (Stahl 2009). Some adolescents were taking higher doses (2–8 mg/day), similar to those prescribed for adults, because they could tolerate them (Roke et al. 2009; Stahl 2009; Pandina et al. 2012).

Despite the high occurrence of hyperprolactinemia, it has occurred at concentrations usually not associated with symptoms and not in values observed in tumors, in which the prolactin levels are usually <100 mg/dL (Casanueva et al. 2006). In none of the cases were clinical repercussions of hyperprolactinemia, such as galactorrhea or gynecomastia, found. However, concerns remain, as it is known that there are potential complications of sustained hyperprolactinemia, such as induction of osteopenia and osteoporosis, which, because of their insidious nature, are not evident in the short term (Roke et al. 2009). There is also an increased risk that hyperprolactinemia, even if asymptomatic, may induce breast cancer and pituitary adenomas, and, although it is still unclear, it has a possible association with prostate cancer (Besnard et al. 2014).

Some symptoms of hyperprolactinemia may not be actively searched for by the physician, as is the case for sexual complaints (Roke et al. 2009). Loss of libido and erectile/ejaculatory dysfunction, orgasmic dysfunction, and vaginal dryness, for example, are complaints that are not part of most routine anamneses of many child and adolescent psychiatrists or pediatricians (Roke et al. 2009).

Certain behavioral changes, moreover, are described as possibly resulting from hypogonadism-induced hyperprolactinemia, especially in females (Tsigkaropoulou et al. 2012). Symptoms include hostility, depression, anxiety, and even psychotic states, among others. It is difficult, however, to assess the specific frequency of the occurrence of psychiatric symptoms caused by the hyperprolactinemia effect, as they can overlap those from the primary psychiatric disorder (Tsigkaropoulou et al. 2012). As for symptoms related to aging and lack of sexual desire, medication-related psychiatric symptoms are not always actively investigated during the consultations, which poses an additional difficulty for their recognition when they result from hyperprolactinemia (Roke et al. 2009). Nevertheless, both sexual and psychiatric symptoms may constitute risk factors for dropping out of treatment (Ho et al. 2011).

Most of the patients evaluated in this study were male adolescents. Considering that, in this sex and age group, risperidone is mainly used for the pharmacological treatment of disruptive disorders linked to impulsive and aggressive behaviors, the distribution of the sexes evaluated in this outpatient sample is in accordance with epidemiologic expectations, because these clusters of symptoms are more common among males (Herpertz-Dahlmann et al. 2013). Depressive and anxiety disorders, in turn, are more prevalent among female adolescents (Herpertz-Dahlmann et al. 2013). Although risperidone does not have antidepressant properties, the fact that female patients in this study have a higher frequency of diagnoses of depressive disorders than males is logical because, among females, impulsive–aggressive disorders, for whose pharmacological treatment risperidone was prescribed, have mostly occurred concomitantly with depressive and anxiety disorders. Because of this co-occurrence, there was also an association with psychiatric drugs among females, especially antidepressants and benzodiazepines.

Among those taking risperidone, either cases or controls, there were no correlations between prolactin levels and chronological age, bone age, BMI, total prescribed dose of the medication, or weight-adjusted dose. In both groups, however, negative correlations with period of time since the beginning of the treatment with risperidone were found.

Among the studied SNPs, there were only verified differences in the frequencies of hyperprolactinemia regarding the allelic distribution of the rs6318 SNP of the HTR2C gene when considering both sexes together and when female patients were considered independently. When only male patients were considered when comparing the distribution of hyperprolactinemia between the hemizigous C/– and the hemizigous G/–, the association was not seen.

The role of dopamine in blocking prolactin secretion is well documented (Yasui-Furukori et al. 2008; Roke et al. 2013). It is also known that, in contrast, serotonin stimulates the secretion of prolactin (Yasui-Furukori et al. 2008; Roke et al. 2013) and the cytochrome P450 2D6 set plays an important role in the metabolism of risperidone, which in turn has an influence on prolactin levels (Zhang and Malhotra 2011), although two previous studies have not shown associations between changes in prolactin levels induced by antipsychotics and polymorphisms related to dopamine (Roke et al. 2013; Yasui-Furukori et al. 2008 ). Correia et al. found, in a sample of 45 patients using risperidone for up to 1 year, associations between the presence of the C allele of the rs6318 SNP in the HTR2C gene (c.68G > C [p.C33S]) and more frequent occurrence of hyperprolactinemia (Correia et al. 2010).

There were no associations between hyperprolactinemia frequencies with sex, skin color, or whether or not the risperidone was the only prescribed psychotropic medication. The skin color variable was used in this study in an attempt to evaluate whether there were any differences between the levels of prolactin within groups with supposedly more homogeneous genetic characteristics. It is known, however, that skin color does not accurately reflect the membership of a given patient to a specific ethnic group, mainly in the Brazilian population, which has very mixed ethnic origins and, therefore, does not have the uniformity of genetic patterns observed in patterns of populations that are more homogeneous from some other countries (Calarge et al. 2009b; Roke et al. 2013). Regarding the SNP of the DRD2 and CYP2D6 genes, unlike in reports in the literature (Roke et al. 2013; Calarge et al. 2009b), in the present study, they were not associated with hyperprolactinemia. It is possible that ethnic differences in the composition of the samples, more genetically homogeneous in studies with positive associations (Aklillu et al. 2007), may account for their findings, such as in a study with only non-Hispanic Caucasian children, who may have different densities of D2 receptors in the caudate nucleus (Calarge et al. 2009b; Roke et al. 2013).

Differences between the distributions of the two polymorphisms of the DRD2 gene, the two polymorphisms of the HTR2C gene, and the rs7799039 SNP of the LEP gene were found between the group using risperidone and the one without exposure to the medication. Although several studies found associations between polymorphisms of the DRD2, CYP2D6, and HTR2C genes, with psychopathological symptoms and/or therapeutic response to antipsychotics (Alenius et al. 2008; Adkins et al. 2011; Lett et al. 2012), there is little information on changes in prolactin levels regarding different polymorphisms of these genes and the other ones evaluated in this study (Roke et al. 2013). The finding of difference in the allelic distributions of the LEP gene SNP between individuals using and those not using risperidone was fortuitous, demanding additional investigation. However, regarding the DRD2 gene and, in females, the HTR2C gene, it is possible that the differences are related to the occurrence of specific psychopathological symptoms, and future studies of this study group should evaluate associations between these genes and psychiatric diagnoses and cluster of symptoms.

Because of the logistical difficulty in obtaining a sample of individuals without exposure to risperidone, without psychiatric disorders, and with the same age distribution as those in the group that used the medication, the group used for the evaluation of the SNPs' distribution was composed predominantly of older participants and females, in whom it is known that the levels of prolactinemia are higher. This precluded making comparisons of prolactin levels between subjects exposed and not exposed to risperidone use. The present findings, however, reinforce the results of other studies on the subject, such as those of Roke and colleagues, who, in a descriptive review of 20 studies on prolactin levels among pediatric patients using risperidone, found a frequency of 61.7% of hyperprolactinemia in a total of 1390 subjects, with a sample sex composition very similar to that of the present study; that is, also with a predominance of males (Roke et al. 2009). Although the average prescribed dose of risperidone in the studies evaluated by Roke et al. was slightly lower (1.6 mg/day, compared with 1.9 mg/day in this study) (Roke et al. 2009), the average age of subjects of their reviewed studies was also younger (9.7 years old, compared with 13.1 years old in this study).

Although some anticonvulsants such as valproic acid and carbamazepine may interfere with serum prolactin levels (Melis et al. 1982; Reis et al. 2013), few patients used either of them (9.2–11%) and there were no differences between the groups that were or were not exposed to them regarding the occurrence of hyperprolactinemia.

Unlike other adverse effects of risperidone, such as weight gain and metabolic changes, hyperprolactinemia appears to be more influenced by the serum concentrations of the active metabolite 9-hydroxyrisperidone than by the risperidone itself, so that subjects with CYP2D6 variants that render them fast and ultrarapid metabolizers tend to be more prone to this complication because they generate the active metabolite more quickly, and 9-hydroxyrisperidone has a long half-life of ∼21 hours (Troost et al. 2007; Roke et al. 2013). In addition, 9-hydroxyrisperidone is only partially metabolized by the liver P450 2D6 cytochrome enzyme set and, compared with risperidone, it is less lipophilic; therefore, it crosses the blood–brain barrier less and has a higher plasma concentration than that of the original drug (Troost et al. 2007). The pituitary gland, in turn, is located outside the blood–brain barrier (Troost et al. 2007). However, this study found no association between prolactin concentrations and the studied polymorphism of CYP2D6, and larger studies and reviews of different age groups are required to conclude about the actual influence of the CYP2D6 in prolactin concentrations. Studies in adults should not be simply extrapolated to pediatric patients, because children and adolescents have specific metabolic characteristics that can bias the interpretation, such as age-dependent differences in the body liquid composition and the distribution of serum proteins (Kearns and Reed 1989; Roke et al. 2013).

There were no sex differences in the frequency of hyperprolactinemia. The reports in the literature on such an association are inconsistent (Saito et al. 2004). There have been studies that found more hyperprolactinemia in males (Aman et al. 2002), others that found more in females (Duval et al. 2008; Roke et al. 2009; Pandina et al. 2012; Besnard et al. 2014; Cabaleiro et al. 2014), and yet others, such as this study, that found no significant differences between the sexes (Saito et al. 2004). However, the present report has shown a considerable discrepancy between the sexes, with a ratio of approximately four males to one female in the group of children and adolescents taking risperidone. The predominance of males in this research is justified because it is not based on a purposive sample, otherwise reflecting the epidemiological context of an outpatient psychiatric service of children and adolescents, where there is a high prevalence of disruptive and externalizing mental disorders (Calarge et al. 2009a). These psychopathological conditions, more prevalent in boys, are conditions for which risperidone is commonly prescribed (Calarge et al. 2009a; Ercan et al. 2011; Farmer et al. 2011; Stadler et al. 2011).

There has also been no association between hyperprolactinemia and skin color. There are few studies that describe skin color being associated with prolactin changes in individuals taking antipsychotics. The Brazilian population is intensely mixed, and the skin color does not reflect substantial ethnic differences in genotypic terms (Gonçalves et al. 2007). However, it is possible that, in studies with more ethnically heterogeneous populations, differences could be verified, as there are differences in the distribution of the risperidone metabolism rate profiles among Caucasian, African, and Asian groups, as they typically have distinct distributions of the CYP2D6 genetic polymorphisms (Troost et al. 2007; Lett et al. 2012; Choong et al. 2013).

Higher dose of risperidone is described as a factor associated with higher concentrations of prolactin in some studies (Duval et al. 2008; Calarge et al. 2009a; Roke et al. 2009; Roke et al. 2013), as well as older age; however, although they are associated in some studies (Duval et al. 2008; Calarge et al. 2009a; Roke et al. 2009), they are not in all of them (Troost et al. 2007). In subjects in the pubertal period, higher physiological concentrations of prolactinemia are expected than in children (Roke et al. 2009). It is suggested, however, that children and adolescents may be more susceptible to risperidone-induced hyperprolactinemia than adults (Saito et al. 2004). In this study, however, there were no observed correlations of prolactin levels with the prescribed dose of risperidone, with weight-adjusted dose, with BMI, or with chronological or bone age.

The negative correlation between prolactinemia and period of time of risperidone use, although weak, is consistent with studies conducted both with children and with adults, as well as is the higher incidence of hyperprolactinemia in cases within the first 12 months of risperidone use (Anderson et al. 2007; Troost et al. 2007). Although there was less research found regarding the evaluation of prolactin levels beyond 6 months of treatment with risperidone, there appears to be a gradual, but incomplete, trend for a decrease in these values (Troost et al. 2007). This finding is consistent with the present study, because it included more cases with more lasting medication usage time than did other similar studies (Troost et al. 2007; Roke et al. 2009). It is possible that the negative correlation could be more intense and the temporal association tests with earlier cutoff points for periods of use of risperidone were statistically significant if more individuals in the early stages of treatment could be included, reflecting the findings from other studies that higher prolactin peaks would occur more in the first weeks of treatment with risperidone (Anderson et al. 2007; Troost et al. 2007).

Although most children and adolescents in this study did not use risperidone as monotherapy, because the association of different clusters of psychopathological symptoms in the same individual is common, and also because sometimes it may be necessary to use other drugs to enhance the actions of risperidone, there were no differences between the occurrence of hyperprolactinemia and association with other psychiatric drugs. This was observed despite antidepressants, which also have the potential to induce hyperprolactinemia (Trenque et al. 2011), being the class of drugs most commonly prescribed together with risperidone in this study. Clonidine was the second most frequently coadministered drug to individuals who were using risperidone as monotherapy. Although it is primarily an antihypertensive drug, its α-2 agonist properties also justify its use in psychopharmacology, and it is commonly prescribed for the treatment of hyperkinetic and anxious syndromes, as well as tic disorders (Stahl 2009). Hyperprolactinemia is not a commonly described side effect of this medication (Stahl 2009).

Among the atypical antipsychotics, risperidone is the one most associated with prolactin increases (Besnard et al. 2014), which may explain the lack of differences in cases in which other antipsychotic drugs were coadministered, in addition to the small number of subjects who were taking more than one of these agents when assessed. The antipsychotics that, in addition to risperidone, are commonly associated with hyperprolactinemia are the first-generation ones, such as haloperidol (Stahl 2008; Besnard et al. 2014), but it was not prescribed together with risperidone for any of the subjects of this study.

Among the limitations of this study, it can be mentioned that many patients had been prescribed drug combinations and not risperidone monotherapy. Methylphenidate has a prodopaminergic action and many antidepressants have proserotonergic actions (Stahl 2008, 2009). However, both actions are contrary to the increase in prolactin and, if they have any clinical significance, they would lead one to underestimate the increases that would occur if risperidone were used as monotherapy. In addition, this study also did not evaluate the incidence of side effects related to hyperprolactinemia. In large trials, however, there are a considerable number of patients who have no clinical effects from hyperprolactinemia (Kleinberg et al. 1999; Yasui-Furukori et al. 2008), perhaps because of the increase in prolactinemia being neither very high nor lasting. However, monitoring for the presence of possible clinical repercussions of hyperprolactinemia, given its high frequency, remains recommended for all children and adolescents who use risperidone (Besnard et al. 2014). Further, although the cross-sectional design of this research did not allow the assessment of pretreatment prolactin concentrations and, therefore, cannot establish causal relationships, cross-sectional designs are not inappropriate in genetic studies, because genes determine predispositions.

Conclusions

In summary, the results found show a high frequency of hyperprolactinemia in children and adolescents who use risperidone, despite doses considered therapeutic for their age. Prolactin concentrations were not correlated with chronological or bone age, with BMI, with the prescribed dose of risperidone, or with weight-adjusted dose. There was a negative correlation with the period of time since the beginning of treatment, but it was not associated with sex or skin color. In the analyzed psychiatic outpatient sample, it was observed that many children and adolescents do not use risperidone as monotherapy, which increases the risk of adverse effects. Some of the analyzed genetic polymorphisms were associated with belonging to the group that was taking risperidone, indicating possible genetic associations with psychopathological clusters of symptoms; however, only the rs6318 SNP of the HTR2C gene was associated with the frequency of hyperprolactinemia in both sexes together and among females alone, but not among males alone. Other psychiatric drugs can also alter prolactin concentrations and hyperprolactinemia. Even if concentrations are high, they tend to be asymptomatic; however, they are not necessarily free from posing medium-term and long-term risks.

Clinical Significance

It is suggested that all children and adolescents who begin the use of risperidone should have prolactin levels checked, for determination of individual control parameters, in addition to undergoing periodic monitoring during the course of treatment. The dosage should also be reevaluated any time new psychiatric drugs are prescribed. Even if the patient is asymptomatic, regular monitoring of prolactin levels is important because the concentration of prolactin can reach levels found in prolactinomas and still be asymptomatic. Even they do not reach such levels, there are symptoms that can be confounded with those of primary psychiatric disorders and there are others that only manifest over the long term if hyperprolactinemia continues. In addition, for all individuals using risperidone, it is recommended to review, at each psychiatric consultation, the possible clinical signs and symptoms that may be consistent with the occurrence of hyperprolactinemia.

Footnotes

Disclosures

No competing financial interests exist.

References

Adkins

, Aberg

, McClay

, Bukszár

, Zhao

, Jia

, Stroup

, Perkins

, McEvoy

, Lieberman

, Sullivan

, van den Oord

: Genomewide pharmacogenomic study of metabolic side effects to antipsychotic drugs. Mol Psychiatry, 16:321–332, 2011.

Aklillu

, Kalow

, Endrenyi

, Harper

, Miura

, Ozdemir

: CYP2D6 and DRD2 genes differentially impact pharmacodynamic sensitivity and time course of prolactin response to perphenazine. Pharmacogenet Genomics, 17:989–993, 2007.

Alenius

, Wadelius

, Dahl

M-L

, Hartvig

, Lindström

, Hammarlund-Udenaes

Gene polymorphism influencing treatment response in psychotic patients in a naturalistic setting. J Psychiatr Res, 42:884–893, 2008.

Aman

, De Smedt

, Derivan

, Lyons

, Findling

: Double-blind, placebo-controlled study of risperidone for the treatment of disruptive behaviors in children with subaverage intelligence. Am J Psychiatry, 159:1337–1346, 2002.

Anderson

, Scahill

, McCracken

, McDougle

, Aman

, Tierney

, Arnold

, Martin

, Katsovich

, Posey

, Shah

, Vitiello

: Effects of short- and long-term risperidone treatment on prolactin levels in children with autism. Biol Psychiatry, 61:545–550, 2007.

Besnard

, Auclair

, Callery

, Gabriel-Bordenave

, Roberge

: Antipsychotic-drug-induced hyperprolactinemia: physiopathology, clinical features and guidance [in French]. Encephale, 40:86–94, 2014.

Cabaleiro

, Ochoa

, López-Rodríguez

, Román

, Novalbos

, Ayuso

, Abad-Santos

: Effect of polymorphisms on the pharmacokinetics, pharmacodynamics, and safety of risperidone in healthy volunteers. Hum Psychopharmacol, 29:459–469, 2014.

Calarge

, Acion

, Kuperman

, Tansey

, Schlechte

: Weight gain and metabolic abnormalities during extended risperidone treatment in children and adolescents. J Child Adolesc Psychopharmacol, 19:101–109, 2009a.

Calarge

, Ellingrod

, Acion

, Miller

, Moline

, Tansey

, Schlechte

: Variants of the dopamine D2 receptor and risperidone-induced hyperprolactinemia in children and adolescents. Pharmacogenet Genomics, 19:373–382, 2009b.

10.

Calarge

, Nicol

, Schlechte

, Burns

: Cardiometabolic outcomes in children and adolescents following discontinuation of long-term risperidone treatment. J Child Adolesc Psychopharmacol, 24:120–129, 2014.

11.

Casanueva

, Molitch

, Schlechte

, Abs

, Bonert

, Bronstein

, Brue

, Cappabianca

, Colao

, Fahlbusch

, Fideleff

, Hadani

, Kelly

, Kleinberg

, Laws

, Marek

, Scanlon

, Sobrinho

, Wass

, Giustina

: Guidelines of the Pituitary Society for the diagnosis and management of prolactinomas. Clin Endocrinol (Oxf), 65:265–273, 2006.

12.

Choong

, Polari

, Kamdem

, Gervasoni

, Spisla

, Jaquenoud Sirot

, Bickel

, Bondolfi

, Conus

, Eap

: Pharmacogenetic study on risperidone long-acting injection: influence of cytochrome P450 2D6 and pregnane X receptor on risperidone exposure and drug-induced side-effects. J Clin Psychopharmacol, 33:289–98, 2013.

13.

Correia

, Almeida

, Santos

, Sequeira

, Marques

, Miguel

, Abreu

, ]oliveira

, Vicente

: Pharmacogenetics of risperidone therapy in autism: association analysis of eight candidate genes with drug efficacy and adverse drug reactions. Pharmacogenomics J, 10:418–430, 2010.

14.

Correll

: Antipsychotic use in children and adolescents: minimizing adverse effects to maximize outcomes. J Am Acad Child Adolesc Psychiatry, 47:9–20, 2008.

15.

Duval

, Guillon

, Mokrani

, Crocq

, Duarte

: Relationship between prolactin secretion, and plasma risperidone and 9-hydroxyrisperidone concentrations in adolescents with schizophreniform disorder. Psychoneuroendocrinology, 33:255–259, 2008.

16.

Ercan

, Basay

, Durak

, Ozbaran

: Risperidone in the treatment of conduct disorder in preschool children without intellectual disability. Child Adolesc Psychiatry Ment Health, 5:10, 2011.

17.

Farmer

, Arnold

, Bukstein

, Findling

, Gadow

, Li

, Butter

, Aman

: The treatment of severe child aggression (TOSCA) study: Design challenges. Child Adolesc Psychiatry Ment Health, 5:36, 2011.

18.

Freeman

, Choate-Summers

, Garcia

, Moore

, Sapyta

, Khanna

, March

, Foa

, Franklin

: The Pediatric Obsessive-Compulsive Disorder Treatment Study II: Rationale, design and methods. Child Adolesc Psychiatry Ment Health, 3:4, 2009.

19.

Gonçalves

, Prosdocimi

, Santos

, Ortega

, Pena

SDJ

: Sex-biased gene flow in African Americans but not in American Caucasians. Genet Mol Res, 6:256–261, 2007.

20.

Herpertz-Dahlmann

, Bühren

, Remschmidt

: Growing up is hard: Mental disorders in adolescence. Dtsch Arztebl Int, 110:432–439, 2013.

21.

, Panagiotopoulos

, McCrindle

, Grisaru

, Pringsheim

: Management recommendations for metabolic complications associated with second generation antipsychotic use in children and youth. J Can Acad Child Adolesc Psychiatry, 20:234–241, 2011.

22.

Kearns

, Reed

: Clinical pharmacokinetics in infants and children. A reappraisal. Clin Pharmacokinet, 17 Suppl 1:29–67, 1989.

23.

Kleinberg

, Davis

, de Coster

, Van Baelen

, Brecher

: Prolactin levels and adverse events in patients treated with risperidone. J Clin Psychopharmacol, 19:57–61, 1999.

24.

Knegtering

, Baselmans

, Castelein

, Bosker

, Bruggeman

, van den Bosch

: Predominant role of the 9-hydroxy metabolite of risperidone in elevating blood prolactin levels. Am J Psychiatry, 162:1010–1012, 2005.

25.

La Torre

, Falorni

: Pharmacological causes of hyperprolactinemia. Ther Clin Risk Manag, 3:929–951, 2007.

26.

Lencz

, Malhotra

: Pharmacogenetics of antipsychotic-induced side effects. Dialogues Clin Neurosci, 11:405–415, 2009.

27.

Lett

, Wallace

TJM

, Chowdhury

, Tiwari

, Kennedy

, Müller

: Pharmacogenetics of antipsychotic-induced weight gain: review and clinical implications. Mol Psychiatry, 17:242–266, 2012.

28.

Melis

, Paoletti

, Mais

, Mastrapasqua

, Strigini

, Fruzzetti

, Guarnieri

, Gambacciani

, Fioretti

: The effects of the gabaergic drug, sodium valproate, on prolactin secretion in normal and hyperprolactinemic subjects. J Clin Endocrinol Metab, 54:485–489, 1982.

29.

Mihara

, Kondo

, Yasui-Furukori

, Suzuki

, Ishida

, Ono

, Kubota

, Iga

, Takarada

, de Vries

, Kaneko

: Effects of various CYP2D6 genotypes on the steady-state plasma concentrations of risperidone and its active metabolite, 9-hydroxyrisperidone, in Japanese patients with schizophrenia. Ther Drug Monit, 25:287–293, 2003.

30.

Pandina

, Kushner

, Karcher

, Haas

: An open-label, multicenter evaluation of the long-term safety and efficacy of risperidone in adolescents with schizophrenia. Child Adolesc Psychiatry Ment Health, 6:23, 2012.

31.

Reis

, de Angelo

, Sakamoto

, Ferriani

, Lara

LAS

: Altered sexual and reproductive functions in epileptic men taking carbamazepine. J Sex Med, 10:493–499, 2013.

32.

Roke

, van Harten

, Boot

, Buitelaar

: Antipsychotic medication in children and adolescents: A descriptive review of the effects on prolactin level and associated side effects. J Child Adolesc Psychopharmacol, 19:403–414, 2009.

33.

Roke

, van Harten

, Franke

, Galesloot

, Boot

, Buitelaar

: The effect of the Taq1A variant in the dopamine D2 receptor gene and common CYP2D6 alleles on prolactin levels in risperidone-treated boys. Pharmacogenet Genomics, 23:487–493, 2013.

34.

Saito

, Correll

, Gallelli

, McMeniman

, Parikh

, Malhotra

, Kafantaris

: A prospective study of hyperprolactinemia in children and adolescents treated with atypical antipsychotic agents. J Child Adolesc Psychopharmacol, 14:350–358, 2004.

35.

Stadler

, Bolten

, Schmeck

: Pharmacotherapeutic intervention in impulsive preschool children: The need for a comprehensive therapeutic approach. Child Adolesc Psychiatry Ment Health, 5:11, 2011.

36.

Stahl

: The Prescriber'S Guide: Stahl'S Essential Psychopharmacology, 3rd ed. New York: Cambridge University Press; 2009.

37.

Stahl

: Stahl's Essential Psychopharmacology: Neuroscientific Basis and Practical Applications, 3rd ed. New York: Cambridge University Press; 2008.

38.

Trenque

, Herlem

, Auriche

, Dramé

: Serotonin reuptake inhibitors and hyperprolactinaemia: A case/non-case study in the French pharmacovigilance database. Drug Saf, 34:1161–1166, 2011.

39.

Troost

, Lahuis

, Hermans

, Buitelaar

, van Engeland

, Scahill

, Minderaa

, Hoekstra

: Prolactin release in children treated with risperidone: impact and role of CYP2D6 metabolism. J Clin Psychopharmacol, 27:52–57, 2007.

40.

Tsigkaropoulou

, Peppa

, Zompola

, Rizos

, Xelioti

, Chatziioannou

, Fillippopoulou

, Lykouras

: Hypogonadism due to hyperprolactinemia and subsequent first episode of psychosis. Gend Med, 9:56–60, 2012.

41.

Van de Kar

, Javed

, Zhang

, Serres

, Raap

, Gray

: 5-HT2A receptors stimulate ACTH, corticosterone, oxytocin, renin, and prolactin release and activate hypothalamic CRF and oxytocin-expressing cells. J Neurosci, 21:3572–3579, 2001.

42.

World Health Organization: Growth Reference 5–19 years. 2007. Available at http://www.who.int/growthref/who2007_bmi_for_age/en/index.html

43.

Yasui-Furukori

, Saito

, Tsuchimine

, Nakagami

, Sato

, Sugawara

, Kaneko

: Association between dopamine-related polymorphisms and plasma concentrations of prolactin during risperidone treatment in schizophrenic patients. Prog Neuropsychopharmacol Biol Psychiatry, 32:1491–1495, 2008.

44.

Zhang

J-P

, Malhotra

: Pharmacogenetics and antipsychotics: Therapeutic efficacy and side effects prediction. Expert Opin Drug Metab Toxicol, 7:9–37, 2011.

45.

Zito

, Derivan

, Kratochvil

, Safer

, Fegert

, Greenhill

: Off-label psychopharmacologic prescribing for children: History supports close clinical monitoring. Child Adolesc Psychiatry Ment Health, 2:24, 2008.