Lower weight gain with the orally disintegrating olanzapine

than with standard tablets in rst-episode never treated

psychotic patients
B. Arranz
1
, L. San
1
*
, R.M. Duenas
2
, M. Centeno
3
, N. Ramirez
1
, J. Salavert
1
and E. del Moral
1
1
Hospital San Rafael, Barcelona, Spain
2
Benito Menni, CASM, Barcelona, Spain
3
Hospital de Granollers, Benito Menni, Barcelona, Spain
Objective Apost-hoc analysis of the data froma randomised clinical trial involving prescription of antipsychotic treatment
to never treated rst-onset psychotic patients was used to compare the weight change after 6-week olanzapine treatment
(standard tablets vs. orally disintegrating formulation).
Method In the subgroup of 38 patients randomised to olanzapine, standard olanzapine tablets were non-randomly and
consecutively prescribed to the rst 19 patients, with the orally disintegrating formulation being prescribed to the following
19 patients.
Results After 6-week treatment with olanzapine, a signicant higher increase in weight was noted in those patients on
standard tablets (mean weight increase 6.3 1.9 Kg) as compared to those on orally disintegrating olanzapine (mean weight
increase 3.3 3.2 Kg) (F7.7; p 0.009). BMI increase was also signicantly higher in the olanzapine tablet group (mean
increase of 2.1 Kg/m
2
as compared with 1.1 Kg/m
2
in the orally disintegrating group) (F4.7; p 0.036). Substantial weight
gain (SWG) (7%increase frombaseline weight) was noted in 84.2%(n 16) of the olanzapine tablet patients and in 31.6%
(n 6) of the orally disintegrating olanzapine patients, with the olanzapine tablet group showing a signicant increase in the
mean percentage of weight gain (F4.0; p 0.014).
Conclusions Partial sublingual absorption occurring with orally disintegrating olanzapine may bypass gastrointestinal
metabolisation and hence lead to differences in metabolite versus parent compound ratios. However, the need arises to
replicate the present study with a longer follow-up. Copyright # 2007 John Wiley & Sons, Ltd.
key words orally disintegrating olanzapine; standard olanzapine tablets; substantial weight gain; rst-episode psychosis
INTRODUCTION
Excess weight is undoubtedly associated with health
risks such as type 2 diabetes, metabolic syndrome and
cardiovascular mortality, and is also a major cause of
non-adherence with antipsychotic medication in
psychiatric patients. Several reviews including pro-
spective, retrospective, randomised, open, non-
comparative and controlled studies have extensively
documented the relationship between conventional
and second generation antipsychotic treatment and
both weight gain (Chue and Cheung, 2004; Haddad,
2005) and type 2 diabetes (Jin et al., 2004; Bergman
and Ader, 2005; Rettenbacher, 2005; Gianfrancesco
and Nasrallah, 2006). Hormones, neuropeptides and
neurotransmitters have been consistently linked to this
antipsychotic-induced weight gain, being the poly-
morphisms of the serotonin 5-HT2A-2C (Templeman
et al., 2005), histamine H1 and leptin receptors (Tighe
and Dinan, 2005) those showing most promising
results.
Methodological problems arising from the patient
sample included (medication-na ve vs. medication-
free), study design (prospective vs. retrospective,
clinical trial vs. naturalistic, short vs. long-term) and
other factors (baseline and outcome measures) have
human psychopharmacology
Hum. Psychopharmacol Clin Exp 2007; 22: 1115.
Published online in Wiley InterScience
(www.interscience.wiley.com) DOI: 10.1002/hup.819
* Correspondence to: L. San, Department of Psychiatry, Hospital
San Rafael, Passeig Vall dHebron, 107-117, 08035 Barcelona,
Spain. E-mail: 12636LSM@COMB.ES
Copyright # 2007 John Wiley & Sons, Ltd.
Received 3 July 2006
Accepted 15 November 2006
led to conicting results regarding the weight gain
attributed to each antipsychotic. However, it seems
that although almost all antipsychotics induce weight
gain to some extent, markedly differences between
pharmacological groups are evident (Allison et al.,
1999).
So far, results from one study (de Haan et al., 2004)
have suggested that switching from standard olanza-
pine tablets to the orally disintegrating olanzapine
causes a decrease in appetite and weight loss.
Therefore, data from a randomised clinical trial
involving prescription of antipsychotic treatment to
antipsychotic-na ve rst-onset psychotic patients was
used to compare the change in weight between both
olanzapine formulations.
METHODS
Patients
The present study corresponds to a data base from an
ongoing research project on rst-episode never
treated psychotic patients admitted to an acute
psychiatric unit during the years 20042005 (study
protocol 02-0463, code F1D-XB-0171), in which
the primary objective was to compare the efcacy
and effectiveness of different antipsychotics during
1 year follow-up. A total of 274 patients were initially
screened to enter the study. Forty-ve patients could
not be randomised because of unmet inclusion criteria
(n 18) or refusal to sign informed consent (n 27),
while the other 229 patients were randomised to
open-label antipsychotic treatment with haloperidol,
olanzapine, risperidone, quetiapine or ziprasidone, at
usual clinical doses.
Given the previous evidence that treatment with the
orally disintegrating olanzapine may cause weight
loss, as compared with standard olanzapine tablets (de
Haan et al., 2004), a post-hoc analysis was performed
in the subgroup of 38 patients randomised to
olanzapine (27 men, 11 female). Standard olanzapine
tablets were prescribed to the rst 19 patients (15 men,
4 female), while the orally disintegrating formulation
was administered to the following 19 patients (12 men,
7 female). This non-randomised chronological assig-
nation was due to the orally disintegrating olanzapine
not being available in our country for the rst
19 patients. DSM-IV-TR diagnosis (conrmed after
1 year follow-up) was Paranoid Schizophrenia
(n 17; 44.7%), Schizophreniform disorder (n 8;
21.1%), Schizoaffective disorder (n 4; 10.5%),
Acute Psychoses (n 4; 10.5%), Psychotic disorder
not otherwise specied (n 3; 7.9%), and Bipolar
Disorder (n 2; 5.3%).
All patients were hospitalised during the 6-week
study length so as to ensure medication compliance,
with dietary and exercise inpatient conditions remain-
ing constant throughout the study. None of the
38 patients dropped-out before study end-point. The
6-week study duration was selected because it is
the usual time frame in clinical practice to determine
treatment outcome and decide on treatment continu-
ation and/or polytherapy. During the study period,
olanzapine was the only treatment prescribed, thus
excluding the effect of other medications (antidepress-
ants, mood stabilizers) on weight gain. Patients
weights were obtained from the same instrument in
fasting conditions (between 7.30 and 8.00 am) both at
baseline and at study end-point (6 weeks). Substantial
weight gain (SWG) was dened as gaining at least 7%
of initial body weight (Lipkovich et al., 2006). All
patients gave written informed consent to the study
procedures. This study was approved by the local
ethical committee.
Statistical analyses
Data were analysed using the Statistical Package for
the Social Sciences (SPSS, Chicago, IL) version 11.0.
Student t test unpaired, 2-tailed, was used to compare
baseline characteristics (age, weight, body mass index
(BMI) and PANSS scores) from both olanzapine
groups. The General Linear Model (repeated
measures) procedure was used to perform analysis
of covariance (ANCOVA) to control for the effect of
age and sex (covariates) on the within-subject variable
(basal and 6-week weight, BMI and PANSS scores),
with type of olanzapine (tablets or orally disintegrat-
ing) as the between-subjects factor. The General
Linear Model (univariate) test was used to perform
analysis of covariance (ANCOVA) to control for the
effect of age and sex (covariates) on the mean SWG
values (dependent variable), with type of olanzapine
(tablets or orally disintegrating) as the xed-effect
factor. The overall signicance level was set at
p 0.05.
RESULTS
As shown in Table 1, both olanzapine groups were not
different in age, gender distribution, BMI, or weight at
baseline (before starting antipsychotic treatment with
olanzapine) DSM-IV-TR diagnosis as conrmed after
1 year follow-up (x
2
8; p 0.15) and baseline total
PANSS scores did not signicantly differ between
Copyright # 2007 John Wiley & Sons, Ltd. Hum. Psychopharmacol Clin Exp 2007; 22: 1115.
DOI: 10.1002/hup
12 b. arranz ET AL.
both treatment groups (Table 1). Mean olanzapine
dose was not signicantly different in both treatment
groups (13.8 6 mg in the olanzapine tablets and
15.8 8 mg in the orally disintegrating olanzapine
group; t 0.88; p 0.38).
After 6-week treatment with olanzapine, a signi-
cant decrease in total PANSS scores was noted in both
treatment groups (mean decrease of 33.4 20 in the
olanzapine tablet group as compared to mean
decrease of 36.6 19 in the orally disintegrating
olanzapine group). However, neither the type of
olanzapine administered (xed effect factor) nor
age or sex (covariates) showed a signicant effect
upon the observed PANSS score decrease (F5.8;
p 0.02; Table 2).
A signicantly higher increase in weight was noted
in those patients on standard tablets (mean weight
increase 6.3 1.9 Kg) as compared to those on orally
disintegrating olanzapine (mean weight increase
3.3 3.2 Kg) (Table 2). BMI increase was also
signicantly higher in the olanzapine tablet group
(mean increase of 2.1 Kg/m
2
as compared with 1.1 Kg/
m
2
in the orally disintegrating group). Age and sex did
not show any signicant inuence in either weight or
BMI increase (Table 2).
SWG (7% increase from baseline weight) was
noted in 84.2% (n 16) of the olanzapine tablet
patients and in 31.6% (n 6) of the orally disin-
tegrating olanzapine patients, with the olanzapine
tablet group showing a signicant increase in the mean
percentage of weight gain (F4.0; p 0.014;
Table 2).
DISCUSSION
Our results indicate that the orally disintegrating
olanzapine may have a lower liability of inducing
weight gain than standard tablets, as higher weight
gain was noted in our patients on standard olanzapine
tablets (6.3 Kg, 9.7% increase in weight) compared to
those taking the orally disintegrating olanzapine
(3.3 Kg, 5.6% increase in weight SWG). SWG seems
to be a good measure of clinically signicant weight
gain (Lipkovich et al., 2006).
The risk for antipsychotic-induced weight gain has
been widely documented (Gentile, 2006). There is
only one study comparing standard and orally
disintegrating olanzapine (de Haan et al., 2004). In
their study, a mean weight reduction of 6.6 Kg was
noted in the nine patients switched during 16 weeks
from standard olanzapine to the orally disintegrating
formulation, while the nine patients continuing
treatment with conventional olanzapine tablets
showed a weight increase of 3.7 Kg.
Causes for these discrepant results may lay in
differences in their study design and in the patient
sample included. de Haans study had a non-
randomised open-label design and was performed in
18 recent-onset, but not antipsychotic-na ve, schizo-
phrenic patients which were already on olanzapine. As
no information was given about the length of
Table 1. Characteristics of both olanzapine groups at study entry
Standard
tablets
Orally
disintegrating
Age (years) 25.8 6.8 23.1 7.2 t 1.17; p 0.24
Sex 15M/4F 12M/7F x
2
1.15; p 0.28
Weight (Kg) 67.6 12.2 61.2 8.8 t 1.86; p 0.07
BMI (Kg/m
2
) 22.3 3.3 21.2 2.6 t 1.19; p 0.24
PANSS score 83.8 17 88.4 20 t 0.92; p 0.36
t Students t test, unpaired, 2-tailed.
Table 2. Baseline and 6-week weight, BMI, PANSS score and Percentage of weight gain
Olanzapine tablets Orally disintegrating
Corrected model Basal 6-week Basal 6-week
Weight (Kg) 67.6 12.2 74.0 12.1 61.2 8.8 64.5 9.0 F7.7; p 0.009
a
BMI (Kg/m
2
) 22.3 3.3 24.4 3.3 21.2 2.6 22.3 2.5 F4.7; p 0.036
a
PANSS score 83.8 17.3 50.3 16 88.4 20 51.8 14 F5.8; p 0.02
a
Weight gain (%) 9.7 3.7 5.6 5.4 F4.0; p 0.014
b
a
General Linear Model, repeated measures. Within-subject variables: Basal and 6-week weight, BMI and PANSS scores. Between-subjects
factor: tablets or orally disintegrating olanzapine. Covariates: age and sex. For weight analyses; age effect: F1.0; p 0.31. Sex
effect: F0.22; p 0.64. Treatment (olanzapine) effect: F13.1; p 0.001. For BMI analyses; age effect: F0.6; p 0.44. Sex effect:
F1.6; p 0.20. Treatment (olanzapine) effect: F13.2; p 0.001. For PANSS score analyses; age effect: F0.02; p 0.88. Sex
effect: F0.29; p 0.59. Treatment (olanzapine) effect: F0.36; p 0.55.
b
General Linear Model, univariate. Fixed-effect factor: tablets or orally disintegrating olanzapine. Dependent variable: mean SWG values.
Covariates: age and sex. For weight gain analyses; age effect: F1.3; p 0.24. Sex effect: F2.8; p 0.09. Treatment (olanzapine) effect:
F10.2; p 0.003.
Copyright # 2007 John Wiley & Sons, Ltd. Hum. Psychopharmacol Clin Exp 2007; 22: 1115.
DOI: 10.1002/hup
olanzapine-induced weight gain 13
antipsychotic treatment with olanzapine, it could be
speculated that their patients had already reached a
stabilisation in weight gain at study entry. In this
respect, it has been well documented that olanzapine
induced weight gain occurs mainly during the rst
40 weeks (10 months) of treatment and then reaches a
plateau (Kinon et al., 2001). The baseline weight
difference of 19.8 Kg and baseline BMI difference of
3.9 Kg/m
2
noted between de Haans and the present
study may conrmthis hypothesis and also account for
the discrepant results. Our sample consisting of never
treated rst-onset psychotic patients avoids the
cumulative effect of prior antipsychotic treatment
possibly being a confounder factor in the antipsycho-
tic weight gain liability (Arranz et al., 2004).
One of the strengths of the present study lies in the
randomisation to antipsychotic treatment, albeit
assignation to type of olanzapine formulation fol-
lowed chronological criteria. This study design ruled
out the possibility of clinicians having prescribed
olanzapine to those patients with lower baseline
weight. Lowbaseline BMI (BMI <25 kg/m
2
) seems to
be a predictor of antipsychotic-induced weight gain
(Basson et al., 2001; Lambert et al., 2005). In our
study, no signicant differences were noted in baseline
BMI values between both olanzapine groups. Only six
patients showed BMI values lower than 18.5 kg/m
2
,
and they were equally distributed between both
olanzapine groups (three patients in each group), thus
avoiding the possible bias of the reported weight gain
being attributed to the effect of low BMI in one of our
olanzapine groups.
Antipsychotic-induced weight gain has also been
related to clinical outcome, with higher weight gain
noted in those patients with better clinical outcome
(Basson et al., 2001). However, this association seems
to be present only in the early weeks of treatment, and
accounts for 4% of the variance in clinical improve-
ment (Zipursky et al., 2005). Our reported lower
liability of the orally disintegrating olanzapine for
inducing weight gain can not be attributed to a worse
clinical improvement, as a similar total PANSS score
decrease was noted in both olanzapine groups.
The underlying biochemical mechanisms behind
olanzapine-induced weight gain are unclear, and so
are those accounting for the reported difference in
weight gain liability between standard and orally
disintegrating olanzapine. The T allele of the
polymorphism (759C/T) of the promoter region of
the 5-HT2C receptor gene seems to protect against
weight gain in a subgroup of patients on olanzapine
treatment (Templeman et al., 2005). Furthermore,
different pharmacokinetic properties between the
orally disintegrating and the standard olanzapine
formulation have been reported, with the orally
disintegrating formulation yielding a more rapid
absorption and earlier detectable plasma concen-
trations (Bergstrom et al., 2004). However, no
differences in AUC, C
max
and T
max
seem to exist
between both formulations (Markowitz et al., 2006).
Partial sublingual absorption occurring with orally
disintegrating olanzapine may bypass gastrointestinal
metabolization and the P-glycoprotein transporter, and
hence lead to differences in metabolite versus parent
compound ratios. In this respect, two major olanza-
pine metabolites have been described so far, the
10-N-glucoronide and 4-N-desmethylolanzapine,
which are devoid of activity at the brain receptors
(Callaghan et al., 1999). However, little is known
about their antagonistic activity for the peripheral
gastrointestinal 5-HT2 and 5-HT3 receptors mediating
feeding and satiety. It could then be speculated that the
reported differences in SWG between the orally
disintegrating and the standard olanzapine tablets
might be attributed to their distinct peripheral
pharmacodynamic prole.
Use of a patient sample of never-treated rst-
episode psychosis, randomisation to olanzapine
treatment and patients being on monotherapy through-
out the study give strength to the reported results.
However, the, main limitations concern the short
treatment duration, the relatively small sample size,
the lack of blind raters, and the non-randomisation to
olanzapine formulation. Therefore, the need arises to
replicate the present results in a study specically
designed to overcome these limitations, and possibly
including measurement of olanzapine metabolites and
their receptor afnities, so as to conrm the clinical
advantages of orally disintegrating olanzapine on
SWG.
CONFLICT OF INTEREST
L. San has received grant/research support, received
honoraria and participated in speakers/advisory
boards from AstraZeneca, Eli Lilly, Pzer, Janssen
and Wyeth. B. Arranz has received grant/research
support from AstraZeneca, Eli Lilly, Pzer, and
Janssen, received honoraria from Eli Lilly and Janssen
and participated in speakers/advisory boards from
Janssen. This manuscript has not been published or
submitted for publication elsewhere. All authors hold
themselves jointly and individually responsible for its
content, and were the only persons involved in the
study design, patient recruitment and follow-up,
Copyright # 2007 John Wiley & Sons, Ltd. Hum. Psychopharmacol Clin Exp 2007; 22: 1115.
DOI: 10.1002/hup
14 b. arranz ET AL.
statistical analyses and discussion of the reported
results.
ACKNOWLEDGEMENTS
This study was performed with the support of a grant
from Fundacio La Marato de TV3 (01/5330) and Lilly
Pharmaceuticals, Spain (F1D-XB-0171).
REFERENCES
Arranz B, Rosel P, Ramirez N, et al. 2004. Insulin resistance and
increased leptin concentrations in noncompliant schizophrenia
patients but not in antipsychotic-naive rst-episode schizophrenia
patients. J Clin Psychiatry 64: 13351342.
Allison DB, Mentore JL, Heo M, et al. 1999. Antipsychotic-induced
weight gain: a comprehensive reasearch synthesis. Am J Psychia-
try 156: 16861696.
Basson BR, Kinon BJ, Taylor CC, Szymanski KA, Gilmore JA,
Tollefson GD. 2001. Factors inuencing acute weight gain in
patients with schizophrenia treated with olanzapine, haloperidol
or risperidone. J Clin Psychiatry 62: 231238.
Bergman RN, Ader M. 2005. Atypical antipsychotics and glucose
homeostasis. J Clin Psychiatry 66: 504514.
Bergstrom RF, Mitchell M, Witcher J, et al. 2004. Rapid onset
of absorption with olanzapine orally did integrating tablets.
Schizophr Res 67: 160.
Callaghan JT, Bergstrom RF, Ptak LR, Beasley CM. 1999. Olanza-
pine pharmacokinetic and pharmacodynamic prole. Clin Phar-
macokinet 37: 177193.
Chue P, Cheung R. 2004. The impact of weight gain associated with
atypical antipsychotic use in schizophrenia. Acta Neuropsychia-
trica 16: 113123.
de Haan L, van Amerlsvoort T, Rosien K, Linszen D. 2004. Weight
loss after switching fromconventional olanzapine tablets to orally
disintegrating olanzapine tablets. Psychopharmacology 175:
389390.
Gentile S. 2006. Long-term treatment with atypical antipsychotics
and the risk of weight gain. Drug Saf 29: 303319.
Gianfrancesco F, Nasrallah HA. 2006. The inuence of study design
on the results of pharmacoepidemiologic studies of diabetes risk
with antipsychotic treatment. Ann Clin Psychiatry 18(1): 917.
Haddad P. 2005. Weight change with atypical antipsychotics in the
treatment of schizophrenia. J Psychopharmacol 19(6): 1627.
Jin H, Meyer JM, Jeste DV. 2004. Atypical antipsychotics and
glucose dysregulation: a systematic review. Schizophr Res 71:
195212.
Kinon BJ, Basson BR, Gilmore JA, Tollefson GD. 2001. Long-term
olanzapine treatment: weight change and weight-related health
factors in schizophrenia. J Clin Psychiatry 62(2): 92100.
Lambert M, Haro JM, Novick D, et al. 2005. Olanzapine vs other
antipsychotics in actual outpatient settings: six months tolerabil-
ity results from the European Schizophrenia Out-patient Health
Outcomes study. Acta Psychiatr Scand 111: 232243.
Lipkovich I, Citrome L, Perlis R, et al. 2006. Early predictors of
substantial weight gain in bipolar patients treated with olanza-
pine. J Clin Psychopharmacol 26: 316320.
Markowitz JS, DeVane CL, Malcolm RJ, et al. 2006. Pharmaco-
kinetics of olanzapine after single-dose oral administration of
standard tablet versus normal and sublingual administration of an
orally disintegrating tablet in normal volunteers. J Clin Pharma-
col 46: 164171.
Rettenbacher MA. 2005. Disturbances of glucose and lipid meta-
bolismduring treatment with newgeneration antipsychotics. Curr
Opin Psychiatry 18: 175179.
Templeman LA, Reynolds GP, Arranz B, San L. 2005. Polymor-
phisms of the 5-HT2C receptor and leptin genes are associated
with antipsychotic drug-induced weight gain in caucasian
subjects with a rst-episode psychosis. Pharmacogenetics
Genomics 15: 195200.
Tighe S, Dinan T. 2005. An overviewof the central control of weight
regulation and the effect of antipsychotic medication.
J Psychopharmacol 19(6): 3646.
Zipursky RB, Gu H, Green AI, et al. 2005. Course and predictors of
weight gain in people with rst-episode psychosis treated with
olanzapine or haloperidol. Br J Psychiatry 187: 537543.
Copyright # 2007 John Wiley & Sons, Ltd. Hum. Psychopharmacol Clin Exp 2007; 22: 1115.
DOI: 10.1002/hup
olanzapine-induced weight gain 15