Role of serotonin 5-HT2C and histamine H1 Receptors in antipsychotic-Induced diabetes: A PharmacoEpidemiological-pharmacodynamic study in VigiBase

European Neuropsychopharmacology (]]]]) ], ]]]–]]]

http://dx.doi.org/10924-977X/& 2015 E

nCorrespondence31000 Toulouse, Fra

E-mail address: f

Please cite this artA.... European Neu

www.elsevier.com/locate/euroneuro

Role of serotonin 5-HT2C and histamine H1

receptors in antipsychotic-induced diabetes:A pharmacoepidemiological-pharmacodynamic study in VigiBase

François Montastruca,b,c,e,n, Aurore Palmaroa,b,e,Haleh Bagheria,b,c,e, Laurent Schmittd,Jean-Louis Montastruca,b,c,e, Maryse Lapeyre-Mestrea,b,e

aService de Pharmacologie Médicale et Clinique, Faculté de Médecine de Toulouse, Centre HospitalierUniversitaire de Toulouse, FrancebINSERM UMR 1027, Pharmacoépidémiologie, Evaluation de l'utilisation et du risque médicamenteux,Université de Toulouse, FrancecPharmacopôle Midi-Pyrénées, Centre Midi-Pyrénées de Pharmacovigilance, de Pharmacoépidémiologieet d'Informations sur le médicament, Toulouse, FrancedService Hospitalo-Universitaire de Psychiatrie et Psychologie Médicale, Faculté de Médecine de Toulouse,Centre Hospitalier Universitaire de Toulouse, FranceeInserm CIC 1436 Toulouse, CIC de Toulouse, Toulouse University Hospital, Place du Dr Baylac,31059 Toulouse Cedex 9, France

Received 20 April 2015; received in revised form 23 June 2015; accepted 14 July 2015

KEYWORDSSerotonin 5-HT2C;Histamine H1;Antipsychotics;Diabetes;Pharmacoepidemiol-ogy;Pharmacovigilance

0.1016/j.euroneurlsevier B.V. and E

to: Service de Phance.rancois.montastru

icle as: Montastruropsychopharmac

AbstractPharmacodynamic mechanisms of diabetes induced by antipsychotic drugs remain unclear,while numerous receptors have been suspected to be involved in the genesis of this AdverseDrug Reaction (ADR). We investigated potential relationships between antipsychotics' receptoroccupancy (serotonin 5-HT1A, 5-HT2A, 5-HT2C, histamine H1, muscarinic M3, adrenergic α1, α2or dopaminergic D2 D3 occupancies) and reports of diabetes using VigiBases, the World HealthOrganization (WHO) global Individual Case Safety Report (ICSR) database. All ADR reports from15 first and second generation antipsychotic drugs recorded in VigiBases were extracted.Logistic regression models, completed by disproportionality analysis, were used to determinethe associations between antipsychotics' receptor occupancy and ICSRs of diabetes onVigiBases. During the study period, 94,460 ICSRs involved at least one of the 15 antipsychoticsof interest. Diabetes was reported in 1799 (1.9%) patients. Clozapine was the most frequently

o.2015.07.010CNP. All rights reserved.

rmacologie Médicale et Clinique, Faculté de Médecine, 37 Allées Jules-Guesde,

[email protected] (F. Montastruc).

c, F., et al., Role of serotonin 5-HT2C and histamine H1 receptors in antipsychotic-induced diabetes:ology (2015), http://dx.doi.org/10.1016/j.euroneuro.2015.07.010

F. Montastruc et al.2

Please cite this article as: MontastruA.... European Neuropsychopharmac

suspected drug (n=953; 53.0%). A significant and positive association was found betweenhistamine H1, muscarinic M3 and serotonin 5-HT2C, 5-HT2A receptor occupancies and reports ofdiabetes. A multivariable stepwise regression model showed that only serotonin 5-HT2c(AOR=2.13, CI 95% 1.72–2.64) and histamine H1 (AOR=1.91, CI 95% 1.38–2.64) predicted therisk for diabetes mellitus (po0.001). Using an original pharmacoepidemiology-pharmacodynamic (PE-PD) approach, our study supports that antipsychotic drugs blockingsimultaneously histamine H1 and serotonin 5-HT2C receptors are more frequently associatedwith diabetes reports in VigiBases than other antipsychotics. These findings should encourageinvestigation of histamine H1 and serotonin 5-HT2C properties for predicting the risk ofglycemic effects in candidate antipsychotics.& 2015 Elsevier B.V. and ECNP. All rights reserved.

1. Introduction

Antipsychotics are known to induce metabolic disturbanceslike weight gain, dyslipidemia, glycemic dysregulations ordiabetes (Newcomer et al., 2002; Caro et al., 2002; Citromeet al., 2004). Over the last two decades, the risk of diabetesassociated to antipsychotics attracted considerable clinicalinterest in psychiatry. The main reason for this emerging issuewas a marked rise in second-generation antipsychotics' (SGAP)use, a class with less extrapyramidal Adverse Drug Reactions(ADRs) than first-generation antipsychotics (FGAP) (Geddeset al., 2000; Verdoux et al., 2010), but more likely to cause orexacerbate diabetes (Expert Group, 2004; Saddichha et al.,2008). Among SGAP, glycemic dysregulations and diabetes aremore frequent with olanzapine or clozapine than with risper-idone, quetiapine or aripiprazole (Starrenburg and Bogers,2009; Reynolds and Kirk, 2010; Meyer et al., 2008). The risk ofdiabetes associated with SGAP has been highlighted in severalstudies, while the underlying pharmacodynamic mechanismsremain largely unknown.

Interesting findings involving serotonin receptors havebeen obtained in animal studies. At least seven types ofserotonin receptors, widely expressed in the central nervoussystem, were characterized (Stahl, 2013). In particular,consequences of 5-HT2C receptor deletion were investigatedin murine knockout models. These studies revealed occur-rences of insulin resistance and type 2 diabetes after deletionof the gene encoding the 5-HT2C receptor in this animalmodel (Hahn et al., 2011; Nonogaki et al., 1998). Anotherstudy by Wade provides evidence that this deletion producesimpairment in glucose homeostasis, which does not seem tobe due exclusively to increased adiposity or body weight(Wade et al., 2008). Moreover, some studies have suggestedthat 5-HT2C receptors in hypothalamus could be involved inglucose homeostasis via the sympathetic nervous system (Qiuet al., 2007). In human, a double-blind, placebo-controlledcrossover trial comparing ketanserin (a 5-HT2C antagonist) toplacebo with regards to weight loss found a significantlydecreased insulin sensitivity in patients taking ketanserin(Gilles et al., 2005). These studies suggest that blockade of5-HT2C receptors has a primary effect on glucose homeostasisthat could be dissociated from its effects on body weight.

Other receptors have been involved in glycemic dysregula-tions observed with antipsychotics: serotonin 5-HT1A and 5-HT2A (Gilles et al., 2005; Nasrallah, 2008; Coccurello and

c, F., et al., Role of serotonin 5-Hology (2015), http://dx.doi.org/10.1

Moles, 2010), histamine H1 (Starrenburg and Bogers, 2009;Hahn et al. 2011; Matsui-Sakata et al., 2005), muscarinic M3

(Nasrallah, 2008; Silvestre and Prous, 2005), adrenergic α1and α2 (Chaouloff and Jeanrenaud, 1987; Sim et al., 2010),dopamine D2 and D3 (Hahn et al., 2011; Coccurello and Moles,2010; Kapur and Seeman, 2001). In particular, antagonism ofmuscarinic M3 and histamine H1 receptors could impairpancreatic insulin secretion and increase diabetes risk(Stahl et al., 2009). Thus, the binding properties of anti-psychotics on serotonin 5-HT1A, 5-HT2A, 5-HT2C, histamine H1,muscarinic M3, adrenergic α1, α2 or dopaminergic D2, D3

receptors are likely to explain occurrence of diabetes andeven variations in diabetes occurrence among drugs. How-ever, we found no consistent evidence that have explicitlylinked antipsychotics binding profiles to antipsychotics-related type 2 diabetes in humans. Finally, the precisepharmacodynamic mechanism still remains speculative.

Our aim was to explore the potential association betweenreceptor occupancy of several antipsychotic (serotonin 5-HT1A, 5-HT2A, 5-HT2C, histamine H1, muscarinic M3, adre-nergic α1, α2 or dopaminergic D2 D3 occupancies) and reportsof diabetes in day-to-day practice using VigiBases, theWorld Health Organization (WHO) monitoring database ofpharmacovigilance.

2. Experimental procedures

2.1. Design

We used logistic regression, completed by disproportionality analy-sis, to determine associations between receptor occupancies byantipsychotics and Individual Case Safety Reports (ICSRs) of dia-betes in the World Health Organization (WHO) global ICSR databaseVigiBases (Lindquist, 2008).

2.2. Data collection

ICSRs entered in VigiBase between 01/01/1994 (year of the firstdiabetes case-report with clozapine) to 29/03/2013 were analyzed.Since 1978, the Uppsala Monitoring Centre (UMC) has been respon-sible for the coordination of the WHO Program for InternationalDrug Monitoring (Lindquist and Edwards, 2001). The UMC collectscases of suspected ADRs and other drug-related problems sponta-neously reported by Health care professionals, hospitals, lawyers,manufacturers or patients via the national pharmacovigilance

T2C and histamine H1 receptors in antipsychotic-induced diabetes:016/j.euroneuro.2015.07.010

3Role of serotonin 5-HT2C and histamine H1 receptors

centers. Currently, VigiBases contains more than 10 million ICSRsfrom 120 countries. The likelihood that the drug has caused thereported event varies from case to case.

Available data through ICSR comprise administrative information(country, type of report, qualification of the reporter), patient data(gender, age and seriousness of the ADR), characteristics of thereported ADR (description with MedDRA terms (Medical Dictionaryfor Regulatory Activities. http://www.meddramsso.com), date ofonset reaction, outcome, and WHO assessment causality) and druginvolved (drug name, drug start- and stop dates, time to onset,dose, Indication, dechallenge and rechallenge). The level ofcompleteness of information varies from case to case. Content ofICSRs is described in details in the UMC website (The UppsalaMonitoring Centre – VigiBase®, 2015).

2.3. Patients

Data extraction was restricted to reports without missing values forage and gender. We analyzed only reports concerning adults (418years). Individual reports in the US were excluded since causalityassessment by medical staff are not always required in each reports(Strom et al., 2012).

2.4. Antipsychotic drugs of interest

All reports involving one of the 15 following FGAP or SGAP (amisul-pride, aripiprazole, asenapine, chlorpromazine, clozapine, fluphena-zine, haloperidol, olanzapine, paliperidone, perphenazine, pimozide,quetiapine, risperidone, sertindole, sulpiride) as suspected, inter-acting or concomitant drug, and for which informations on bindingaffinities were available, were included in the study.

2.5. Antipsychotics' receptor occupancy: definition andmeasurement

We estimated the degrees of receptor occupancy for serotonin 5-HT1A,5-HT2A, 5-HT2C, histamine H1, muscarinic M3, adrenergic α1, α2 ordopaminergic D2 D3 receptors. Degrees of receptor occupancy wereconsidered as a quantitative variable and were calculated according toan equation derived from the pharmacological receptor theory's model(Kenakin, 2004; Yamada et al., 2002) (see Table 1). Receptoroccupancy theory is a useful approach to predict quantitativelyreceptor-mediated pharmacological actions (Ayuhara et al., 2009):

Degree of receptor occupancyð%Þ ¼ ½Cr�ðKiþ Cr½ �Þ � 100

where [Cr] represents the concentration of unbound antipsychotic.[Cr] were estimated according the “therapeutic reference ranges”reported in the “AGNP Consensus Guidelines for Therapeutic DrugMonitoring in Psychiatry” (Hiemke et al., 2011) and data of plasmaticbinding reported in pharmacological reference textbooks (Bruntonet al., 2011; Brayfield, 2014).

Constant Ki characterizes affinity, defined as the ability of a drugto bind to a receptor, for antagonist drugs like antipsychotics. The Kirefers to the equilibrium dissociation constant of a ligand determinedin inhibition studies. The Ki for a given ligand is typically determinedin a competitive radioligand binding study by measuring bindinginhibition of a reference radioligand by the competing ligand ofinterest under equilibrium conditions (Rang et al., 2012).

We collected Ki of the 15 studied antipsychotics for 9 differentreceptors potentially involved in diabetes pathophysiology anddescribed above. These values were obtained through a search onthe website iPHACE (Integrative navigation in PHarmaCologicalspacE) (iPHACE, 2015) (see Supplemental Table 1). iPHACE repre-sents a new integrative conceptual framework to navigate in thepharmacological space covered by drugs, and contains data related

Please cite this article as: Montastruc, F., et al., Role of serotonin 5-HA.... European Neuropsychopharmacology (2015), http://dx.doi.org/10.1

to interactions between drugs and targets extracted from IUPHAR(The International Union of Basic and Clinical Pharmacology)database (IUPHAR, 2015) and PDSP (Psychoactive Drug ScreeningProgram) of the National Institute of Mental Health (PDSP – HomePage, 2015). The 15 selected drugs act on these receptors asantagonist agent.

2.6. Identification of diabetes cases

In VigiBases, reports of diabetes were identified using MedDRAdictionary with the following Preferred Terms (PT): “Decreasedinsulin requirement”, “Diabetes complicating pregnancy”, “Diabetesmellitus”, “Gestational diabetes”, “Increased insulin requirement”,“Insulin resistant diabetes”, “Insulin-requiring type 2 diabetes melli-tus”, “Type 2 diabetes mellitus”, and “Type 3 diabetes mellitus”.ADR terms related to diabetes complications were not included.

2.7. Potential confounders

Several potential confounders were identified. These factors included:gender, age, comorbidities (anorectic drugs (ATC codes A08 andA10BX06, as a proxy of overweight or obesity)), exposure to hypo-and hyperglycemic drugs, characteristics of the reporter (medicaldoctor, pharmacist, patient…), geographic area and time since firstmarketing of the antipsychotic drug. Time since marketing wascalculated as the year of reporting minus the year of first marketing.

Countries were divided in 10 geographical areas accordingclassification of the WHO regional offices (WHO | WHO RegionalOffices, 2015): 3 geographical (continental) regions (Africa, South-East Asia, Eastern Mediterranean) and 7 geographical sub-regions(Canada, Central/South America, Eastern Europe, Northern Europe,Southern Europe, Western Europe and Western Pacific).

2.8. Statistical methods

Association between receptors' occupancy and diabetes was inves-tigated using binary logistic regression models.

Quantitative values of occupancy, dichotomized into “high levelof occupancy” (Z50%) and “low level of occupancy” (o50%), wereincluded in the regression models. For instance, a report involvingolanzapine was quoted “High level of occupancy” for 5-HT2Creceptor and “High level of occupancy” for H1 receptor. If severaldrugs were involved, all levels of occupancies were entered. In caseof several antipsychotics with different receptors occupancies onthe same receptor, the highest degree of occupancy was selected.

Logistic regressions used diabetes cases as the dependentvariable. Potential explanatory variables included categorizedvalues of occupancies (“high level of occupancy” versus “low levelof occupancy” (reference)) of involved antipsychotic drugs regard-ing the 9 different receptors investigated. To compare reports ofdiabetes to other ADRs, we used Pearson's χ2 test or Fisher's exacttest for categorical variables (gender, WHO area…) and the t-testor Mann–Whitney parametric test for quantitative variables (degreeof receptor occupancy, age, time since first marketing) (α=5%). Foreach quantitative variable, we tested the log-linearity hypothesisand chose the corresponding adequate coding.

In a first step, simple univariate logistic models were constructedseparately for each receptor. Then, in a second step, all receptors'levels of occupancy were considered simultaneously in a singlelogistic regression model. In order to identify the most significantreceptor binding associated with diabetes' reports and to constructa relevant statistical model from a pharmacological point of view,receptors exhibiting a “protective effect” (ORo1) were not main-tained in the multivariable analysis. Then, receptor binding valuespositively associated with diabetes occurrence in the univariateanalysis were entered in a multivariable logistic model. Variables

T2C and histamine H1 receptors in antipsychotic-induced diabetes:016/j.euroneuro.2015.07.010

Table 1 Receptor occupancy for the 15 antipsychotics of interest.

Antipsychotics Adrenergicα1 (%)

Adrenergicα2 (%)

DopamineD2 (%)

DopamineD3 (%)

HistamineH1 (%)

MuscarinicM3 (%)

Serotonin5-HT1A (%)

Serotonin5-HT2A (%)

Serotonin5-HT2C (%)

Amisulpride 0.72 5.46 97.33 96.66 0.72 0.72 0.72 0.91 0.72Aripiprazole 3.56 1.48 42.98 24.76 3.81 0.02 16.55 7.47 1.86Asenapine 6.50 6.50 2.25 18.02 35.57 0.00 1.09 25.84 73.45Chlorpromazine 68.85 0.74 3.62 43.96 37.31 3.24 0.14 33.11 13.00Clozapine 68.27 7.89 5.97 3.77 86.40 26.31 4.30 62.55 49.01Fluphenazine 1.42 0.02 18.58 18.58 0.72 0.01 0.01 0.07 0.03Haloperidol 2.08 0.02 14.46 5.08 0.02 0.00 0.01 0.27 0.00Olanzapine 2.21 0.60 6.24 5.48 64.19 3.61 0.11 44.47 19.84Paliperidone 53.63 12.71 42.19 26.78 6.80 0.04 0.58 82.16 5.48Perphenazine 0.06 0.00 1.17 1.84 0.07 0.00 0.00 0.12 0.00Pimozide 0.03 0.00 2.66 4.15 0.01 0.00 0.01 0.14 0.00Quetiapine 58.31 0.27 5.63 7.77 73.62 1.62 5.63 10.00 1.73Risperidone 28.30 4.53 39.03 22.23 11.10 0.00 0.45 87.78 5.28Sertindole 4.13 0.03 3.70 2.48 0.06 0.00 0.03 18.45 18.11Sulpiride 0.35 5.26 52.58 87.49 0.02 1.73 21.78 0.14 0.18

Degrees of receptor occupancy were calculated according to an equation derived from the pharmacological receptor theory's model(Kenakin, 2004; Yamada et al., 2002).

Degree of receptor occupancy ð%Þ ¼ ½Cr �ðKiþ Cr½ �Þ � 100

where [Cr] represents the concentration of unbound antipsychotic. [Cr] were estimated according the “therapeutic reference ranges”reported in the “AGNP Consensus Guidelines for Therapeutic Drug Monitoring in Psychiatry” (Hiemke et al., 2011) and data of bound inplasma reported in pharmacological reference textbooks (Brunton et al., 2011; Brayfield, 2014). Constant Ki characterizes affinity,defined as the ability of a drug to bind to a receptor, for antagonist drugs like antipsychotics (see Supplemental Table 1)

F. Montastruc et al.4

from the final model were selected using a stepwise descendingprocedure. Fits of final models were checked using Hosmer andLemeshow χ2 test. The association between receptor's occupancyand diabetes was estimated as crude and an Adjusted Odds Ratioswith their corresponding 95% confidence interval (95% CI).

To complete the analysis, we performed a case/non-case com-parison between the 15 drugs to measure the disproportionality ofdrug-related diabetes reporting between each antipsychotic. Thecase/non-case method compares cases of diabetes ADRs to othercases of ADRs regarding the exposure to one antipsychotic incomparison to other antipsychotics (Montastruc et al., 2011;Rothman et al., 2004). The method allowed us to calculate anADR Reporting Odds Ratio (ROR) for risk factors of diabetes and itsconfidence interval. The ROR provides a ratio of the odds ofexposure in reports of cases and non-cases. The ROR was calculatedby dividing the number of cases with one antipsychotic as thesuspected drug divided by the number of cases with other anti-psychotics by the number of non-cases with the same antipsychoticas the suspected drug divided by the number of non-cases withother antipsychotics (see Supplemental Table 2).

All analyses were performed using SAS software, version 9.4 (SASInstitute. Inc., Cary. North Carolina, USA). Statistical significancewas defined as a p threshold of 0.05.

3. Results

3.1. Study population

During the study period, 94,460 reports involved at least oneof the 15 antipsychotics of interest. Male/female sex ratiowas 1.2 and mean age 43.7 years (717.1). The greatestnumber of reports concerned clozapine (43.6%) followed by

Please cite this article as: Montastruc, F., et al., Role of serotonin 5-HA.... European Neuropsychopharmacology (2015), http://dx.doi.org/10.1

risperidone (16.1%) and olanzapine (13.6%). Table 2 presentsthe characteristics of all reports regarding the 15 antipsy-chotics in VigiBase®. Reports came mainly from physicians,Northern Europe and top year of reporting was 2009. During2007, a new import process was developed at the UMC withthe consequence that almost none ICSR was processed thatyear. The majority of ICSRs received in 2007 were processedthe following year, and 2008 therefore shows a peak.

3.2. Main analysis

Diabetes mellitus was reported in 1780 (1.9%) of thepatients. Clozapine was the most frequent suspected drug(n=951; 53.4%), followed by olanzapine (n=545; 30.6%) andrisperidone (n=138; 7.8%). Since 1994, reports of diabetesincreased, with a peak in 2011.

Table 3 shows the main characteristics of diabetes'reports and all other reports. Cases of diabetes includedpatients significantly older, with more hypo- or hyper-glycemic drugs, mainly from Northern Europe. These caseswere frequently reported by patients and physicians.

A significant and positive association was found betweenhistamine H1, muscarinic M3 and serotonin 5-HT2C, 5-HT2Areceptor occupancies and reports of diabetes (Table 4). Inthe multivariate models, in comparison to reports with low5-HT2C receptor occupancies (reference), Adjusted OddsRatio (aOR) for developing diabetes was 2.89 (95% CI,2.41–3.46) for high 5-HT2C receptor occupancies. Similarly,a positive association was found for serotonin 5-HT2Areceptor (aOR 3.65, 95% CI 2.76–4.82), histamine H1

T2C and histamine H1 receptors in antipsychotic-induced diabetes:016/j.euroneuro.2015.07.010

Table 2 Characteristics of all reports for the 15 antipsychotics of interest in VigiBase (n=94,460).

Characteristics Mean age (year)

Age group. Year (n, %)0–30 24,543 26.0 24.731–45 32,036 33.9 37.946–65 25,794 27.3 54.0465 12,087 12.8 76.1Female (n, %) 42,940 45.5 46.6Reporting year (median, min–max) 2009 1994–2013 43.3Years since marketing (n, %)o10 21,239 22.5 42.910–15 24,092 25.5 43.915–20 21,025 22.3 42.5420 28,104 29.8 45.2Hypoglycemic drugs (n, %) 1846 1.9 54.8Hyperglycemic drugs (n, %) 4657 4.9 53.7Anorectic drugs (n, %) 74 0.1 38.8WHO area (n, %)Africa 554 0.6 40.6Americas 4604 4.9 38.4Canada 6766 7.2 47.2South-East Asia 3061 3.2 37.5Eastern Mediterranean 344 0.4 36.8Western Pacific 15,201 16.1 41.9Southern Europe 7494 7.9 46.9Northern Europe 33,050 35.0 43.0Eastern Europe 830 0.9 40.5Western Europe 22,556 23.9 46.3Reporter (n, %)Physicians 41,995 44.5 44.3Pharmacist 3681 3.9 48.7Others Health professional 5078 5.4 43.5Patients 1604 1.7 42.0Literature 1629 1.7 45.2Hospitals 12,438 13.2 43.2Drug companies 4702 5.0 41.6Others 8445 8.9 43.6Unknown 14,888 15.8 42.4

5Role of serotonin 5-HT2C and histamine H1 receptors

receptor (3.89, 95% CI 2.94–5.13) and muscarinic M3 (2.87,95% CI 2.50–3.37). We were unable to find any positiveassociation with the remaining receptors.

The second analysis (a stepwise regression multivariablemodel) including receptor binding values positively asso-ciated (i.e. serotonin 5-HT2A, 5-HT2C, histamine H1, mus-carinic M3 receptors) and adjusting for all potentialconfounders, showed that only serotonin 5-HT2c(AOR=2.13, CI, 95% 1.72–2.64) and histamine H1

(AOR=2.04, CI 95% 1.46–2.84) receptor occupancy pre-dicted the risk for diabetes mellitus (po0.001).

3.3. Case/non-case analysis

Results of disproportionality analysis are presented inFigure 1. The association between antipsychotic drugs anddiabetes was significant for two SGAP. ROR estimates were2.64 (CI 95%, 2.39–2.92) for olanzapine and 1.49 (CI 95%,1.36–1.64) for clozapine.

Please cite this article as: Montastruc, F., et al., Role of serotonin 5-HA.... European Neuropsychopharmacology (2015), http://dx.doi.org/10.1

In Supplemental Figures 1 and 2 , degrees of 5-HT2c and H1

receptor occupancy of antipsychotics are plotted against thelogarithm of ROR, together with a line describing the logisticmodel. Drugs that did not follow the predicted model werehaloperidol, perphenazine, paliperidone fluphenazine and sulpir-ide (for which degrees of receptor occupancies were too lower).

4. Discussion

4.1. Main results

This study investigated the association between degrees ofantipsychotics' receptor occupancy and subsequent risk ofdiabetes in day-to-day practice using VigiBases, the WHOmonitoring database of pharmacovigilance. We found apositive and statistically significant association between 5-HT2C and H1 receptor degrees of occupancy and diabetesoccurrences. This association persisted after adjustment onpatients' characteristics, type and time of reports and

T2C and histamine H1 receptors in antipsychotic-induced diabetes:016/j.euroneuro.2015.07.010

Table 3 Univariate and multivariate binary logistic regression for diabetes (n=94,460).

Reports characteristic Cases (n=1780) Non-cases (n=92,680) Crude OR 95% CI AORa 95% CI p

Age. Year (n)0–30 312 24,231 1 1 o0.00131–45 759 31,277 1.88 1.65–2.15 1.75 1.53–2.0046–65 617 25,177 1.90 1.66–2.20 1.62 1.41–1.87465 92 11,995 0.60 0.46–0.74 0.54 0.43–0.69Female (n) 697 42,243 0.76 0.69–0.84 0.81 0.73–0.90Reporting year (mean) 2008 2007 1.03 1.02–1.04Years since marketing (n)o10 355 20,884 1 1 o0.00110–15 555 23,537 1.39 1.21–1.59 1.37 1.19–1.5715–20 457 20,568 1.31 1.14–1.50 1.17 1.00–1.36420 413 27,691 0.88 0.76–1.00 0.71 0.60–0.83Hypoglycemic drugs (n) 173 1673 5.83 4.95–6.87 5.54 4.66–6.59 o0.001Hyperglycemic drugs (n) 137 4520 1.61 1.35–1.92 1.38 1.15–1.66 o0.001Anorectic drugs (n) 3 71 2.20 0.69–6.93 –

WHO area (n)Western Europe 309 22,247 1 1 o0.001Africa 12 542 1.59 0.89–2.85 1.68 0.93–3.00Americas 6 4598 0.09 0.04–0.21 0.10 0.05–0.23Canada 217 6549 2.39 2.00–2.84 1.82 1.50–2.20South-East Asia 15 3046 0.35 0.21–0.59 0.47 0.28–0.80Eastern Mediterranean 2 342 0.42 0.10–1.70 0.55 0.13–2.22Western Pacific 215 14,986 1.03 0.87–1.23 0.85 0.69–1.05Southern Europe 68 7426 0.66 0.51–0.86 0.75 0.58–0.98Northern Europe 927 23,123 2.07 1.82–2.36 2.33 2.01–2.71Eastern Europe 9 821 0.79 0.40–1.55 0.75 0.39–1.47Reporter (n)Physicians 784 41,211 1 1 o0.001Pharmacist 43 3638 0.62 0.46–0.85 0.56 0.41–0.76Others Health professional 78 5000 0.82 0.65–1.04 0.64 0.50–0.81Patients 89 1515 3.09 2.46–3.87 3.54 2.78–4.49Literature 24 1605 0.78 0.52–1.18 0.82 0.54–1.24Hospitals 115 12,323 0.49 0.40–0.60 0.71 0.58–0.87Drug companies 87 4615 0.99 0.79–1.24 1.60 1.20–2.14Others 171 8274 1.08 0.92–1.28 0.92 0.77–1.12Unknown 389 8274 1.41 1.23–1.60 0.77 0.66–0.89

OR: Odds Ratio; AOR: Adjusted Odds Ratio; CI: Confidence Interval.aAdjusted for age, gender, years since marketing, hypoglycemic drugs, hyperglycemic drugs, anorectic drugs, type of reporter and

WHO area.

F. Montastruc et al.6

concomitant medications. Olanzapine and clozapine werethe main antipsychotics involved in diabetes reports.

4.2. Relation with previous experimental findings

Involvement of histamine H1 receptors in the regulation ofenergy homeostasis has been described in several animals'models (Sakata et al., 1990; Lecklin et al., 1998). HistamineH1 receptors in the ventromedial hypothalamic nucleus andin the paraventricular nucleus of the brain play a central rolein food intake and metabolic regulation (Masaki et al., 2004;Gotoh et al., 2013). Mice without histamine H1 receptorsactivity develop obesity and glycemic dysregulation(Jørgensen et al., 2006). Therefore, inhibition of histamine

Please cite this article as: Montastruc, F., et al., Role of serotonin 5-HA.... European Neuropsychopharmacology (2015), http://dx.doi.org/10.1

H1 receptors could lead to dysregulation of feeding rhythm,weight gain and insulin resistance. Drug-induced insulinresistance is not only in relation with weight gain, but alsoseems to be linked with the increased level of leptin(Morimoto et al., 1999; Melkersson and Hulting, 2001).Antagonism of histamine H1 receptors may impair metabolicfunction of leptin, leading to insulin resistance and impairedglucose tolerance (Coccurello and Moles, 2010).

As highlighted in our results, glucose sensitivity may bealtered by the action of antipsychotics through an effect onserotonin 5-HT2C receptors. Hypothalamic 5-HT2C receptorsmediate glucose homeostasis via the sympathetic nervoussystem, which then activate glucose metabolic organs (adi-pose tissues, liver, muscle) (Zhou et al., 2007). As explainedin introduction, insulin resistance and type 2 diabetes have

T2C and histamine H1 receptors in antipsychotic-induced diabetes:016/j.euroneuro.2015.07.010

Table 4 Univariate and multivariate binary logistic regression for diabetes reports including receptor occupancies.

Receptor Diabetes reports Crude OR CI 95% AORa CI 95%

Cases n (%) Non-cases n (%)

Adrenergic α1 1181 (66.4) 66,078 (71.3) 0.79 0.72–0.88 0.52 0.46–0.58Adrenergic α2 1103 (62.0) 59,813 (64.5) 0.89 0.81–0.99 0.64 0.57–0.71Dopamine D2 193 (10.8) 29,823 (32.2) 0.26 0.22–0.30 0.32 0.28–0.38Dopamine D3 190 (10.7) 25,612 (27.6) 0.31 0.27–0.36 0.35 0.30–0.41Histamine H1 1726 (97.00) 77,857 (84.0) 6.08 4.64–7.98 3.89* 2.94–5.13Muscarinic M3 1597 (89.7) 64,618 (69.7) 3.79 3.25–4.42 2.87* 2.50–3.37Serotonin 5-HT1A 1224 (68.8) 70,285 (75.8) 0.70 0.63–0.78 0.44 0.40–0.50Serotonin 5-HT2A 1726 (97.0) 78,333 (84.5) 5.85 4.46–7.67 3.65* 2.76–4.82Serotonin 5-HT2C 1639 (92.1) 71,283 (76.9) 3.49 2.93–4.15 2.89* 2.41–3.46

OR: Odds Ratio; AOR: Adjusted Odds Ratio; CI: Confidence IntervalFor each receptor: high receptor occupancy (Z50%) versus low receptor occupancy (o50%) set as the baseline level.

aAdjusted for age, gender, years since marketing, hypoglycemic drugs, hyperglycemic drugs, anorectic drugs, type of reporter andWHO area.

nPositively associated (po0.001).

Figure 1 Association between antipsychotic exposure and dia-betes occurrence in VigiBases according to the case–non-case study.Log ROR: Logarithm of Reporting Odds Ratio; 95% CI: 95% ConfidenceInterval. Antipsychotics with less than 3 reports of diabetes are notpresented (i.e. asenapine, sertindole and pimozide).

7Role of serotonin 5-HT2C and histamine H1 receptors

been observed after deletion of gene encoding the 5-HT2Creceptor in animal models (Hahn et al., 2011; Nonogakiet al., 1998). Moreover, some pharmacogenomic studies haveidentified an association between serotonin 5-HT2C receptorpolymorphisms and occurrence of metabolic syndrome inpatients using antipsychotics (Kang et al., 2014; Risseladaet al., 2012; Mulder et al., 2009). Recent clinical studies alsosuggested that diabetic patients using lorcaserin, a serotonin5-HT2C agonist (Nigro et al., 2013), as an appetite suppres-sant, had significant reductions in their HbA1c level.

Our multivariable model showed that both histamine H1

and serotonin 5-HT2C receptors are independently asso-ciated with the risk of diabetes. Then, the combinedblockade of histamine H1 and serotonin 5-HT2C receptors isexpected to be highly associated with diabetes. Thisassumption is supported by our case/non-case study. Thecombined blockade of histamine H1 and serotonin 5-HT2C

Please cite this article as: Montastruc, F., et al., Role of serotonin 5-HA.... European Neuropsychopharmacology (2015), http://dx.doi.org/10.1

receptors could explain why SGAP such as olanzapine orclozapine, which have high degree of occupancy of H1 and5-HT2C receptors, were the most associated with diabetes inour case/non-case study. In contrast, chlorpromazine withhigh histamine H1 antagonist property and low 5-HT2Ceffects was not associated in the disproportionality analysis(ROR =0.3; CI 95% [0.19–0.48]). This results confirm themechanism proposed by Reynolds and Kirk (2010) and Stahlet al. (2009) involving a combined H1 and 5-HT2C blockadeto explain metabolic disturbances with antipsychotics.

4.3. Limitations

Our analyses are subject to the inherent limitations of data-mining approaches from pharmacovigilance databases. Themain drawback of this data is under-reporting (Martin et al.,1998; Montastruc et al., 2006). However, there is no reason formajor differential underreporting among cases and non-cases,excepted for olanzapine and clozapine reports. The highnumbers of reports with these two drugs suggest a notorietybias (due to safety alerts) that would have underestimatedRORs (Pariente et al., 2007) in the case/non-case approach.Therefore, we did a sensitivity analysis to test potentialcompetition biases (Salvo et al., 2013; Pariente et al.,2012). Drug-related competition bias is caused by numerousreports of the ADR of interest (here, diabetes) involving otherdrugs than the drugs of interest (here, olanzapine or cloza-pine), thus increasing the relative exposure to other drugsamong the cases, and also underestimating the ROR of thedrug of interest (Pariente et al., 2012). Consequently, in thesensitivity analysis for case/non-case study, we excludedolanzapine or clozapine (drug competitors). These sensitivityanalyses led to similar results (data not shown).

Measurement of receptor occupancy has several limita-tions. First, we restricted analyses to antipsychotics forwhich receptor binding properties as well as therapeuticreference ranges are studied, and published. Second, the [Cr](the concentration of unbound antipsychotic) was based onthe therapeutic reference ranges at recommended doses

T2C and histamine H1 receptors in antipsychotic-induced diabetes:016/j.euroneuro.2015.07.010

F. Montastruc et al.8

(Hiemke et al., 2011). Moreover, data on Cerebral SpinalFluid concentrations always remains unknown. No sufficientdetail on doses used by patients was reported in ICSRs. Thirdlimitation is that level of occupancy is based on theconcentration and Ki, but this does not account the intrinsicactivity of drugs. However, the investigated antipsychotics doact as antagonists on receptors associated with diabetes onthe multivariable model, (i.e. serotonin 5-HT2C and hista-mine H1 receptors).

We did not consider US reports because some of them arenot always assessed by medical staff. Additionally, wehypothesized that including US reports would have led tooverestimate the magnitude of the association found, dueto the imbalance in reporting across antipsychotics. Indeed,the notoriety bias in favor of olanzapine and clozapine biasseems to be far more important in US due to several classactions (Lenzer, 2007). Including or not this geographicalarea would not have affected our results, since relationsbetween receptors occupancy and diabetes occurrence isnot likely to be impacted by the population considered.

4.4. Originality

To the best of our knowledge, this is the first study examiningdiabetes related to antipsychotics in day-to-day clinicalpractice using a pharmacovigilance database, through ananalysis of different receptor occupancies. This originalapproach should be designed as a pharmacoepidemiology-pharmacodynamic study (PE-PD study). Previous studies(Matsui-Sakata et al., 2005; Silvestre and Prous, 2005) havealready approximated the risk of diabetes using resultsavailable in the literature. In contrast to our work, thesestudies could not adjust the outcome on potential confoun-ders. The unique study using a comparable design wasperformed by DeBruin et al. on VigiBases, who used a similarmethod to investigate the relation between anti-HERGactivity and risk of cardiac arrhythmias (De Bruin et al.,2005). However, the authors only considered one receptorvalue (anti-HERG activity) in their multivariable models.

One important strength of our study is the data source,VigiBase. It is the largest pharmacovigilance database in theworld, which enables to evaluate unfrequently reported ADRs(Star et al., 2011). VigiBase allowed us to investigate risk ofdiabetes in one of the biggest sample of patient (n=94,460)exposed to antipsychotics. Consequently, the sample of diabetesreports involving antipsychotics is relatively important (n=1799).

5. Conclusion

Using an original pharmacoepidemiology-pharmacodynamic(PE-PD) approach, our study supports for the first time inhumans that antipsychotic drugs blocking simultaneouslyhistamine H1 and serotonin 5-HT2C receptors, such asolanzapine and clozapine, are more associated with dia-betes reports in VigiBases than other antipsychotics withdifferent binding profiles, suggesting a higher glycemicdysregulation risk. These findings should encourage investi-gation of histamine H1 and serotonin 5-HT2C propertiesduring pre-clinical tests in order to predict the risk ofglycemic effects in candidate antipsychotics.

Please cite this article as: Montastruc, F., et al., Role of serotonin 5-HA.... European Neuropsychopharmacology (2015), http://dx.doi.org/10.1

Conflict of interest

None to declare.

Contributors

François Montastruc, Haleh BagherI, Laurent Schmitt, Jean-LouisMontastruc, Maryse Lapeyre-Mestre participated in the study con-ception and planning. François Montastruc, Aurore Palmaro, MaryseLapeyre-Mestre took part in the acquisition, analysis or interpreta-tion of data. François Montastruc, Aurore Palmaro, Jean-LouisMontastruc, Maryse Lapeyre-Mestre drafted the manuscript. Fran-çois Montastruc, Aurore Palmaro, Kristina Star, Haleh BagherI,Laurent Schmitt, Jean-Louis Montastruc, Maryse Lapeyre-Mestrecritically revised the manuscript for important intellectual content.All the authors approved the final version of the manuscript.

Role of funding source

François Montastruc was a beneficiary of the “Année-Recherche”funding of the French Ministry of Research and Education (Faculty ofMedicine, University of Toulouse, France). He was also funded byINSERM UMR 1027 (Toulouse, France) to participate at 15th inter-national pharmacovigilance training course, which will take place inUppsala Monitoring Centre, Sweden, in May, 2013.

Acknowledgments

The authors would like to thank Kristina Star and the UppsalaMonitoring Centre (UMC) that provided and gave permission to usethe data analyzed in the present study.

The authors are indebted to the National Pharmacovigilance Centresthat contributed with data. The opinions and conclusions in this studyare not necessarily those of the various centers or of the WHO.

Appendix A. Supporting information

Supplementary data associated with this article can befound in the online version at http://dx.doi.org/10.1016/j.euroneuro.2015.07.010.

References

Ayuhara, Hideaki, Takayanagi, Risa, Okuyama, Kiyoshi, Yoshimoto,Koichi, Ozeki, Takeshi, Yokoyama, Haruko, Yasuhiko, Yamada,2009. Receptor occupancy theory-based analysis of interindivi-dual differences in antiemetic effects of 5-HT3 receptorantagonists. Int. J. Clin. Oncol. 14 (6), 518–524. http://dx.doi.org/10.1007/s10147-009-0912-5.

Brayfield, Alison, 2014. Martindale: The Complete Drug Reference,38th revised edition Pharmaceutical Press, United Kingdom.

Brunton, Laurence, Chabner, Bruce A., Knollman, Bjorn, 2011.Goodman and Gilman's the Pharmacological Basis of Therapeu-tics, 12th edition McGraw-Hill Medical, New York.

Caro, J Jaime, Alexandra, Ward, Carey, Levinton, Kimberly, Robin-son, 2002. The risk of diabetes during olanzapine use comparedwith risperidone use: a retrospective database analysis. J. Clin.Psychiatry 63 (12), 1135–1139.

Chaouloff, F., Jeanrenaud, B., 1987. 5-HT1A and alpha-2 adrenergicreceptors mediate the hyperglycemic and hypoinsulinemiceffects of 8-hydroxy-2-(di-N-propylamino)tetralin in the con-scious rat. J. Pharmacol. Exp. Ther. 243 (3), 1159–1166.

T2C and histamine H1 receptors in antipsychotic-induced diabetes:016/j.euroneuro.2015.07.010

9Role of serotonin 5-HT2C and histamine H1 receptors

Citrome, Leslie, Jaffe, A.r.i., Levine, Jerome, Allingham, Baerbel,James, Robinson, 2004. Relationship between antipsychoticmedication treatment and new cases of diabetes among psy-chiatric inpatients. Psychiatr. Serv. 55 (9), 1006–1013. http://dx.doi.org/10.1176/appi.ps.55.9.1006.

Coccurello, Roberto, Moles, Anna, 2010. Potential mechanisms ofatypical antipsychotic-induced metabolic derangement: cluesfor understanding obesity and novel drug design. Pharmacol.Ther. 127 (3), 210–251. http://dx.doi.org/10.1016/j.pharmthera.2010.04.008.

De Bruin, M.L., Pettersson, M., Meyboom, R.H.B., Hoes, A.W.,Leufkens, H.G.M., 2005. Anti-HERG activity and the risk ofdrug-induced arrhythmias and sudden death. Eur. Heart J. 26(6), 590–597. http://dx.doi.org/10.1093/eurheartj/ehi092.

Expert Group ‘Schizophrenia and Diabetes 2003’ Expert ConsensusMeeting, 2004, Dublin, 3–4 October 2003: consensus summary,Br. J. Psychiatry 184 (Suppl. 47), S112-S114.

Geddes, J., Freemantle, N., Harrison, P., Bebbington, P., 2000. Atypicalantipsychotics in the treatment of schizophrenia: systematic over-view and meta-regression analysis. BMJ 321, 1371–1376.

Gilles, Maria, Annette, Wilke, Daniel, Kopf, Annemarie, Nonell,Hendrik, Lehnert, Michael, Deuschle, 2005. Antagonism of theserotonin (5-HT)-2 receptor and insulin sensitivity: implicationsfor atypical antipsychotics. Psychosom. Med. 67 (5), 748–751.http://dx.doi.org/10.1097/01.psy.0000174994.91245.34.

Gotoh, Koro, Masaki, Takayuki, Chiba, Seiichi, Ando, Hisae, Fujiwara,Kansuke, Shimasaki, Takanobu, Mitsutomi, Kimihiko, 2013. Brain-derived neurotrophic factor, corticotropin-releasing factor, andhypothalamic neuronal histamine interact to regulate feedingbehavior. J. Neurochem. 125 (4), 588–598. http://dx.doi.org/10.1111/jnc.12213.

Hahn, Margaret, Chintoh, Araba, Giacca, Adria, Xu, Li, Loretta,Lam, Steve, Mann, Paul, Fletcher, 2011. Atypical antipsychoticsand effects of muscarinic, serotonergic, dopaminergic andhistaminergic receptor binding on insulin secretion in vivo: ananimal model. Schizophr. Res. 131 (1–3), 90–95. http://dx.doi.org/10.1016/j.schres.2011.06.004.

Hiemke, C., Baumann, P., Bergemann, N., Conca, A., Dietmaier, O.,Egberts, K., Fric, M., 2011. AGNP consensus guidelines fortherapeutic drug monitoring in psychiatry: update 2011. Phar-macopsychiatry 44 (6), 195–235.

iPHACE,2015. Available from: ⟨http://agave.health.unm.edu/iphace/main.php?text=fluoxetine&source=all⟩. (accessed 2015).

IUPHAR – International Union of Basic and Clinical Pharmacology,2015. Available from: ⟨http://www.iuphar.org/⟩. (accessed 2015).

Jørgensen, Emilie A., Vogelsang, Thomas W., Knigge, Ulrich,Watanabe, Takeshi, Warberg, Jørgen, Andreas, Kjaer, 2006.Increased susceptibility to diet-induced obesity in histamine-deficient mice. Neuroendocrinology 83 (5–6), 289–294. http://dx.doi.org/10.1159/000095339.

Kang, Shi Hyun, Jong-il Lee, Hye Ree, Han, Minah Soh, Jin, PyoHong, 2014. Polymorphisms of the leptin and HTR2C genes andclozapine-induced weight change and baseline BMI in patientswith chronic schizophrenia. Psychiatr. Genet. 24 (6), 249–256.http://dx.doi.org/10.1097/YPG.0000000000000053.

Kapur, S., Seeman, P., 2001. Does fast dissociation from thedopamine d(2) receptor explain the action of atypical antipsy-chotics? A new hypothesis. Am. J. Psychiatry 158 (3), 360–369.

Kenakin, Terry, 2004. Principles: receptor theory in pharmacology.Trends Pharmacol. Sci. 25 (4), 186–192. http://dx.doi.org/10.1016/j.tips.2004.02.012.

Lecklin, A., Etu-Seppälä, P., Stark, H., Tuomisto, L., 1998. Effects ofintracerebroventricularly infused histamine and selective H1, H2and H3 agonists on food and water intake and urine flow inWistar rats. Brain Res. 793 (1–2), 279–288.

Lenzer, Jeanne, 2007. Drug company seeks to suppress internalmemos. BMJ 334 (7584), 59. http://dx.doi.org/10.1136/bmj.39090.484074.DB.

Please cite this article as: Montastruc, F., et al., Role of serotonin 5-HA.... European Neuropsychopharmacology (2015), http://dx.doi.org/10.1

Lindquist, Marie, 2008. VigiBase, the WHO global ICSR databasesystem: basic facts. Drug Inform. J. 42 (5), 409–419. http://dx.doi.org/10.1177/009286150804200501.

Lindquist, M., Edwards, I.R., 2001. The WHO programme forinternational drug monitoring, its database, and the technicalsupport of the uppsala monitoring center. J. Rheumatol. 28 (5),1180–1187.

Martin, R.M., Kapoor, K.V., Wilton, L.V., Mann, R.D., 1998. Under-reporting of suspected adverse drug reactions to newly mar-keted (“black triangle”) drugs in general practice: observationalstudy. BMJ 317 (7151), 119–120.

Masaki, Takayuki, Chiba, Seiichi, Yasuda, Tohru, Noguchi, Hitoshi,Kakuma, Tetsuya, Watanabe, Takeshi, Sakata, Toshiie, Hironobu,Yoshimatsu, 2004. Involvement of hypothalamic histamine H1receptor in the regulation of feeding rhythm and obesity.Diabetes 53 (9), 2250–2260.

Matsui-Sakata, Akiko, Ohtani, Hisakazu, Sawada, Yasufumi, 2005.Receptor occupancy-based analysis of the contributions ofvarious receptors to antipsychotics-induced weight gain anddiabetes mellitus. Drug Metab. Pharmacokinet. 20 (5), 368–378.

Melkersson, K.I., Hulting, A.L., 2001. Insulin and leptin levels inpatients with schizophrenia or related psychoses—a comparisonbetween different antipsychotic agents. Psychopharmacology154 (2), 205–212.

Meyer, Jonathan M, Davis, Vicki G, Goff, Donald C, McEvoy, JosephP, Nasrallah, Henry A, Davis, Sonia M, Rosenheck, Robert A,2008. Change in metabolic syndrome parameters with antipsy-chotic treatment in the CATIE schizophrenia trial: prospectivedata from phase 1. Schizophr. Res. 101 (1–3), 273–286. http://dx.doi.org/10.1016/j.schres.2007.12.487.

Montastruc, Jean-Louis, Agnès, Sommet, Haleh, Bagheri, Maryse,Lapeyre-Mestre, 2011. Benefits and strengths of thedisproportionality analysis for identification of adverse drugreactions in a pharmacovigilance database. Br. J. Clin.Pharmacol. 72 (6), 905–908. http://dx.doi.org/10.1111/j.1365-2125.2011.04037.x.

Montastruc, Jean-Louis, Sommet, Agnès, Isabelle, Lacroix, Pascale,Olivier, Geneviève, Durrieu, Christine, Damase-Michel, Maryse,Lapeyre-Mestre, Haleh, Bagheri, 2006. Pharmacovigilance forevaluating adverse drug reactions: value, organization, andmethods. Joint Bone Spine: Rev. Du Rhum. 73 (6), 629–632.http://dx.doi.org/10.1016/j.jbspin.2006.09.002.

Morimoto, T., Yamamoto, Y., Mobarakeh, J.I., Yanai, K., Watanabe,T., Watanabe, T., Yamatodani, A., 1999. Involvement of thehistaminergic system in leptin-induced suppression of foodintake. Physiol. Behav. 67 (5), 679–683.

Mulder, Hans, Dan, Cohen, Hans, Scheffer, Christine, Gispen-deWied, Johan, Arends, Frederik, W.Wilmink, Barbara, Franke,Antoine, C.G.Egberts, 2009. HTR2C gene polymorphisms and themetabolic syndrome in patients with schizophrenia: areplication study. J. Clin. Psychopharmacol. 29 (1), 16–20.http://dx.doi.org/10.1097/JCP.0b013e3181934462.

Nasrallah, H.A., 2008. Atypical antipsychotic-induced metabolic sideeffects: insights from receptor-binding profiles. Mol. Psychiatry 13(1), 27–35. http://dx.doi.org/10.1038/sj.mp.4002066.

Newcomer, John W., Dan, W. Haupt, Robert, Fucetola, Angela, K.Melson, Julie, A. Schweiger, Benjamin, P. Cooper, Gregg, Selke,2002. Abnormalities in glucose regulation during antipsychotictreatment of schizophrenia. Arch. Gen. Psychiatry 59 (4),337–345.

Nigro, Stefanie C., Luon, Darren, Baker, William L., 2013. Lorca-serin: a novel serotonin 2C agonist for the treatment of obesity.Curr. Med. Res. Opin. 29 (7), 839–848. http://dx.doi.org/10.1185/03007995.2013.794776.

Nonogaki, K., Strack, A.M., Dallman, M.F., Tecott, L.H., 1998.Leptin-independent hyperphagia and type 2 diabetes in micewith a mutated serotonin 5-HT2C receptor gene. Nat. Med. 4(10), 1152–1156. http://dx.doi.org/10.1038/2647.

T2C and histamine H1 receptors in antipsychotic-induced diabetes:016/j.euroneuro.2015.07.010

F. Montastruc et al.10

Pariente, Antoine, Avillach, Paul, Salvo, Francesco, Thiessard,Frantz, Miremont-Salamé, Ghada, Fourrier-Reglat, Annie, Har-amburu, Françoise, Bégaud, Bernard, Moore, Nicholas, Associa-tion Française des Centres Régionaux de Pharmacovigilance(CRPV), 2012. Effect of competition bias in safety signal gen-eration: analysis of a research database of spontaneous reportsin France. Drug Saf.: Int. J. Med. Toxicol. Drug Exp. 35 (10),855–864. http://dx.doi.org/10.2165/11631780-000000000-00000.

Pariente, Antoine, Gregoire, Fleur, Fourrier-Reglat, Annie, Haram-buru, Françoise, Nicholas, Moore, 2007. Impact of safety alertson measures of disproportionality in spontaneous reportingdatabases: the notoriety bias. Drug Saf. 30 (10), 891–898.

« PDSP – Home Page, 2015. Available from: ⟨http://pdsp.med.unc.edu/indexR.html⟩. (accessed 2015).

Qiu, Jian, Xue, Changhui, Bosch, Martha A., Murphy, Jonathan G,Fan, W.e.i., Rønnekleiv, Oline K., Kelly, Martin J., 2007.Serotonin 5-hydroxytryptamine2C receptor signaling in hypotha-lamic proopiomelanocortin neurons: role in energy homeostasisin females. Mol. Pharmacol. 72 (4), 885–896. http://dx.doi.org/10.1124/mol.107.038083.

Rang H.P., Dale M.M., Ritter J.M., Flower R.J., Henderson G., 2012.Rang & Dale's Pharmacology. 7th ed.

Reynolds, Gavin P, Kirk, L. Shona, 2010. Metabolic side effects ofantipsychotic drug treatment—pharmacological mechanisms.Pharmacol. Ther. 125 (1), 169–179. http://dx.doi.org/10.1016/j.pharmthera.2009.10.010.

Risselada, A.J., Vehof, J., Bruggeman, R., Wilffert, B., Cohen, D.,Al Hadithy, A.F., Arends, J., Mulder, H., 2012. Associationbetween HTR2C gene polymorphisms and the metabolic syn-drome in patients using antipsychotics: a replication study.Pharmacogenomics J. 12 (1), 62–67. http://dx.doi.org/10.1038/tpj.2010.66.

Rothman, Kenneth J, Stephan, Lanes, Susan, T Sacks, 2004. Thereporting odds ratio and its advantages over the proportionalreporting ratio. Pharmacoepidemiol. Drug Saf. 13 (8), 519–523.http://dx.doi.org/10.1002/pds.1001.

Saddichha, S., Manjunatha, N., Ameen, S., Akhtar, S., 2008.Diabetes and Schizophrenia – effect of disease or drug? resultsfrom a randomized, double-blind, controlled prospective studyin first-episode schizophrenia. Acta Psychiatr. Scand. 117 (5),342–347. http://dx.doi.org/10.1111/j.1600-0447.2008.01158.x.

Sakata, T., Fukagawa, K., Ookuma, K., Fujimoto, K., Yoshimatsu,H., Yamatodani, A., Wada, H., 1990. Hypothalamic Neuronalhistamine modulates ad libitum feeding by rats. Brain Res. 537(1–2), 303–306.

Salvo, Francesco, Leborgne, Florent, Thiessard, Frantz, Moore,Nicholas, Bégaud, Bernard, Antoine, Pariente, 2013. A potentialevent-competition bias in safety signal detection: results from aspontaneous reporting research database in France. Drug Saf. 36(7), 565–572. http://dx.doi.org/10.1007/s40264-013-0063-5.

Silvestre, J.S., Prous, J., 2005. Research on adverse drug events. I.Muscarinic M3 receptor binding affinity could predict the risk of

Please cite this article as: Montastruc, F., et al., Role of serotonin 5-HA.... European Neuropsychopharmacology (2015), http://dx.doi.org/10.1

antipsychotics to induce type 2 diabetes. Methods Find. Exp.Clin. Pharmacol. 27 (5), 289–304. http://dx.doi.org/10.1358/mf.2005.27.5.908643.

Sim, Yun-Beom, Park, Soo-Hyun, Kang, Yu-Jung, Kim, Seon-Mi, Lee,Jin-Koo, Jung, Jun-Sub, Suh, Hong-Won, 2010. The regulation ofblood glucose level in physical and emotional stress models:possible involvement of adrenergic and glucocorticoid systems.Arch. Pharmacal Res. 33 (10), 1679–1683. http://dx.doi.org/10.1007/s12272-010-1018-3.

Stahl, S.M., Mignon, L., Meyer, J.M., 2009. Which comes first:atypical antipsychotic treatment or cardiometabolic risk? ActaPsychiatr. Scand. 119 (3), 171–179. http://dx.doi.org/10.1111/j.1600-0447.2008.01334.x.

Stahl, Stephen M., 2013. Stahl's Essential Psychopharmacology, 4thedition Cambridge University Press.

Star, Kristina, Norén, G. Niklas, Nordin, Karin, Ralph Edwards, I., 2011.Suspected adverse drug reactions reported for children worldwide:an exploratory study using VigiBase. Drug Saf. 34 (5), 415–428. http://dx.doi.org/10.2165/11587540-000000000-00000.

Starrenburg, F.C., Bogers, J.P., 2009. How can antipsychotics causediabetes mellitus? insights based on receptor-binding profiles,humoral factors and transporter proteins. Eur. Psychiatry: J.Assoc. Eur. Psychiatr. 24 (3), 164–170. http://dx.doi.org/10.1016/j.eurpsy.2009.01.001.

Strom, Brian L., Stephen, E. Kimmel, Sean, Hennessy, 2012.Pharmacoepidemiology, 5th edition Wiley-Blackwell, Chichester,West Sussex, UK.

« The Uppsala Monitoring Centre – VigiBaseTM, 2015. Consulté lefévrier 16. ⟨http://www.umc-products.com/DynPage.aspx?id=73590&mn1=1107&mn2=1132⟩.

Verdoux, H., Tournier, M., Bégaud, B., 2010. Antipsychotic pre-scribing trends: a review of pharmaco-epidemiological studies.Acta Psychiatr. Scand. 121 (1), 4–10. http://dx.doi.org/10.1111/j.1600-0447.2009.01425.x.

Wade, Jennifer M., Juneja, Punita, Adrienne, W. MacKay, James,Graham, Peter, J. Havel, Laurence, H. Tecott, Evan, H. Gould-ing, 2008. Synergistic impairment of glucose homeostasis in Ob/ob mice lacking functional serotonin 2C receptors. Endocrinol-ogy 149 (3), 955–961. http://dx.doi.org/10.1210/en.2007-0927.

« WHO | WHO Regional Offices, 2015. WHO. Available from: ⟨http://www.who.int/about/regions/en/⟩. (accessed 2015).

Yamada, Yasuhiko, Ohno, Yoshiyuki, Nakashima, Yoshifumi, Fukuda,Masato, Takayanagi, Risa, Sato, Hitoshi, Tsuchiya, Fumito,Sawada, Yasufumi, Tatsuji, Iga, 2002. Prediction and assessmentof extrapyramidal side effects induced by risperidone based ondopamine D(2) receptor occupancy. Synapse 46 (1), 32–37. http://dx.doi.org/10.1002/syn.10111.

Zhou, Ligang, Sutton, Gregory M., Justin, J. Rochford, Robert, K.Semple, Daniel, D. Lam, Laura, J. Oksanen, Zoe, D. Thornton-Jones, et al., 2007. Serotonin 2C Receptor agonists improve type2 diabetes via melanocortin-4 receptor signaling pathways. CellMetab. 6 (5), 398–405, http://dx.doi.org/10.1016/j.cmet.2007.10.008.

T2C and histamine H1 receptors in antipsychotic-induced diabetes:016/j.euroneuro.2015.07.010