Add-on mirtazapine enhances antipsychotic effect of first generation antipsychotics in schizophrenia: A double-blind, randomized, placebo-controlled trial

Add-on mirtazapine enhances antipsychotic effect of first generation antipsychotics in schizophrenia: A double-blind, randomized, placebo-controlled trial
of 8
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
  This article appeared in a journal published by Elsevier. The attachedcopy is furnished to the author for internal non-commercial researchand education use, including for instruction at the authors institutionand sharing with colleagues.Other uses, including reproduction and distribution, or selling orlicensing copies, or posting to personal, institutional or third partywebsites are prohibited.In most cases authors are permitted to post their version of thearticle (e.g. in Word or Tex form) to their personal website orinstitutional repository. Authors requiring further informationregarding Elsevier’s archiving and manuscript policies areencouraged to visit:  Author's personal copy Add-on mirtazapine enhances antipsychotic effect of   fi rst generationantipsychotics in schizophrenia: A double-blind, randomized, placebo-controlled trial Grigori Joffe a,b, ⁎ ,1 , Viatcheslav Terevnikov b,c,1 , Marina Joffe a,1 , Jan-Henry Stenberg a,1 ,Mark Burkin c,1 , Jari Tiihonen d,e,1 a Helsinki Univercity Central Hospital (HUCH), Helsinki, Finland b Kellokoski Hospital, Kellokoski, Finland c University of Petrozavodsk, Russia d Department of Forensic Psychiatry, University of Kuopio, Finland e Department of Mental Health and Alcohol Research, National Public Health Institute, Helsinki, Finland a r t i c l e i n f o a b s t r a c t  Article history: Received 6 October 2008Received in revised form 3 December 2008Accepted 4 December 2008Available online 13 January 2009 Background:  Mirtazapine,anantidepressantwithabroadspectrumofreceptoraf  fi nitymay,if combined with  fi rst generation antipsyhotics (FGAs), improve clinical pro fi le of the FGAs.However, potentiation of the antipsychotic effect by mirtazapine has not been reported thusfar. We explored the ef  fi cacy of adjunctive mirtazapine on symptoms of schizophrenia inpatients having an insuf  fi cient response to different FGAs. Methods:  Schizophrenia-diagnosed patients with a prolonged disease and a history of a poorresponse to numerous antipsychotics, who were at least moderately ill despite their FGAstreatment, received add-on mirtazapine ( n =20) or placebo ( n =19) in a 6-week double-blindrandomized controlled trial (RCT). The analysis was made on a Modi fi ed Intent-to-Treat (MITT)basis with Last Observations Carried Forward (LOCF). Results:  Mirtazapine outrangedplacebo on almost allmeasures.The clear-cutclinical relevanceofthis fi ndingwasdemonstratedbyalargeeffectsizeof1.00(95%CI0.23 – 1.67,  p =0.003)onthetotal Positive and Negative Syndrome Scale (PANSS) scores (the primary outcome). The PANSSpositive subscale scores decreased by 17.2% with mirtazapine vs.1.6% with placebo (  p b 0.001),and the PANSS negative subscale scores by 12% and 3% (  p b 0.001), correspondingly. Conclusions:  This is the  fi rst RCT reporting a robust additive antipsychotic effect of anadjunctive antidepressant. Mirtazapine-FGAs combination appears to be a safe, well-toleratedand ef  fi cacious treatment option in this challenging population. These  fi ndings are importantdue to the current re-emerging attention to FGAs. The focus of further studies should beexpanded to include combinations with or switching to novel antipsychotics, differentsubpopulations of patients with schizophrenia,  fi nding of optimal doses, and comparison withclozapine. Trial registration code: ISRCTN00721331, © 2008 Elsevier B.V. All rights reserved. Keywords: MirtazapineSchizophreniaFirst generation antipsychoticsAntidepressantRCTCombination 1. Introduction All antipsychotics are antagonists at dopamine D2 recep-tors (Butcher, 2000). Clozapine, the prototypical second Schizophrenia Research 108 (2009) 245 – 251 ⁎  Corresponding author. Department of Psychiatry, Helsinki UniversityCentral Hospital, Välskärinkatu 12, FI-00290 Helsinki, Finland. Tel.: +358 947163700, +358 40 5136500 (mobile). E-mail address:  grigori.joffe@hus. fi  (G. Joffe). 1 Location of work: Psychiatric Hospital, Matrosy, Pria ž a District, Republicof Karelia, Russia, and Day Treatment Unit, Psychoneurological Dispensary,Krsano fl otskaya Street 29, Petrozavodsk, Republic of Karelia, Russia.0920-9964/$  –  see front matter © 2008 Elsevier B.V. All rights reserved.doi:10.1016/j.schres.2008.12.002 Contents lists available at ScienceDirect Schizophrenia Research  journal homepage:  Author's personal copy generation antipsychotic (SGA), demonstrates higher ef  fi cacyin treatment-refractory schizophrenia and lower incidenceof extrapyramidal symptoms (EPS) than FGAs  —  featuresreferred to as  “ atypicality ” . These advantages of clozapineare attributed to its high af  fi nity to serotonin 5HT2 receptorsand low af  fi nity to dopamine D2 receptors, as opposed to fi rst generation antipsychotics (FGAs), whose D2 af  fi nity ishigh (Meltzer, 1994). Based on this concept of serotonindopamine antagonism (SDA), a number of novel SGAs havebeen developed to replace clozapine, due to its seriousadverse effects.Much optimism accompanied initial reports that favoredSGAs over FGAs in a vast majority of important clinicaldomains. However, the superiority of SGAs has recently beenquestioned since in large independent studies SGAs failed tooutrange FGAs in effectiveness or safety ( Jones et al., 2006;Tiihonenetal.,2006).Nevertheless,evidenceexistsinfavorof SGA in terms of e.g., negative symptoms of schizophrenia(Davis et al., 2003), chronic schizophrenia with a history of suboptimal response to treatment (Volavka et al., 2002),adverse effect pro fi le (Dossenbach et al., 2004), compliance(Kahn et al., 2008) and (with olanzapine) discontinuation of treatment (Lieberman et al., 2005).Like the SGAs, antidepressants trazodone, mianserine,nefazodone and mirtazapine are also inhibitors at 5HT2receptors. They are, however, lacking D2 receptor af  fi nityand, thus, antipsychotic ef  fi cacy. If the SDA theory was true, acombination of these compounds with modestly dosed FGAscould result in a SGA-like receptor blockade pro fi le andimproved clinical properties. Indeed, the addition of trazo-done, mianserin (Hayashi et al., 1997), or nefazodone ( Joffe et al., 1999) to FGAs may reduce negative symptoms of schizophrenia. Moreover, nefazodone (Wynchank and Berk,2003) and mianserin (Poyurovsky et al., 1999) may alleviate the FGAs-induced EPS, possibly due to the preponderance of serotonin 5HT2 antagonism over dopamine D2 blockadeachieved by this combination.Mirtazapine, in addition to the 5HT2 receptor blockade,demonstrates inhibition at adrenergic  α 2 , serotoninergic5HT3 and histaminergic H1 receptors, as well as indirectagonism at serotonin 5HT1a receptors (Berendsen andBroekkamp,1997). Also these receptors appear to be involvedin the schizophrenic process (Hertel et al.,1999; Stahl, 2008).Inthisrespect,mirtazapineresemblesclozapine(Stahl,2008),and thus as adjunct might have advantages over antidepres-santsthatexhibitpurely5HT2inhibition.Indeed,inpreclinicalstudies, adjunctive mirtazapine enhanced the antipsychotic-like effects and tolerability of FGAs drugs in a fashionpredictive for atypicality (Berendsen et al., 1998; Pinderet al.,1998). In two recent small RCTs adjunctive mirtazapinewas superior to placebo in negative, but not positivesymptoms in haloperidol (Berk et al., 2001) and clozapine-treated schizophrenia (Zoccali et al., 2001). Mirtazapine alsoimproved both antipsychotic-induced akatisia (Hieber et al.,2008) and neuro-cognitive performance (Delle Chiaie et al.,2007).Thisstudywas designedto explore theeffectofadjunctivemirtazapine on symptoms of schizophrenia in patients whohad negligible or sub-optimal response to different FGAs instable doses. The study was initiated by the investigators andindependent of any commercial interests. 2. Materials and methods The studyprotocol and its amendments were approved bythe Ethics Committee of the Karelian Republic, RussianFederationandconductedincompliancewiththeDeclarationof Helsinki, Guideline for Good Clinical Practice (ICH_GCP),and the current National Regulations.Thisstudywasadouble-blind,RCTofmirtazapineaddedtoon-going FGA treatment. The inclusion criteria were: in-/outpatients, aged 18 to 65 years, diagnosed with schizophre-nia or schizoaffective disorder (depressive type) (DSM-IVTR;APA, 2000), presence of positive or negative symptoms orboth, resulting in the illness of at least moderate severity ( ≥ 4on the Clinical Global Impression (CGI) severity scale) (Guy,1976); and the clinical condition which had remainedunchanged during  ≥ 6 previous weeks. The patients had toreceive  ≥ 1 FGAs at a daily dose of   ≥ 400 mg chlorpromazineequivalents, which had remained unchanged for  ≥ 6 previousweeks (8 weeks for depot FGAs). Soon after start of recruit-ment (during the run-in period of the  fi rst patient), the dosewas amended to  ≥ 200 mg due to the study sites' currentmodest dosage practices and the SDA-theoretical premise of suf  fi ciencyoflowD2receptorblockadeinthepresenceofhigh5HT2 receptor blockade.Exclusion criteria were: a previous lack of response to anadd-on antidepressant with af  fi nity to 5HT2, current SGA, ahistory of non-response to any SGA, a serious medicalcondition, (in order to avoid a possible exacerbation of theillness provoked bythe antidepressant) a history of bipolarorschizoaffective disorder, bipolar type; substance misuse,expected poor compliance, suicidality, treatment with anyantidepressant, mood stabilizer, buspiron or triptan, anyantipsychotics other than those currently in use withinprevious  ≤ 6 weeks, regular use of benzodiazepines (unlessnecessary)oraccidentaluseofbenzodiazepinesindailydosesof   ≥ 30 mg of diazepam equivalents, and (for fertile females)pregnancy, lactation, or inability/unwillingness to usecontraception.After a complete description of the study, subjects gavewritten informed consent, and after a 1-week placebo run-inperiod, all those eligible were randomly assigned, half tomirtazapine and half to placebo, in a double-blind fashion for6 weeks, with their maintained FGA dosages. Either mirtaza-pine 30 mg or placebo was administered QD in identicalgelatin capsules. Blocked randomization was performed witha randomization table. Thick envelopes with randomizationcodes were open only when the database was closed.Adherence tothe studymedicationwas monitored bycapsulecount during each visit.Clinical ef  fi cacy was assessed with the PANSS (Kay et al.,1987), Clinical Global Impression (CGI), severity (CGIs) andCGI, improvement (CGIi) scales. Patients' subjective percep-tions in change of their condition were measured with thePatient Global Impression (PGI; Guy,1976). Neurological sideeffects were evaluated with the Simpson-Angus Scale forExtrapyramidal Side-effects (SAS; Simpson and Angus,1970).Clinical assessments were performed at 1 week before thestudy ( − 1, screening), week 0 (baseline) and after 1, 2, 4, and6 weeks (or at drop-out).Theprimaryef  fi cacyvariableswerethePANSStotalscores.Secondary ef  fi cacy parameters included the PANSS subscale 246  G. Joffe et al. / Schizophrenia Research 108 (2009) 245 –  251  Author's personal copy scores, CGI, PGI and number of responders (those with  ≥ 20%decreaseonthetotalPANSSscores).Theanalysiswasmadeona MITT basis (i.e., all randomized patients with  ≥ 1 on-therapyevaluations) with LOCF.Cross-sectional statistical differences between the studygroups were tested using an independent samples  t  -test forcontinuous variables and either Chi-Square or Fisher's exacttests for categorical variables. Within-group changes in theef  fi cacy variables over time were tested by the paired sample t  -test. Repeated measures ANOVA was used to examine andtest the differences between treatments over time. Time wastreatedasawithin-subjectfactor,andtreatmentasabetweensubject factor in the statistical model. Baseline values wereused as covariates. Post hoc type comparisons of thetreatments were based on estimated marginal means whenrepeated measures ANOVA was used. Spearman's correlationcoef  fi cient was applied to study associations between base-line depression and primary outcome. 2-Tailed  p -values lessthan 0.05 were considered statistically signi fi cant. Analyseswere performed using SPSS for Windows 14.0 software. 3. Results Of the 46 subjects screened, 41 were randomized. Oneplacebo patient was excluded due to a protocol violation (i.e.,too low FGA dose). One more placebo patient was withdrawnimmediately after randomization and could not be taken intoanalyses, asthe baselineassessmentswerenot performed.Nowithdrawals occurred in either group afterwards, and theMITTwasthusequaltoconventionalITT.Of14patientswithaprevious history of clozapine treatment, clozapine was usedin the recommended therapeutic doses of   ≥ 300 mg in twopatients. None of the patients had a history of other SGAs.Thirty nine subjects (20 with mirtazapine and 19 withplacebo) formed the MITT population (Fig. 1). One placebopatient had schizoaffective disorder, depressive type, and theremaining patients had schizophrenia. Patients receivedmono- or polytherapy with haloperidol ( n =14), tri fl uoper-azine ( n =13), haloperidol decanoate ( n =10),  fl uphenazinedecanoate ( n =8) levomepromazine ( n =6), chlorprotixene( n =5), zuclopentixol decanoate ( n =5), zuclopentixol ( n =4), fl upentixol ( n =2), periciazine ( n =2), sulpiride ( n =1), orcombination of   ≥ 2 of these FGAs. Baseline demographic andclinical characteristics, medication history or current FGAdosages did not signi fi cantly differ between the mirtazapineand placebo groups, with the exception of higher PANSSpositive scores in the mirtazapine group (Table 1).Incidence of adverse events (10 with mirtazapine and 10with placebo) did not differ. All adverse events wereregistered once only, except headache ( n =2, both withplacebo) and hypersedation ( n =3, all with mirtazapine).None of adverse events led to withdrawal from the study.With the paired  t  -test (baseline and endpoint datacompared), changes in the PANSS total scores for themirtazapine group outranged that of placebo by an averageof 10 points (  p =0.003). Computed with adjustment for thebaseline PANSS differences, the similar result remained(  p =0.006).The effect size was 1.00 (CI95% 0.34 – 1.67). Within groups,mirtazapine treatment resulted in a statistically signi fi cantimprovement of all measured parameters, and placebointervention showed improvement on the PANSS negativesubscale. The between-group difference in the change frombaseline to week 6 favored mirtazapine on all scales, exceptthe PANSS general subscale, SAS, and CGIs (with a similartrendforthelatter).Nocomparisonsfavoredplacebo(Table2).Repeated measures ANOVA performed for all 5 time-points resulted in  p =0.002 with the baseline adjustment forthe baseline PANSS differences (without adjustment,  p =0.001), again in favour of mirtazapine. In post hoc tests,thedifferencebetweentreatmentssurfacedatweek4(Fig.2).Depression (item 20 of the PANSS) did not differ between the  Table 1 Baseline demographic and clinical data of schizophrenia patients withsuboptimal response to FGAsMirtazapine30 mg ( n =20)Placebo( n =19)Test statisticsAge, years 43.40 (9.24) 48.21 (9.68)  t  =1.59,  p =0.121Gender, males:  n ,(%) 11 (55%) 9 (47.4%)  X  2 =0.227,  p =0.632In-patients ( n , %) 11 (55%) 7 (36.8%)  X  2 =1.293,  p =0.256Age at onset, years: 23.45 (6.47) 24.11 (6.69)  t  =0.31,  p =0.758Duration of disease, years19.95 (9.08) 24.95 (9.43)  t  =1.69,  p =0.100Current FGAs dose,CPZ equivalents, mg330.75 (123.53) 316.58 (164.87) t  =0.31,  p =0.762Number of previousantipsychotics8.50 (2.69) 8.11 (3.49)  t  =0.40,  p =0.694PANSS positive 22.15 (4.10) 18.89 (5.13)  t  =2.20,  p =0.034PANSS negative 30.90 (4.68) 29.26 (5.06)  t  =1.05,  p =0.301PANSS gen 53.05 (6.20) 51.42 (8.65)  t  =0.68,  p =0.502PANSS total 106.10 (10.20) 99.58 (16.06)  t  =1.52,  p =0.137CGIs 4.45 (0.61) 4.21 (0.42)  t  =1.43,  p =0.161SAS 12.05 (5.18) 10.00 (4.70)  t  =1.29,  p =0.204Data (other than gender and in-patients) are given as mean (sd).For all parameters, df=37. Fig. 1.  Subject  fl owchart.247 G. Joffe et al. / Schizophrenia Research 108 (2009) 245 –  251  Author's personal copy groups at baseline ( t  =0.473,  p =0.640), nor did baselinedepression correlate with the primary outcome ( r  s =0.242,  p =0.137).Thein-/out-patientstatusdidnotaffectthechangesin clinical parameters (data not shown). There were 4responders (20.0%) in the mirtazapine group and 1 responder(5.3%) in the placebo group (  p =0.342). 4. Discussion In this study, a statistically signi fi cant favorable effect wasfound in a wide range of clinical parameters when mirtaza-pinewasaddedtoFGAs.Toourknowledge,thisisthe fi rstRCTreporting an additive antipsychotic effect of an adjunctiveantidepressant. Robust statistically signi fi cant differencesfavored mirtazapine treatment when compared to placeboin both within group and between group analyses, despitehigher initial psychotic symptom scores in the mirtazapinegroup. The difference was also clinically signi fi cant, which isevident by an impressive effect size of 1.00 (CI95% 0.34 – 1.67)[vs.  ≤ 0.3 small, 0.4 – 0.7 medium, and  ≥ 0.7 large Effect Size(Cohen, 1988)] on the PANSS total scores, which was theprimary outcome measure. This effect size was higher thanthat of SGAs vs. FGAs (0.49, 0.29, 0.25, and 0.21 for clozapine,amisulpride,risperidone, andolanzapine,respectively)(Daviset al., 2003). This, however, has to be taken with caution, assmall trials tend to generate higher effect sizes than largerones (Contopoulos-Ioannidis et al., 2005). Twenty percent of patients on mirtazapine were responders vs. only 5.3% onplacebo. Though not statistically signi fi cant in our smallpopulation, this would correspond to a Number Needed toTreat (NNT) of 7, meaning a clinical signi fi cance comparablewith that of family intervention in schizophrenia (Pharoahet al., 2006) or acetylcholinesterase inhibitors in vasculardementia (Demaerschalk and Wingerchuk, 2007) and greaterthan considered acceptable for common somatic treatments(e.g.,NNT=20 – 30forstatinesorNNT=53 – 60forantibioticsinrheumatic fever) (Robertson et al., 2005; McElduff et al.,2006).Theimprovementachievedwithclozapineinthestudyby Kane et al. (1998) was, however, clinically more signi fi cant(NNT=4).While the effectof mirtazapine on the negative symptomshas been noted earlier (Berk et al., 2001; Zoccali et al., 2001),an improvement in the positive symptoms with adjunctivemirtazapine was a novel  fi nding, and the magnitude of thiseffect was unexpectedly high (17.2% on mirtazapine vs. 1.6%on placebo,  p b 0.001).The antipsychotic effect of current drugs has been at-tributed to D2 receptor blockade, while adjunctive anti-depressants have been used in schizophrenia (withcon fl icting results) for treating negative (Rummel et al.,2005) or affective (Whitehead et al., 2005) symptoms. However, there is a body of preclinical (Hertel et al., 1999)and clinical (Litman et al., 1996) evidence to indicate that α 2 -adrenoreceptor inhibition may enhance the antipsycho-tic effects of D2 blockade.  α 2  receptor blockade alsoenhances neurogenesis (Rizk et al., 2006), which may beof bene fi t in schizophrenia. In a small trial ( n =17), Litmanand co-authors (1996) reported the ef  fi cacy of combiningthe FGA  fl uphenazine with the  α 2  inhibitor idazoxanequivalent to that of clozapine in therapy-resistant patients.Moreover, 5HT1A receptor stimulation could contribute toenhancing the antipsychotic effect (Berendsen et al., 1998).In our earlier report on nefazodone, a 5HT2 blocker that isdevoid of other receptor af  fi nities of mirtazapine, residualpsychotic symptoms were abated ( Joffe et al.,1999), but the fi ndings of that small, open-label pilot should not beoverestimated. Indeed, the role of 5HT2 receptor antagon-ism might rather be seen in negative, mood (Duinkerke Fig. 2.  PANSS Total Scores of FGA-treated schizophrenia patients with add-onmirtazapine ( fi lled circles,  n =20) or placebo (open squares,  n =19). Data areanalyzed on Modi fi ed Intent-to-Treat basis, with Last Observations CarriedForward.PANSStotalscoremeansareadjustedforthePANNSscoresatbaseline.  Table 2 Changes in clinical parameters for FGAs-treated patients with schizophreniaafter 6 weeks of add-on treatment with mirtazapine 30 mg ( n =20) or placebo( n =19)Change withmirtazapine:Mean (sd),change % t  =,  p =Change withplacebo: mean(sd), change % t  =,  p =Change withmirtazapinevs. changewith placebo t  =,  p =/ b PANSS positive 4.05 (3.59), 17.2% 0.47 (2.25), 1.6% 3.58 t  =5.04,  p b 0.001  t  =0.92,  p =0.370  t  =3.71,  p b 0.001PANSS negative 3.60 (2.58), 12.0% 0.89 (1.73), 3.0% 2.71 t  =6.23,  p b 0.001  t  =2.26,  p =0.037  t  =3.82,  p b 0.001PANSS general 5.65 (7.03), 10.2% 1.84 (5.24), 3.5% 3.81 t  =3.60,  p =0.002  t  =1.53,  p =0.143  t  =1.91,  p =0.064PANSS total ⁎  13.30 (11.31), 12.5% 3.21 (8.03), 3.2% 10.09 t  =5.26,  p b 0.001  t  =1.74,  p =0.098  t  =3.20,  p =0.003CGI severity 0.55 (0.51), 12.8% 0.21 (0.54), 4.7% 0.34 t  =4.82,  p b 0.001  t  =1.71,  p =0.104  t  =2.03,  p =0.050CGI improvement 0.80 (0.62), 20% 0.05 (0.52), 1.3% 0.75 t  =5.18,  p b 0.001  t  =0.44,  p =0.67  t  =4.07,  p b 0.001PGI 0.70 (0.57),17.5% 0.16 (0.77), 4.0% 0.54 t  =5.48,  p b 0.001  t  =0.90,  p =0.380  t  =2.52,  p =0.016SAS 2.05 (3.50), 9.8% 0.42 (2.67), 0.2% 1.63 t  =2.62,  p =0.017  t  =0.69,  p =0.501  t  =1.64,  p =0.112Absolute values are given as scores at week 0 (baseline) minus those at week6 (endpoint), i.e. positive values mean improvement.Statistical tests were calculated for the absolute values of parameters (i.e.,data not adjusted for baseline differences).For all parameters,  df  =19 in mirtazapine group,  df  =18 in placebo group, and df  =37 between these groups. ⁎  = primary outcome.248  G. Joffe et al. / Schizophrenia Research 108 (2009) 245 –  251
Similar documents
View more...
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks