Methylphenidate Side Effect Profile Is Influenced by Genetic Variation in the Attention-Deficit/Hyperactivity Disorder-Associated CES1 Gene

Abstract

Objective:

A naturalistic, prospective study of the influence of genetic variation on dose prescribed, clinical response, and side effects related to stimulant medication in 77 children with attention-deficit/hyperactivity disorder (ADHD) was undertaken. The influence of genetic variation of the CES1 gene coding for carboxylesterase 1A1 (CES1A1), the major enzyme responsible for the first-pass, stereoselective metabolism of methylphenidate, was investigated.

Methods:

Parent- and teacher-rated behavioral questionnaires were collected at baseline when the children were medication naïve, and again at 6 weeks while they were on medication. Medication dose, prescribed at the discretion of the treating clinician, and side effects, were recorded at week 6. Blood and saliva samples were collected for genotyping. Single nucleotide polymorphisms (SNPs) were selected in the coding, non-coding and the 3′ flanking region of the CES1 gene. Genetic association between CES1 variants and ADHD was investigated in an expanded sample of 265 Irish ADHD families. Analyses were conducted using analysis of covariance (ANCOVA) and logistic regression models.

Results:

None of the CES1 gene variants were associated with the dose of methylphenidate provided or the clinical response recorded at the 6 week time point. An association between two CES1 SNP markers and the occurrence of sadness as a side effect of short-acting methylphenidate was found. The two associated CES1 markers were in linkage disequilibrium and were significantly associated with ADHD in a larger sample of ADHD trios. The associated CES1 markers were also in linkage disequilibrium with two SNP markers of the noradrenaline transporter gene (SLC6A2).

Conclusions:

This study found an association between two CES1 SNP markers and the occurrence of sadness as a side effect of short-acting methylphenidate. These markers were in linkage disequilibrium together and with two SNP markers of the noradrenaline transporter gene.

Introduction

Attention-deficit/hyperactivity disorder (ADHD) is one of the most common neurodevelopmental disorders of childhood, with worldwide prevalence rates of ∼5% (Polanczyk et al. 2007). Methylphenidate (MPH) is a stimulant medication commonly prescribed in the management of ADHD. There is considerable variance in the reported rate of response to MPH (Green 1995), and no reliable means to predict an individual's future response. Certain clinical parameters such as body weight are currently used as a guide to predict the likely dose required to achieve a positive therapeutic response. The inter-individual variability in response to MPH may be partly influenced by genetic factors (Polanczyk et al. 2010).

Previous pharmacogenetic studies have focused on genes related to the dopaminergic system, a prime site of action of MPH. Studies investigated the association between response to MPH and possession of gene polymorphisms associated with the dopamine transporter (DAT) (Winsberg and Comings 1999; Joober et al. 2007; Gruber et al. 2009); the dopamine receptors DRD2 (Leddy et al. 2009), DRD4 (Cheon et al. 2007), and DRD5 (Tahir et al. 2000); and the enzyme catecol-O-methyltransferase (COMT) (Kereszturi et al. 2008). Researchers have also studied the association between medication response and polymorphisms of genes of the noradrenergic system, including genes coding for the norepinephrine transporter (NET) (Yang et al. 2004) and the adrenergic α2A receptor (Cheon et al. 2009), the serotonergic and nicotinic systems, other genetic markers associated with ADHD (synaptosomal-associated protein 25 [SNAP-25], monoamine oxidase A [MAOA]), and a combination of these risk genes (Contini et al. 2012). These studies have yielded mixed results (see Polanczyk et al. 2010 for a review). One reason for the lack of consistency may be that genetic variability in the drug metabolism system influences plasma concentrations of MPH before MPH reaches the brain (Sun et al. 2004).

Orally administered MPH is absorbed completely from the gut (Hungund et al. 1979), undergoes first-pass stereoselective clearance in humans, and readily passes through the blood–brain barrier. MPH has a higher affinity for NET than DAT in both in vitro (Eshleman et al. 1999) and in vivo (Hannestad et al. 2010) studies, and it occupies both the DAT (Volkow et al. 1998) and the NET (Hannestad et al. 2010) at clinically relevant doses in healthy adults. The therapeutic efficacy of MPH may be a function of blockade of both the NET and DAT. Previous studies in healthy adult volunteers (Srinivas et al. 1991) and children with ADHD (Hungund et al. 1979; Srinivas et al. 1991; Shader et al. 1999) have shown that significant inter-individual variability exists in plasma concentrations of MPH following administration of standardized doses.

MPH is metabolized by the carboxylesterase 1A1 (CES1A1) enzyme in the liver (Sun et al. 2004). Expression of CES1A1 is similar across age groups (1–4, 6–9, 12–18, and 75–85 years) (Pope et al. 2005; Zhu et al. 2009) and the sexes (Zhu et al. 2009). It is of note that although there were no significant differences among the age groups noted in the Zhu et al. (2009) study, there were inter-individual differences in the level of CES1A1 expressed in the human liver (Zhu et al. 2009). Both environmental and genetic factors may explain this variability. Environmental factors, such as alcohol use and exposure to environmental pollutants and to lipophilic drugs, may have an influence on CES1A1 expression in adults (Hosokawa et al. 1995). CES1A1 activity is decreased by some commonly prescribed drugs, including aripiprazole, perphenazine, thioridazine, and fluoxetine (Zhu et al. 2010). Genetic variation may also influence the level of CES1A1 expression. Hosokawa and colleagues characterized the gene (CES1, MIM 114835) encoding human CES1A1 and sequenced its 5′ flanking promoter region (Hosokawa et al. 2008). Two CES1 variants were identified that were associated with significantly reduced enzyme activity, but they are both rare (Zhu et al. 2008; Walter Soria et al. 2010).

Only a handful of studies have examined the role of genes that influence the pharmacokinetic profile of stimulant medication in ADHD (Michelson et al. 2007; Ramoz et al. 2009). Nemoda and colleagues assessed the relationship between the CES1 Gly143Glu polymorphism and medication response in a sample of children with ADHD (n=122) (Nemoda et al. 2009). Analysis of the categorical grouping found no significant difference in the number of Gly/Glu heterozygotes in the responder and nonresponder groups (Nemoda et al. 2009). The Gly/Glu heterozygote group (n=5) required significantly less MPH than the Gly/Gly homozygote group (n=85) (Nemoda et al. 2009), suggesting that the heterozygote form was associated with reduced enzyme CES1A1 activity, resulting in higher plasma drug levels. No data regarding side effects were collected in this study (Nemoda et al. 2009). Bruxel and colleagues investigated the association between the single nucleotide polymorphism (SNP) −75 T>G (rs3815583) of the CES1, MPH treatment, and appetite reduction as a side effect in stimulant-naïve children with ADHD (Bruxel et al. 2013). Possession of the G allele was associated with a worsening of anorexia with MPH treatment (Bruxel et al. 2013).

The CES1 gene maps ∼99 kbp downstream of the SLC6A2 gene, which codes for the noradrenaline transporter (NET1) on chromosome 16. Our recent work has indicated a significant association between the Irish ADHD sample and a SLC6A2 haplotype (Hawi et al. 2013). We elected, therefore, to examine whether the investigated CES1 markers (of this study) were also associated with ADHD, as a consequence of potential linkage disequilibrium (LD) between two genes.

The general aim of this study was to examine the influence of DNA variation in CES1 on clinical response to MPH in a sample of medication-naïve children with ADHD. DNA variation in the CES1 gene variants may be associated with individual differences in the amount of CES1A1 in the liver. Hypothetically, variation in the rate of hydrolysis of MPH would vary the bioavailability of MPH, leading to differential prescriptions, symptom changes, and side effect profiles. We hypothesised that genetic variation in the CES1 gene would be associated with differential clinical response to MPH; specifically that CES1 genetic variants would influence the dose provided by the treating physician at the end of a 6 week period, any improvement in symptoms, and side effect frequency. This study used a naturalistic design, in that community consulting psychiatrists determined the dose of MPH for their child patients. There were four specific aims of this research. The first was to test the influence of CES1 genetic variants on the dose of MPH prescribed in mg/kg at 6 weeks of treatment in our sample of school-aged children with ADHD. The second aim was to test the influence of CES1 genetic variants on clinical response after 6 weeks of MPH treatment. The third aim was to test the influence of CES1 genetic variants on the side effect profile at 6 weeks of treatment with MPH. The fourth aim was to test for genetic association between CES1 variants and ADHD in an expanded sample (n=265) of ADHD trios.

Methods

A naturalistic, prospective pharmacogenetic study of stimulant response in ADHD in a sample of stimulant-naïve Irish children attending community-based child and adolescent psychiatry clinics was conducted in the Eastern region of the Republic of Ireland. Clinical data were collected at baseline, and after 6 weeks, 6 months, and 1 year of MPH treatment. Here, we focus on the data from the baseline and 6 week time points. Ethical approval for the study was granted by eight local research ethics committees, in accordance with the Declaration of Helsinki. All parents and/or legal guardians provided written consent for participation in the study, and older children additionally provided written assent.

Participants

For inclusion, participants were required to be between 4 and 15 years of age, meet American Psychiatric Association, Diagnostic and Statistical Manual of Mental Disorders, 4th ed. (DSM IV) (American Psychiatric Association 1994) criteria for ADHD, be stimulant naïve, and be willing to provide a blood or saliva sample for genetic analysis. Exclusion criteria included an intelligence quotient (IQ)<70, epilepsy, fragile X syndrome, fetal alcohol syndrome, maternal drug abuse during pregnancy, primary diagnosis of pervasive developmental disorder, Tourette's syndrome, psychosis, or bipolar disorder, and current treatment with other nonstimulant psychotropic medications.

Clinical assessment of DSM IV ADHD diagnoses were confirmed by a child psychiatrist (E.B.) using the parent version of the Child and Adolescent Psychiatric Assessment (CAPA) (Angold et al. 1995) and The Child Attention-Deficit Hyperactivity Disorder Teacher Telephone Interview (CHATTI) (Holmes et al. 2004). Sixty-three (82%) children were diagnosed with combined type ADHD, 7 (9%) were diagnosed with the hyperactive-impulsive subtype, and 7 (9%) were diagnosed with the inattentive subtype. Forty-six (60%) of the children were also diagnosed with oppositional defiant disorder and 12 (16%) were diagnosed with conduct disorder. A primary diagnosis of pervasive developmental disorder was ruled out using the Asperger Syndrome Diagnostic Interview (Gillberg et al. 2001). ADHD symptom severity was rated using the revised Conners' Parent and Teacher Rating Scale Short Version (Conners et al. 1998) at baseline and after 6 weeks on medication. IQ was assessed using four subtests (picture completion, vocabulary, information, block design) of the Wechsler Intelligence Scale for Children (WISC-III) (Wechsler 1992) and full IQ estimates were calculated using Sattler's method (Sattler 1992). Parents completed a Side-Effects Rating Scale (Barkley et al. 1990) after their child had been on medication for 6 weeks. As this was a naturalistic study, stimulant medication was prescribed at the discretion of the treating community-based consultant child and adolescent psychiatrists. Other medications taken concomitantly included paracetamol, salbutamol and steroid inhalers for asthma, and ampicillin antibiotics; these medications are not known to be metabolized by CES1A1.

Missing data

For one child at baseline and for two children at 6 weeks, only the Conners' Teacher ratings were available, and these were used to replace the missing Conners' Parent rating. The baseline body weight data of one child was missing, but body weight at 6 months was available; therefore his baseline weight was extrapolated based on the average weight gain of the cohort of 3.9% over the 6 month period.

Expanded sample used in the association study

Of the 77 children who participated in the pharmacogenetic analysis, 75 had at least one parent who provided a DNA sample. This set of families was extended in size by the inclusion of an additional 190 ADHD nuclear families that had been used previously for genetic association studies. All probands in these families fulfilled the ADHD DSM IV criteria for ADHD. Further clinical information can be found in Hawi et al. (2010).

CES1 genetic variants and genotyping

The CES1 gene is mapped to chromosome 16q13-q22.1 and contains 14 exons. Seven SNPs (Table 1) were selected in the CES1 gene, covering 39.4 kb of the gene and the surrounding regions, with average SNP density of 5.56 per kbp. All genotypes were in Hardy–Weinberg equilibrium. DNA was extracted from either blood or saliva samples. Saliva was collected using the Oragene DNA self-collection kits (http://www.dnagenotek.com/). The selected SNPs were commercially genotyped at KBiosciences using the KASPar assay technique (http://www.kbioscience.co.uk/chemistry_Kasp_intro.html).

Table 1.

CES1A1 Single Nucleotide Polymorphisms (SNPs) with Their Chromosomal Positions, Hardy–Weinberg Equilibrium, Genotyping Success Rate, Allele Frequencies, and Gene Position

Marker	Position	Hardy-Weinberg p value	% Genotyped	Minor allele frequency	Gene position
rs3815589 (C/G)	54384420	0.501	95.6	0.135	3′ Flanking
rs2287194 (T/C)	54385642	1	99.3	0.028	3′ Flanking
rs2244613 (A/C)	54402110	1	98	0.173	Intron 10
rs2002577 (G/C)	54407003	0.221	98.6	0.181	Intron 8
rs2307244 (C/T)	54414953	0.882	91.8	0.208	Intron 4
rs2307240 (G/A)	54420213	1	96.7	0.037	Exon 2 (Ser-Asn)
rs12443580 (A/G)	54423838	0.914	95.9	0.284	Intron 1

C, cytosine; G, guanine; T, thymine; A, adenine.

Outcome measures

The MPH dose was expressed in mg/kg/day and was calculated using the prescription at the 6 week time point. The mean dose for the full sample was 0.57 (SD 0.19), ranging from 0.18 to 1.0. The mean dose for the short-acting MPH group was 0.54 (SD 0.20), ranging from 0.18 to 1.0. It is of note that 0.5mg/kg/day is a moderate dose compared with other studies with doses of 0.7mg/kg/day and higher (e.g., Vitiello et al. 2007).

Clinical response was measured using the Conners' Parent Rating Scale ADHD Index T-score at baseline in comparison with the 6 week time point.

Categoric variables indicating the presence or absence of individual side effects after taking MPH were created, using 1 as the point of presence of the side effect. The side effects selected for this study, based on clinical experience, were decreased appetite, weight loss, headache or abdominal pain, irritability, sadness, insomnia, and tics. These data were solicited from parental responses to the Barkley Side Effect Rating Scale and a weighing of the child at the 6 week time point. Given that insomnia and tics can occur as part of the symptomology of ADHD, three possible responses were created to distinguish the onset of these side effects: 1) absent or 2) worse since on MPH or 3) new-onset since on MPH. There were too few children presenting with tics as a side effect (two with a new problem, five with an old problem worsening with MPH) to further analyze the data.

Statistical analysis

All analyses were conducted using IBM SPSS version 20. For each SNP, the individuals with the heterozygote combination of alleles were grouped with individuals with the minor homozygous allele combination, to create two groups per SNP (see Table 2).

Table 2.

The Number of Participants with Homozygote and Heterozygote Alleles at Each of the SNP Markers Investigated: Full Data Set (n=77)

Marker	Hom	Het	Hom	Missing genotype	Total	Grouping
rs3815589 (C/G)	60	15	0	2	77	C/C (60) vs. C/G (15)
rs2287194 (C/T)	0	5	70	2	77	C/T (5) vs. T/T (70)
rs2244613 (A/C)	51	19	3	4	77	A/A (51) vs. A/C & C/C (22)
rs2002577 (C/G)	5	19	52	1	77	C/C & C/G (24) vs. G/G (52)
rs2307244 (C/T)	52	17	6	2	77	C/C (52) vs. C/T & T/T (23)
rs2307240 (A/G)	0	4	72	1	77	A/G (4) vs. G/G (72)
rs12443580 (A/G)	37	32	7	1	77	AA (37) vs. A/G & G/G (39)

Ritalin Only Sample (n=45)
Marker	Hom	Het	Hom	Missing genotype	Total	Grouping
rs3815589 (C/G)	35	9	0	1	45	C/C (35) vs. C/G (9)
rs2287194 (C/T)	0	3	40	2	45	C/T (3) vs. T/T (40)
rs2244613 (A/C)	29	9	3	4	45	A/A (29) vs. A/C & C/C (12)
rs2002577 (C/G)	5	9	30	1	45	C/C (5) & C/G vs. G/G (39)
rs2307244 (C/T)	30	10	3	2	45	C/C (30) vs. C/T & T/T (13)
rs2307240 (A/G)	0	2	42	1	45	A/G (2) vs. G/G (42)
rs12443580 (A/G)	24	15	5	1	45	AA (24) vs. A/G & G/G (20)

SNP, single nucleotide polymorphism.

With reference to the first aim of this study, any difference between the genotype groups for the MPH dose (mg/kg/day) per child at 6 weeks was assessed using univariate analyses of variance (ANOVAs) and univariate analyses of covariance (ANCOVAs) with gender and age as covariates. With reference to the second aim of this study, the clinical response to MPH was assessed using a repeated measures ANCOVA with time (baseline, 6 weeks) as the within-subjects factor. Gender, age, and drug dose (mg/kg/day) were entered as covariates. The clinical response to MPH as a function of genotype and gender was assessed using a repeated measures ANCOVA with time (baseline, 6 weeks) as the within-subjects factor, and gender (male, female) and genotype (1,2) as the between-subjects factors. Age and drug dose were entered as covariates. With reference to the third aim of this study, a logistic regression model was developed to assess the capability of the genotype to predict side effect profiles of the children at 6 weeks. Here gender, age, and drug dose (mg/kg/day) were first entered into the model with genotype entered in the second step. Where three categorical variables were present as the dependent variable (e.g., tic side effect absent/new onset/exacerbation) multinomial logistic regression was used. With reference to the fourth aim of this study, transmission disequilibrium test (TDT) and haplotype analyses were performed on all examined markers using the program Haploview.

For all comparisons, the α level was set at 0.05 and Bonferroni adjustments were used for within-test post-hoc comparisons throughout the analyses. No attempt was made to adjust for multiple comparisons across the SNPs for the first three aims, as this study was exploratory. For the TDT, multiple testing adjustments across the SNPs and haplotypes were conducted by performing 20,000 permutations.

Results

Of 132 eligible families, 130 families provided consent and 108 of these families met inclusion criteria. From the 108 families, 96 children with ADHD proceeded to prescription of stimulant medication. Six children refused to provide DNA samples and three children refused to take part in the study at the 6 week point. The children were treated with either an MPH preparation (short or long-acting) or dexamphetamine. Of the 87 children suitable for the pharmacogenetic study, only children prescribed an MPH preparation were included in the analyses outlined herein. The final sample consisted of 77 children (66 males and 11 females; mean age 8.0 years, SD 2.6; age range 4–15; mean IQ 92.7, SD 15.9). We examined our hypotheses in all children prescribed MPH (irrespective of preparation) and in a subgroup of 45 children prescribed short-acting MPH (Ritalin) only (38 males; mean age 7.1 years, SD 1.7; age range 4–11; mean IQ 94.4, SD 16.4).

Aim 1: Influence of DNA variation in CES1 gene variants on MPH dosage at 6 weeks

There was no significant effect of genotype on dose of MPH prescribed at 6 weeks for any of the 7 SNP studies. This was the case for both the whole group (n=77) and the short-acting MPH group (n=45), with and without gender and age as covariates.

Aim 2a: MPH clinical response

There was a significant interaction between gender as a covariate and time; therefore, the analysis was rerun with gender entered as a between-subjects factor in the ANOVA. The male group (n=66; mean age 8.1, SD 2.6; age range 4–15; mean IQ 92.8, SD 16.5; mean dose 0.58, SD 0.19) did not differ significantly from the female group (n=11; mean age 7.6, SD 2.1; age range 5–11; mean IQ 92.0, SD 12.6; mean dose 0.52, SD 0.19) on age, IQ, or dose taken. Significant time and gender main effects were further explained by a significant time by gender interaction, F(1,73)=7.705, p=0.007, ηp2=0.095. Pairwise comparisons suggested that at baseline, the female group (mean 85.7, SD 9.2) presented with a significantly higher Conners' ADHD Index score than the male group (mean 75.8, SD 6.0), p<0.001. After six weeks of MPH treatment, there was no significant difference in the Conners' ADHD Index score between the female (mean 59.0, SD 7.4) and male (mean 58.8, SD 9.9) groups, p=0.964. Both the male and female groups presented with significantly reduced ADHD symptoms at the 6 week time point compared with the baseline, p<0.001. The result was the same with and without age and drug dose as covariates. For the short-acting MPH (Ritalin) subset, a similar set of results was found. A significant time by gender interaction was found, F(1,41)=6.040, p=0.018, ηp2=0.128. Pairwise comparisons indicated that at baseline, the female group (n=7, mean 86.8, SD 7.9) presented with a significantly higher Conners' ADHD Index score than the male group (n=38, mean 76.2, SD 6.6), p<0.001. There was no significant difference between the female (mean 58.2, SD 8.4) and male (mean 59.0, SD 9.7) groups after 6 weeks of short-acting MPH treatment. Both groups demonstrated a significant reduction in the ADHD Index score between the two time points, p<0.001. The result was the same with and without age and drug dose as covariates.

Aim 2b: MPH clinical response and CES1 genotypes

For SNP rs12443580, a time by gender by genotype group three-way interaction was found, F(1,70)=6.990, p=0.010, ηp2=0.091. This interaction was broken down by genotype group (see Fig. 1). For the A/A group, significant time and gender main effects were further explained by a significant time by gender interaction, F(1,33)=16.439, p<0.001, ηp2=0.333. At baseline, the female group (n=7, mean 89.9, SD 0.4) presented with a significantly higher Conners' ADHD Index score than did the male group (n=30, mean 76.0, SD 6.2), p<0.001. By 6 weeks, there was no significant difference between the female (mean 57.9, SD 8.6) and male (mean 61.4, SD 9.5) groups, p=0.366. Both the male and female groups presented with significantly reduced ADHD symptoms at the 6 week time point compared with the baseline, p<0.001. For the A/G and GG group, a time main effect was found, but no significant time by gender interaction was found, (female n=4, baseline mean 78.0, SD 12.8; 6 weeks mean 61.0, SD 5.0; male n=35, baseline mean 75.7, SD 6.2; 6 weeks mean 56.5, SD 9.9). The result was the same with and without age and drug dose as covariates.

FIG. 1.

The three-way interaction between time, gender and genotype for rs12443580.

No significant genotype effects were found for the other six SNPs, or for the short-acting MPH subset of participants.

Aim 3: Capability of CES1 genotypes to predict side effect profiles (for the full data set [n=77])

MPH side effect frequencies at 6 weeks are shown in Table 3. The side effect profile of one child prescribed short-acting MPH was not available.

Table 3.

The Frequency of Side Effects at 6 Weeks

Side effect	Full sample: Frequency, no. (n=77)	Ritalin only sample: Frequency, no. (n=45)
Reduced appetite	50 (66%)	26 (59%)
Weight loss	26 (34%)	9 (20%)
Headache or abdominal pain	38 (50%)	20 (45%)
Irritability	35 (46%)	23 (52%)
Sadness	29 (38%)	18 (41%)
Insomnia – new onset versus exacerbation	25 (33%) versus 23 (30%)	12 (27%) versus 13 (30%)
Tics – new onset versus exacerbation	2 (3%) versus 5 (7%)	1 (2%) versus 2 (5%)

A logistic regression model designed to assess the predictability of genotype on side effect occurrence was developed for the 6 week data. Age, gender, and drug dose (mg/kg) were first entered into the model, followed by the SNP genotype.

Decreased appetite

Drug dose was a significant predictor of whether or not a child presented with reduced appetite, for all of the SNP models. The mean drug dose for the children with reduced appetite (n=50) was 0.61 mg/kg/day (SD 0.18) compared with the mean drug dose of 0.50 mg/kg/day (SD 0.20) of children without this side effect (n=26). As an example, the results for the SNP rs12443580 were β=3.3 (SE 1.4), p=0.017, Exp(β)=28.3, (2, 443). Age, gender, and each of the SNPs were not significant predictors of reduced appetite.

Weight loss

Drug dose was a significant predictor of whether or not a child presented with weight loss, for all of the SNP models. The mean drug dose for the children experiencing weight loss (n=26) was 0.68 mg/kg/day (SD 0.17), compared with the mean drug dose of 0.51 (SD 0.19) for children not experiencing weight loss (n=50). As an example, the results for the SNP rs12443580 were β=5.2 (SE 1.6), p=0.001, Exp(β)=189.8, (8, 4386). Age, gender, and each of the SNPs were not significant predictors of weight loss.

Sadness

Age was a significant predictor of whether or not a child presented with sadness, for all of the SNP models. The mean age of the children (n=29) experiencing sadness was 6.8 years (SD 2.5), compared with the mean age 8.7 years (SD 2.3) of the children not experiencing sadness (n=47). As an example, the results for the SNP rs12443580 were were β=−0.41 (SE 0.14), p=0.003, Exp(β)=0.66, [0.51, 0.87]. Gender, drug dose, and each of the SNPs were not significant predictors of sadness.

Other side effects

None of the factors entered into the logistic regression were significant predictors of whether or not a child presented with headache or abdominal ache, irritability, or insomnia.

Aim 3: Capability of CES1 genotypes to predict side effect profiles (for the short-acting MPH data set [n=45])

Weight loss

A trend was noted for drug dose to be a predictor of whether or not a child presented with weight loss, for all of the SNP models, with the reduced sample size.

Sadness

Age was a significant predictor of whether or not a child presented with sadness, for all of the SNP models. The mean age of the children (n=18) experiencing sadness was 6.0 years (SD 1.0), compared with the mean age 7.7 years (SD 1.6) of the children not experiencing sadness (n=26). Gender and drug dose did not predict the presence of sadness as a side effect.

Two SNPs were associated with sadness as a side effect, when age was entered into the model in the previous step (see Table 4). rs2244613 was a significant predictor of sadness, B=−2.66 (SE 1.2), p=0.032, Exp(β)=0.07, (0.01, 0.80). This model explained a reasonable amount of variance, R² =0.59 (Hosmer–Lemeshow), 0.42 (Cox–Snell), 0.58 (Nagelkerke), model χ² (4)=21.93, p<0.001. Children with two copies of the A allele (n=29) on short-acting MPH were significantly more likely to experience sadness as a side effect than were children with one or no copies of the A allele (n=12).

Table 4.

Logistic Regression Analyses Predicting Sadness as a Side-Effect From Gender, Drug Dose, Age and SNP rs2244613.

Predictor	B	S.E. B	β	95% CI for β
Step 1 Constant	−0.511	0.327	0.60
Step 2 Gender	0.985	1.061	2.68	[0.33, 21.45]
Drug dose^a	−0.416	1.763	0.66	[0.02, 20.89]
Age	−1.206^*	0.433	0.30	[0.13, 0.70]
Constant	6.947	3.028	1039.82
Step 3 Gender	1.211	1.256	3.36	[0.29, 39.37]
Drug dose	0.527	2.055	1.69	[0.03, 95.09]
Age	−1.388^*	0.449	0.25	[0.10, 0.60]
SNP rs2244613	−2.657^*	1.241	0.07	[0.01, 0.80]
Constant	8.083	3.113	3240.21

(mg/kg/day).

R ²=0.59 (Hosmer & Lemeshow), 0.42 (Cox & Snell), .58 (Nagelkerke). Model χ² (4)=21.93, p<0.001.

p<.05. ^** p<.001.

SNP, single nucleotide polymorphism.

A trend was found for rs2002577, B=−1.75 (SE 0.97), p=0.071, Exp(β)=0.17, (0.03, 1.12). This model also explained a reasonable amount of variance, R² =0.67 (Hosmer–Lemeshow), 0.36 (Cox–Snell), 0.49 (Nagelkerke), model χ² (4)=19.236, p<0.001. There was a trend for children with two copies of the G allele (n=30) taking short-acting MPH to experience sadness as a side effect compared with children with one or no copies of the G allele (n=14).

Other side-effects

None of the factors entered into the logistic regression were significant predictors of whether or not a child presented with reduced appetite, headache or abdominal ache, irritability, or insomnia.

Aim 4: Test of genetic association between CES1 variants and ADHD in an expanded sample of ADHD trios

TDT analysis on all examined markers is presented in Table 5. Nominally significant associations between ADHD and the SNPs rs2287194, rs2244613, rs2002577, and rs12443580 were observed (Table 5). Only SNPs rs2244613 and rs2002577 remained significant (p-corrected=0.0248 and 0.0248, respectively) after the 20,000 permutation test. It is important to note that these markers are not independent, as they are in very strong LD (D’=1, r² =99). Haplotype analysis (based on the Haploview default setting) identified one haplotype block between SNPs rs2244613 and rs2002577. A significant association of ADHD and a haplotype comprising the allele A of the SNP rs2244613 and the allele G of the SNP rs2002577 was observed (χ²=10.129, p=0.0015, OR=1.66) which remained significant after the 20,000 permutation test (χ²=10.129, p-corrected=0.0003).

Table 5.

Transmission Disequilibrium Test Analysis of CES1 in 265 Attention-Deficit/ Hyperactivity Disorder (ADHD) Nuclear Families

Marker	OTA	T : UT	χ²	p value	Odds ratio	Empirical p value from 20,000 permutations
rs3815589 (C/G)	C	48:43	0.275	0.600	1.12	0.995
rs2287194 (C/T)	T	17:5	6.545	0.011	3.40	0.072
rs2244613 (A/C)	A	73:42	8.357	0.0038	1.74	0.024
rs2002577 (C/G	G	73:42	8.357	0.0038	1.74	0.024
rs2307244 (C/T)	T	65:53	1.22	0.269	1.23	0.869
rs2307240 (A/G)	-	13:13	0	1	1	1
rs12443580 (A/G)	A	87:63	3.84	0.05	1.38	0.299

OTA, overtransmitted allele; T, transmitted allele; UT, untransmitted allele.

Discussion

A naturalistic, prospective pharmacogenetic study of stimulant response in a sample of medication-naïve Irish children diagnosed with ADHD was conducted. The specific study aims were to examine the influence of selected CES1 SNPs on MPH dose prescribed, ADHD symptom change, side effect profile at 6 weeks of treatment, and a test for genetic association between CES1 variants and ADHD in an expanded sample (n=265) of ADHD trios. None of the CES1 gene variants were associated with the dose of MPH provided or the clinical response recorded at the 6 week time point. The dose of MPH significantly predicted decreased appetite and weight loss in the whole sample; the greater the dose the greater the side effects of reduced appetite and weight loss. These side effects of MPH are well known (Gittelman-Klein et al. 1976; Barkley et al. 1990; NICE 2009). In both the whole cohort and the short-acting MPH subset, age of the child significantly predicted sadness as a side effect of MPH; the younger the child, the more likely that sadness would present as a side effect. After accounting for variation associated with age, two SNPs were associated with sadness as a side effect of short-acting MPH. Possession of two copies of the A allele of rs2244613 was a significant predictor of sadness, whereas possession of two copies of the G allele of rs2002577 was a trend predictor of sadness. These two CES1 markers that were in LD were significantly associated with ADHD in the larger sample of ADHD trios and were in LD with two SNP markers of the noradrenaline transporter gene SLC6A2 (see Fig. 2). Although these novel genetic findings all require replication, they offer important information regarding both the genetic etiology of ADHD and appropriate psychological treatment of children prescribed short-acting MPH.

FIG. 2.

Linkage disequilibrium relations between SLC6A2 and CES1A1 markers. NET1 markers are numbered 1 (rs2397771) to 10 (rs9930182) and CES1A1 markers are numbered 11 (rs3815589) to marker 17 (rs12443580).

CES1A1 is predominantly expressed in the liver, with low levels of expression in the heart and lungs (Islam et al. 1999). It is unlikely, therefore, that this enzyme would have a direct effect on the development of ADHD, a neurodevelopmental disorder. The CES1 gene maps next to the SCL6A2 gene that codes for the NET, on chromosome 16. Our recent work has indicated a significant association between the Irish ADHD sample and a SLC6A2 haplotype (Hawi et al. 2013). The NET adjusts the extracellular concentrations of noradrenaline and dopamine in the prefrontal cortex (Arnsten 2011). Noradrenaline plays a critical role in prefrontal functioning, including attention regulation, behavioral inhibition, working memory, and planning functions (for a recent review, see Arnsten 2011). Alterations in the levels of noradrenaline and dopamine can impair functioning of the prefrontal cortex, resulting in poor attention and impulse control (Aston-Jones et al. 2000); symptoms that are related to ADHD. MPH blocks both the DAT and NET (Arnsten 2009). The association of CES1A1 with ADHD found in this study may, therefore, reflect its LD relationship with a variant at SLC6A2 that itself is associated with ADHD. The CES1A1 markers rs2244613 and rs2002577 are in LD (D’=0.53-0.80) with the SLC6A2 markers rs5569 and rs2242447 (Fig. 2). Both of the SLC6A2 SNPs showed evidence of association with ADHD in our sample (Hawi et al. 2013). Alternatively, the CES1A1 associated variants may be functionally important and influence the expression of another gene that lies in a pathway involved in the etiology of ADHD. These speculations however, require examination in future studies.

The most common side effects of MPH are insomnia, decreased appetite, stomachaches, and headaches (Gittelman-Klein et al. 1976; Barkley et al. 1990; Stein et al. 1996; Efron et al. 1997; Schachter et al. 2001). Although it is less common, sadness as a sideeffect of MPH has been noted previously in school-aged children (Gittelman-Klein et al. 1976; Elia et al. 1991; Buhrmester et al. 1992) and in preschool children (Schleifer et al. 1975; Firestone et al. 1998). A recent literature review of the medication response in preschool children highlighted dysphoria, and also crying, whining, irritability, and solitary play, as more frequent side effects of MPH in this agegroup compared with older children (Ghuman et al. 2008). An analysis investigating the effects of age on the development of side effects is rarely seen in the literature. In the present study, neither the dose of MPH nor the gender of the child were significantly related to the prediction of sadness as a side effect. This age effect is suggestive of a developmental change in the ability of a child to physiologically manage MPH. An increase in sadness in particular, points to an interaction of MPH with the neurotransmitter systems related to emotion processing and mood; these include serotonin, noradrenaline, and dopamine (Elhwuegi 2004). Acute treatment with MPH acts to increase the extracellular levels of both noradrenaline and dopamine, but does not affect serotonin levels in mice (Koda et al. 2010) or rats (Berridge et al. 2006; Weikop et al. 2007). It is unclear at this stage if MPH modifies extracellular serotonin levels in humans, but the affinity of MPH for the serotonin transporter is substantially reduced compared with its affinity for NET and DAT (Eshleman et al. 1999). This increase in sadness as a side effect of MPH in the younger children probably relates to induced alterations in the noradrenaline and dopamine systems.

After accounting for variation associated with age, possession of two copies of the A allele of CES1 SNP rs2244613 and the G allele of CES1 SNP rs2002577 (at trend level), were found to be predictors of sadness as a side effect of the short-acting MPH only. This suggests that the timing of release of MPH into the gastrointestinal system is important in the development of sadness as a side effect. It is unclear if these CES1 markers play a functional role in varying the amount of liver enzyme CES1A1 and the metabolism of MPH. Hypothetically, variation in the rate of hydrolysis of MPH as a function of the CES1 markers could lead to variation in the side effect profile. Further pharmacologic research is required. These two markers are in LD together and are significantly associated with ADHD in our larger ADHD trios sample. Intriguingly, they are in LD with two SNP markers of the SLC6A2 NET1 tranporter gene. Further research is required to clarify the specific genetic roles of these markers.

Girls with ADHD scored significantly higher on the Conners' Parent Rating Scale ADHD Index score than boys. This has been noted previously in a large cohort of children with combined-type ADHD (Müller et al. 2011), but is not typical of the literature. A number of reasons could explain the higher Conners ADHD Index rating. Inattention symptoms are a major contributor to the questions related to this index, and it may, therefore, be reflecting the common finding of greater inattentive symptoms in girls, compared with boys, with ADHD (Biederman et al. 2002). Alternatively, the girls in this sample may represent an extreme of the ADHD behavior profile, or genetics might underpin this difference. The sex difference in baseline Conners ADHD Index scores was modified by the possession of a SNP marker of the CES1 gene. Girls with ADHD with two copies of the A allele of rs12443580 were rated with significantly higher Conners ADHD Index scores than boys with two copies of the A allele at baseline. Boys and girls with one or no copies of the A allele did not differ in their ADHD Index scores. MPH administration ameliorated these differences in behavior by the 6 week time point. Given that these are baseline behavioral differences, and that the children were MPH naïve, it is impossible for the CES1 genetic variation to have had a pharmacokinetic influence on these results. The most likely interpretation of this result is that the SNP rs12443580 may be in LD with another genetic marker that is associated with ADHD. Further investigation into sex differences in the genetics of ADHD is warranted.

Strengths

This naturalistic pharmacogenetic study examined data collected prospectively from stimulant-naïve children diagnosed with ADHD after a thorough clinical examination. As the medication was prescribed by community-based child and adolescent psychiatrists, a “real-life” representation of MPH response and side-effect experience resulted.

Limitations

A superior study design to this naturalistic study would have been a crossover design with a forced MPH-dose titration protocol. Serum measures of MPH were not obtained in our study, and would have been highly desirable. An important intermediate step would have been to test the effect of CES1A1 genetic variants on serum MPH measurements following the administration of a range of MPH doses to medication-naïve children with ADHD. Zhu et al (2008) reported their findings from such a study conducted with healthy adult volunteers. Our study assumed such an effect in children with ADHD, but this remains to be demonstrated. Our study did not consider environmental factors impacting on CES1A1 function, other than the concomitant prescription of medications known to be metabolized by CES1A1.

Conclusions

In conclusion, this study finds an association between two CES1 SNP markers and the occurrence of sadness as a side effect of short-acting MPH in previously medication-naïve children with ADHD. These two CES1 markers were significantly associated with ADHD in the larger ADHD trios sample, and were in significant LD together and with two SNP markers of the noradrenaline transporter gene SLC6A2. These novel findings are of interest, and indicate new research directions for pharmacogenetic studies of MPH in ADHD.

Clinical Significance

An association was found between markers of a gene coding for the major enzyme involved in the metabolism of MPH and the occurrence of sadness as a side effect of short-acting MPH in children newly diagnosed with ADHD. The genetic markers were also in LD with two markers of the noradrenline transporter gene. Side-effect profiles related to MPH may be related to genetic markers associated with the CES1 gene.

Footnotes

Acknowledgments

We thank the families that participated in the study, Marie Cox for help with data collection, and all referring consultants: Dr. Atkins, Dr. Dunne, Prof. Fitzpatrick, Dr. Halpin, Dr. Holmes, Dr. Lawlor, Dr. Leader, Dr. McCarthy, Dr. Moran, Dr. Noone, Dr. O'Donovan, and Dr. Sharkey.

Disclosures

No competing financial interests exist.

References

American Psychiatric Association: Diagnostic and Statistical Manual of Mental Disorders, 4th ed. Washington, DC: American Psychiatric Association; 1994.

Angold

, Predergast

, Cox

, Harrington

, Simonoff

, Rutter

: The Child and Adolescent Psychiatric Assessment (CAPA). Psychol Med, 25:739–753, 1995.

Arnsten

: Toward a new understanding of attention-deficit hyperactivity disorder pathophysiology. CNS Drugs, 23:33–41, 2009.

Arnsten

: Catecholamine influences on dorsolateral prefrontal cortical networks. Biol Psychiatry, 69:e89–99, 2011.

Aston–Jones

, Rajkowski

, Cohen

: Locus coeruleus and regulation of behavioral flexibility and attention. Prog Brain Res, 126:165–182, 2000.

Barkley

, McMurray

, Edelbrock

, Robbins

: Side effects of methylphenidate in children with attention deficit hyperactivity disorder: A systemic, placebo-controlled evaluation. Pediatrics, 86:184–192, 1990.

Berridge

, Devilbiss

, Andrzejewski

, Arnsten

, Kelley

, Schmeichel

, Hamilton

, Spencer

: Methylphenidate preferentially increases catecholamine neurotransmission within the prefrontal cortex at low doses that enhance cognitive function. Biol Psychiatry, 60:1111–1120, 2006.

Biederman

, Mick

, Faraone

, Braaten

, Doyle

, Spencer

, Wilens

, Frazier

, Johnson

: Influence of gender on attention deficit hyperactivity disorder in children referred to a psychiatric clinic. Am J Psychiatry, 159:36–42, 2002.

Bruxel

, Salatino–Oliveira

, Genro

, Polanczyk

, Chazan

, Rohde

, Hutz

: Association of a carboxylesterase 1 polymorphism with appetite reduction in children and adolescents with attention-deficit/hyperactivity disordertreated with methylphenidate. Pharmacogenomics J, 13:476–480, 2013.

10.

Buhrmester

, Whalen

, Henker

, MacDonald

, Hinshaw

: Prosocial behavior in hyperactive boys: Effects of stimulant medication and comparison with normal boys. J Abnorm Child Psychol, 20:103–121, 1992.

11.

Cheon

, Cho

, Koo

, Song

, Namkoong

: Association between homozygosity of a G allele of the alpha-2a-adrenergic receptor gene and methylphenidate response in Korean children and adolescents with attention-deficit/hyperactivity disorder. Biol Psychiatry, 65:564–570, 2009.

12.

Cheon

, Kim

, Cho

: Association of 4-repeat allele of the dopamine D4 receptor gene exon 111 polymorphism and response to methylphenidate treatment in Korean ADHD children. Neuropsychopharmacology, 32:1377–1383, 2007.

13.

Conners

, Sitarenios

, Parker

JDA

, Epstein

: The revised Conner's Parent Rating Scale (CPRS-R): factor structure, reliability, and criterion validity. J Abnorm Child Psychol, 26:257–269, 1998.

14.

Contini

, Victor

, Bertuzzi

, Salgado

, Picom

, Grevet

, Rohde

, Belmonte-de-Abreu

, Bau

: No significant association between genetic variants in 7 candidate genes and response to methylphenidate treatment in adult patients with ADHD. J Clin Psychopharmacol, 32:820–823, 2012.

15.

Efron

, Jarman

, Barker

: Side effects of methylphenidate and dexamphetamine in children with attention deficit hyperactivity disorder: A double-blind, crossover trial. Pediatrics, 100:662–666, 1997.

16.

Elhwuegi

: Central monoamines and their role in major depression. Prog Neuropsychopharmacol Biol Psychiatry, 28:435–451, 2004.

17.

Elia

, Borcherding

, Rapoport

, Keysor

: Methylphenidate and dextroamphetamine treatments of hyperactivity: Are there true nonresponders?. Psychiatry Res, 36:141–155, 1991.

18.

Eshleman

, Carmolli

, Cumbay

, Martens

, Neve

, Janowsky

: Characteristics of drug interactions with recombinant biogenic amine transporters expressed in the same cell type. J Pharmacol Exp Ther, 289:877–885, 1999.

19.

Firestone

, Monteiro Musten

, Pisterman

, Mercer

, Bennett

: Short-term side effects of stimulant medication are increased in preschool children with attention-deficit/hyperactivity disorder: A double-blind placebo-controlled study. J Child Adolesc Psychopharmacol, 8:13-25, 1998.

20.

Ghuman

, Arnold

, Anthony

: Psychopharmacological and other treatments in preschool children with attention-deficit/hyperactivity disorder: Current evidence and practice. J Child Adolesc Psychopharmacol, 18:413–447, 2008.

21.

Gillberg

, Gillberg

, Råstram

, Wentz

: The Asperger Syndrome (and high-funtioning autism) Diagnostic Interview (ASDI): A preliminary study of a new structured clinical interview. Autism, 5:57–66, 2001.

22.

Gittelman–Klein

, Klein

, Katz

, Saraf

, Pollack

: Comparative effects of methylphenidate and thioridazine in hyperkinetic children. Arch Gen Psychiatry, 33:1217–1231, 1976.

23.

Green

: The treatment of attention-deficit hyperactivity disorder with nonstimulant medications. Child Adolesc Psychiatr Clin N Am, 4:169–195, 1995.

24.

Gruber

, Joober

, Grizenko

, Leventhal

, Cook

EHJ

, Stein

: Dopamine transporter genotype and stimulant side effect factors in youth diagnosed with attention-deficit/hyperactivity disorder. J Child Adolesc Psychopharmacol, 19:233–239, 2009.

25.

Hannestad

, Gallezot

J-D

, Planeta–Wilson

, Lin

S-F

, Williams

, van Dyck

, Malison

, Carson

, Ding

Y-S

: Clinically relevant doses of methylphenidate significantly occupy norepinephrine transporters in humans in vivo. Biol Psychiatry, 68:854–860, 2010.

26.

Hawi

, Kent

, Hill

, Anney

, Brookes

, Barry

, Franke

, Banaschewski

, Buitelaar

, Ebstein

, Miranda

, Oades

, Roeyers

, Rothenberger

, Sergeant

, Sonuga–Barke

, Steinhausen

, Faraone

, Asherson

, Gill

: ADHD and DAT1: further evidence of paternal over-transmission of risk alleles and haplotype. Am J Med Genet B Neuropsychiatr Genet, 153B:97–102, 2010.

27.

Hawi

, Matthews

, Barry

, Kirley

, Wagner

, Wallace

, Heussler

, Vance

, Gill

, Bellgrove

: A high density linkage disequilibrium mapping in 14 noradrenergic genes: evidence of association between SLC6A2, ADRA1B and ADHD. Psychopharmacology (Berl), 225:895–902, 2013.

28.

Holmes

, Lawson

, Langley

, Fitzpatrick

, Trumper

, Pay

, Harrington

, Thapar

: The Child Attention-Deficit Hyperactivity Disorder Teacher Telephone Interview (CHATTI): Reliability and validity. Br J Psychiatry, 184:74–78, 2004.

29.

Hosokawa

, Endo

, Fujisawa

, Hara

, Iwata

, Sato

, Satoh

: Interindividual variation in carboxylesterase levels in human liver microsomes. Drug Metab Dispos, 23:1922–1027, 1995.

30.

Hosokawa

, Furihata

, Yaginuma

, Yamamoto

, Watanabe

, Tsukada

, Ohhata

, Kobayashi

, Satoh

, Chiba

: Structural organization and characterization of the regulatory element of the human carboxylesterase (CES1A1 and CES1A2) genes. Drug Metab Pharmacokinet, 23:73–84, 2008.

31.

Hungund

, Perel

, Hurwic

, Sverd

, Winsberg

: Pharmacokinetics of methylphenidate in hyperkinetic children. Br J Clin Pharmacol, 8:571–576, 1979.

32.

Islam

, Waheed

, Shah

, Tomatsu

, Sly

: Human egasyn binds beta-glucuronidase but neither the esterase active site of egasyn nor the C terminus of beta-glucuronidase is involved in their interaction. Arch Biochem Biophys, 372:53–61, 1999.

33.

Joober

, Grizenko

, Sengupta

, Amor

, Schmitz

, Schwartz

, Karama

, Lageix

, Fathalli

, Torkaman–Zehi

, Ter-Stepanian

: Dopamine transporter 3'-UTR VNTR genotype and ADHD: a pharmaco-behavioural genetic study with methylphenidate. Neuropsychopharmacology, 32:1370–1376, 2007.

34.

Kereszturi

, Tarnok

, Bognar

, Lakatos

, Farkas

, Gadoros

, Sasvari–Szekely

, Nemoda

: Catechol-O-methyltransferase Val158Met polymorphism is associated with methylphenidate responsein ADHD children. Am J Med Genet B Neuropsychiatr Genet, 147B:1431–1435, 2008.

35.

Koda

, Ago

, Cong

, Kita

, Takuma

, Matsuda

: Effects of acute and chronic administration of atomoxetine and methylphenidate on extracellular levels of noradrenaline, dopamine and serotonin in the prefrontal cortex and striatum of mice. J Neurochem, 114:259–270, 2010.

36.

Leddy

, Waxmonsky

, Salis

, Paluch

, Gnagy

, Mahaney

, Erbe

, Pelham

, Epstein

: Dopamine-related genotypes and the dose-response effect of methylphenidate on eating in attention-deficit/hyperactivity disorder. J Child Adolesc Psychopharmacol, 19:127–136, 2009.

37.

Michelson

, Read

, Ruff

, Witcher

, Zhang

, McCracken

: CYP2D6 and clinical response to atomoxetine in children and adolescents with ADHD. J Am Acad Child Adolesc Psychiatry, 46:242–251, 2007.

38.

Müller

, Asherson

, Banachewski

, Buitelaar

, Ebstein

, Eisenberg

, Gill

, Manor

, Miranda

, Oades

, Roeyers

, Rothenberger

, Sergeant

, Sonuga–Barke

EJS

, Thompson

, Faraone

, Steinhausen

H-C

: The impact of study design and diagnostic approach in a large multi-centre ADHD study: Part 2: Dimensional measures of psychopathology and intelligence. BMC Psychiatry, 11:55, 2011.

39.

Nemoda

, Angyal

, Tarnok

, Gadoros

, Sasvari–Szekely

: Carboxylesterase 1 gene polymorphism and methylphenidate response in ADHD. Neuropharmacology, 57:731–733, 2009.

40.

NICE: Attention Deficit Hyperactivity Disorder: The NICE guideline on diagnosis and management of ADHD in children, young people and adults. London: The British Psychological Society and the Royal College of Psychiatrists; 2009.

41.

Polanczyk

, Bigarella

, Hutz

, Rohde

: Pharmacogenetic approach for a better drug treatment in children. Curr Pharm Des, 16:2462–2473, 2010.

42.

Polanczyk

, Silva de Lima

, Lessa Horta

, Biederman

, Rohde

: The worldwide prevalence of ADHD: a systematic review and metaregression analysis. Am J Psychiatry, 164:942–948, 2007.

43.

Pope

, Karanth

, Liu

, Yan

: Comparative carboxylesterase activitys in infant and adult liver and their in vitro sensitivity to chlorpyrifos oxon. Regul Toxicol Pharmacol, 42:64–69, 2005.

44.

Ramoz

, Boni

, Downing

, Close

, Peters

, Prokop

, Allen

, Hamon

, Purper–Ouakil

, Gorwood

: A haplotype of the norepinephrine transporter (Net) gene Slc6a2 is associated with clinical response to Atomoxetine in attention-deficit hyperactivity disorder (ADHD). Neuropsychopharmacology, 34:2135–2142, 2009.

45.

Sattler

: Assessment of Children: WISC-III and WPPSI-R Supplement. Ithaca, NY: Jerome M. Sattler Inc.; 1992.

46.

Schachter

, Pham

, King

, Langford

, Moher

: How efficacious and safe is short-acting methylphenidate for the treatment of attention-deficit disorder in children and adolescents? A meta-analysis. CMAJ, 165:1475–1488, 2001.

47.

Schleifer

, Weiss

, Cohen

, Elman

, Cvejic

, Kruger

: Hyperactivity in preschoolers and the effect of methylphenidate. Am J Orthopsychiatry, 45:38–50, 1975.

48.

Shader

, Harmatz

, Oesterheld

, Parmalee

, Sallee

, Greenblatt

: Population pharmacokinetics of methylphenidate in children with attention-deficit hyperactvity disorder. J Clin Pharmacol, 39:775–785, 1999.

49.

Srinivas

, Hubbard

, Quinn

, Korchinski

, Midha

: Extensive and enantioselective presystemic metabolism of dl-threo-methylphenidate in humans. Prog Neuropsychopharmacol Biol Psychiatry, 15:213–220, 1991.

50.

Stein

, Blondis

, Schnitzler

, O'Brien

, Fishkin

, Blackwell

, Szumowski

, Roizen

: Methylphenidate dosing: twice daily versus three times daily. Pediatrics, 98:748–756, 1996.

51.

Sun

, Murry

, Sanghani

, Davis

, Kedishvili

, Zou

, Hurley

, Bosron

: Methylphenidate is stereoselectively hydrolyzed by human carboxylesterase CES1A1. J Pharmacol Exp Ther, 310:469–476, 2004.

52.

Tahir

, Yazgan

, Cirakoglu

, Ozbay

, Waldman

, Asherson

: Association and linkage of DRD4 and DRD5 with attention deficit hyperactivity disorder (ADHD) in a sample of Turkish children. Mol Psychiatry, 5:396–404, 2000.

53.

Vitiello

, Abikoff

, Chuang

, Kollins

, McCracken

, Riddle

, Swanson

, Wigal

, McGough

, Ghuman

, Wigal

, Skrobala

, Davies

, Posner

, Cunningham

, Greenhill

: Effectiveness of methylphenidate in the 10-month continuation phase of the preschoolers with ADHD treatment study (PATS). J Child Adolesc Psychopharmacol, 17:593–603, 2007.

54.

Volkow

, Wang

, Fowler

, Gatley

, Logan

, Ding

Y-S

, Hitzemann

, Pappas

: Dopamine transporter occupancies in the human brain induced by therapeutic doses of oral methylphenidate. Am J Psychiatry, 155:1325–1331, 1998.

55.

Walter Soria

, Belaus

, Galvan

, Ana Pasquali

, Velez

, Del Carmen Montes

, Beltramo

: A simple allele-specific polymerase chain reaction method to detect the Gly143Glu polymorphism in the human carboxylesterase 1 gene: Importance of genotyping for pharmacogenetic treatment. Genet Test Mol Biomarkers, 14:749–751, 2010.

56.

Wechsler

: WISC-III UK. London: Harcourt Brace & Company; 1992.

57.

Weikop

, Yoshitake

, Kehr

: Differential effects of adjunctive methylphenidate and citalopram on extracellular levels of serotonin, noradrenaline and dopamine in the rat brain. Eur Neuropsychopharmacol, 17:658–671, 2007.

58.

Winsberg

, Comings

: Association of the dopamine transporter gene (DAT1) with poor methylphenidate response. J Am Acad Child Adolesc Psychiatry, 38:1474–1477, 1999.

59.

Yang

, Wang

, Li

, Faraone

: Association of norepinephrine transporter gene with methylphenidate response. J Am Acad Child Adolesc Psychiatry, 43:1154–1158, 2004.

60.

Zhu

, Appel

, Jiang

, Markowitz

: Age- and sex-related expression and activity of carboxylesterase 1 and 2 in mouse and human liver. Drug Metab Dispos, 37:1819–1825, 2009.

61.

Zhu

, Appel

, Peterson

, Wang

, Markowitz

: Identification of selected therapeutic agents as inhibitors of carboxylesterase 1: Potential sources of metabolic drug interactions. Toxicology, 270:59–65, 2010.

62.

Zhu

, Patrick

, Yuan

H–J

, Wang

J-S

, Donovan

, DeVane

, Malcolm

, Johnson

, Youngblood

, Sweet

, Langaee

, Markowitz

: Two CES1 gene mutations lead to dysfunctional carboxylesterase 1 activity in man: clinical significant and molecular basis. Am J Hum Genet, 82:1241–1248, 2008.