International Journal of Neuropsychopharmacology (2004), 7 (Supplement 1), S21–S25.

Copyright f 2004 CINP

S U P P LE M E N T
DOI : 10.1017/S1461145704004134

How do we choose between atypical

antipsychotics? The advantages of amisulpride

Ann M. Mortimer
Department of Psychiatry, The University of Hull, East Riding Campus, Beverley Road, Willerby, Hull HU10 6NS, UK

Abstract
Clinician choice of an atypical antipsychotic may depend on a number of factors such as perceived
efficacy, tolerability and cost. It is also important that the choice of treatment takes into consideration
the previous response to treatment, experience of side-effects and personal clinical characteristics. The
receptor-affinity profiles of the atypical antipsychotics differ ; with the exception of amisulpride, a selective
D2/D3 antagonist, all the atypical antipsychotics exhibit a greater affinity for the serotonin-2A receptors
than dopamine receptors. However, there is no evidence that the variation in receptor affinities is relevant
to efficacy. Indeed, the crucial factor may be fast dissociation from/low affinity for the D2 receptor.
Tolerability also varies between the atypical antipsychotics and the side-effect profile may be related to

Downloaded from http://ijnp.oxfordjournals.org/ by guest on June 7, 2016

the receptor-affinity profile of the individual drugs. Extrapyramidal side-effects are generally less of a
problem with most atypical drugs than with conventional drugs, but weight gain, loss of glycaemic
control, sedation and hyperprolactinaemia remain problematic in some patients. Amisulpride is effective
for the treatment of both positive and negative symptoms, and is well tolerated with regard to weight gain,
glucose tolerance and sedation. In two clinical trials, the AMIRIS and SOLIANOL studies, amisulpride dem-
onstrated clear advantages over some other atypical antipsychotics with respect to negative symptoms,
depressive symptoms and weight gain.
Received 27 July 2003 ; Reviewed 17 September 2003 ; Revised 25 October 2003 ; Accepted 29 October 2003

Key words : Amisulpride, atypical antipsychotics, schizophrenia.

Introduction thirdly, those that have affinity for a broad range of

central receptors, such as clozapine, olanzapine, zote-
In the UK, the National Institute of Clinical Excellence
pine and quetiapine. With the exception of amisul-
(NICE) recommends the use of atypical antipsychotics
pride, all the atypical antipsychotics exhibit a greater
for all newly diagnosed sufferers of schizophrenia and
affinity for 5-HT-2A receptors than dopamine re-
also for sufferers who are experiencing troublesome
ceptors and this was once thought to explain their
side-effects on the older drugs (NICE, 2002). However,
efficacy for treating negative symptoms of schizo-
with the availability of numerous atypical agents, how
phrenia and to confer a level of protection against ex-
do clinicians choose which atypical agent to prescribe ?
trapyramidal symptoms (EPS). However, amisulpride
Atypical antipsychotics differ in their receptor-
is an efficacious atypical antipsychotic for negative
affinity profiles and can be divided into three groups :
symptoms of schizophrenia (Boyer et al., 1995 ; Danion
firstly, amisulpride, a selective dopamine type-2/
et al., 1999 ; Loo et al., 1997 ; Paillere-Martinot et al.,
type-3 (D2/D3) receptor antagonist, stands alone in
1995). Together with the recent proposal that the
this group ; secondly, atypical antipsychotics that have
most powerful predictor of atypicality is fast dis-
affinity mainly for dopamine and serotonin (5-HT)-2A
sociation from the low dopamine D2 receptor (Kapur
receptors, such as risperidone and ziprasidone ; and
and Seeman, 2001), affinities for other receptors
would now appear to be of limited relevance in the
Address for correspondence : Professor A. Mortimer, Department
therapeutic action of antipsychotic drugs, although
of Psychiatry, The University of Hull, East Riding Campus,
Beverley Road, Willerby, Hull HU10 6NS, UK.
they may still relate to the side-effects profiles of
Tel. : +44 1482 466 960 Fax : +44 1482 464 569 individual atypical antipsychotics. The clinical ad-
E-mail : A.M.Mortimer@hull.ac.uk vantages and disadvantages of a number of atypical
S22 A. Mortimer
antipsychotics, therefore, are discussed here. In ad-
dition, interim findings from trials comparing the
efficacy and side-effect profiles of amisulpride with
risperidone and olanzapine are described in this
paper.
Advantages and disadvantages of atypical
antipsychotics
Clozapine
Clozapine is the only atypical antipsychotic indicated
for treatment-resistant schizophrenic patients and in
patients who have severe, untreatable neurological
adverse reactions to other antipsychotic agents. Treat-
ment resistance is defined as a lack of satisfactory
clinical improvement despite the use of adequate
doses of at least two different antipsychotic agents,
including an atypical antipsychotic agent, prescribed
for adequate duration. Hence, clozapine is a third-line
treatment.
Although serum levels can be monitored to maxi-
mize patients ’ responses, clozapine is probably the
worst tolerated of all the atypical antipsychotics.
Approximately 1 in 43 people develop neutropenia
(Munro et al., 1999) : the drug has to be stopped in
these patients. White cell counts need to be measured
weekly for 18 wk, then fortnightly and then monthly
after the first year of treatment, making the use of
clozapine cumbersome. Treatment with clozapine is
also associated with some disturbing side-effects such
as weight gain, hyper-salivation, sedation, tachycardia
and a loss of glycaemic control that can amount to
diabetes in treatment-resistant patients. Acquisition
costs plus monitoring costs can result in clozapine
therapy being expensive.
Risperidone
Risperidone has been shown to be more effective and
well tolerated than conventional antipsychotics for the
treatment of both positive and negative symptoms of
schizophrenia. Moreover, risperidone is advantageous
in that the illness can be managed with low dosages
(f6 mg/d) and consequently, EPS are equivalent to
those seen with placebo. However, adverse effects can
include orthostatic hypotension, sexual dysfunction,
sedation, minor weight gain and hyperprolactinaemia,
which cause a variety of related clinical phenomena
including menstrual cycle disruption. These side-
effects can partly be resolved by using lower doses of
risperidone, but they may still create a clinical man-
agement difficulty.
Olanzapine
Like risperidone, olanzapine is more effective than
conventional antipsychotics for treating positive and
negative symptoms of schizophrenia. EPS are uncom-
mon at any doses, and while there is no hyperpro-
lactinaemia, mild intransient rises occur that appear
to have no clinical relevance. However, olanzapine
therapy can produce substantial weight gain, which
may become worse as time passes. Recent recognition
that olanzapine can compromise glycaemic control has
caused some authorities to suggest that glucose levels
be monitored in patients who begin treatment with
olanzapine and clozapine. However, as it is likely that
other antipsychotics may have this effect, it would
seem prudent to monitor glycaemic control routinely
in any patient prescribed any antipsychotic until the
picture becomes clearer. Sedation and dizziness can
also be annoying side-effects of olanzapine. Unlike
Downloaded from http://ijnp.oxfordjournals.org/ by guest on June 7, 2016
risperidone, where doses are recommended to be
lower and lower, the recommended dosage of olan-
zapine, 10 mg, can be insufficient. In England, the
average dose of olanzapine is much higher at 17 mg,
which can make it expensive.
Quetiapine
Quetiapine is as effective as conventional antipsy-
chotics for treating positive symptoms (Davis J.M.,
Chen N., Glick I.D., 2003), and seems to be well toler-
ated as EPS and hyperprolactinaemia are absent and
weight gain is minimal. Thus, quetiapine treatment is
a good choice for drug-naive, first-episode patients
who are especially sensitive to extrapyramidal side-
effects. However, quetiapine may be no better for
negative symptoms than conventional antipsychotics.
Considering this drug produces virtually no EPS, and
that Parkinsonism may be mistaken for negative
symptoms (Mortimer A. and Spence S., 2001), this is a
paradox that nobody seems able to explain. Further-
more, quetiapine treatment can be expensive in some
countries because, like olanzapine, quetiapine is often
required at higher doses than anticipated.
Ziprasidone
Ziprasidone, which is only available in some Euro-
pean countries and North America, is as effective as
conventional antipsychotics for treating positive symp-
toms (Davis et al., 2003). While ziprasidone is in the
same receptor-affinity group as risperidone, unlike the
latter, EPS occur less frequently (Davis and Markham,
1997 ; Tarsy et al., 2002) and effects on prolactin levels
are minimal. However, ziprasidone can cause, albeit
 How do we choose between atypical antipsychotics ? S23

uncommonly, somnolence and gastrointestinal side- p = 0.036 p = 0.02 p = 0.042

effects. A question mark also exists over whether %
80

Responders 6 months
Maintenance >day 56
ziprasidone is any better than conventional antipsy- 70 77
60 72
chotics for relieving negative symptoms, and current 65
58
65
50
concerns exist regarding cardiac side-effects. 52
40
30
20
Amisulpride 10
0
Amisulpride is effective in treating both positive and PANSS* BPRS* CGI-2**
negative symptoms of schizophrenia, and is, in fact,
Figure 1. Efficacy of amisulpride vs. risperidone after 6
the most thoroughly evaluated atypical antipsychotic
months of treatment. Responders were defined as :
for the treatment of negative symptoms. At low doses
(*) patients who had greater than 50 % improvement from
(f300 mg), amisulpride produces placebo-level EPS. baseline in their symptoms or (**) patients who were
The safety profile of amisulpride is favourable with considered much or very much improved on CGI-I score.
respect to weight gain, glucose tolerance and sedation, %, Amisulpride (n=121) ; &, risperidone (n=123).
and amisulpride does not compromise cognitive (Adapted from Sechter et al., 2002.)
functions in healthy volunteers, unlike conventional
treatment (Peretti et al., 1997). Disadvantages of
p = 0.033
amisulpride include anxiety, agitation, insomnia and

Downloaded from http://ijnp.oxfordjournals.org/ by guest on June 7, 2016

49% ns
hyperprolactinaemia. p ⭐ 0.10
38%
35% 33%
31%
Direct comparative trials with amisulpride
23%
The efficacy and safety of amisulpride was directly
compared with risperidone and olanzapine in two
separate, randomized, controlled trials – the AMIRIS SOFAS responders SOFAS responders SOFAS responders
study and the SOLIANOL study. (30%) (40%) (50%)

Figure 2. Response rates after 6 months for the SOFAS

The AMIRIS study
In the AMIRIS study, amisulpride was compared with
risperidone in a 6-month, randomized, controlled trial
of 304 patients with chronic schizophrenia (Sechter
et al., 2002). No significant differences were evident
between the demographic and disease characteristics
of the two groups. The patients, who had all been ill
for approx. 10 yr, were in their late 30s and had the
usual distribution of subtypes of schizophrenia. Drop-
out rates were as expected and no significant differ-
ences occurred between the two treatment arms.
The percentage of patients that responded to treat-
ment after 6 months was significantly greater with
amisulpride (200–800 mg) than risperidone (2–8 mg)
on PANSS (Positive and Negative Syndrome Scale ;
65 % vs. 52 %, p=0.036), BPRS (Brief Psychiatric Rating
Scale ; 72 % vs. 58 %, p=0.02) and CGI-2 (Clinical
Global Impression-2 ; 77 % vs. 65 %, p=0.042 ; Figure 1).
Responders were classified as those patients who had
greater than 50 % improvement in their symptoms.
The superiority of amisulpride therapy was also
reflected in the functional response as measured on
the Social and Occupational Functioning Assessment
Scales (SOFAS). For example, data for patients with
SOFAS improvements of 30 % showed that amisulpride
scores. %, Amisulpride (n=121) ; &, risperidone (n=123).
(Adapted from Sechter et al., 2002.)
was significantly greater than risperidone (49 % vs.
35 %, p=0.033 ; Figure 2). As the SOFAS criteria
became more stringent, from 40 to 50 % increase in
function, statistical significance was lost but the trend
was still present (Figure 2). The patients themselves
also rated amisulpride therapy superior to risperidone
therapy using the Van Putten Questionnaire : 93 %
of amisulpride-treated patients expressed a positive
subjective response to treatment after 6 months com-
pared to 83 % in the risperidone group (p=0.015 ;
Figure 3).
Both amisulpride and risperidone were well toler-
ated, with few EPS being experienced by patients in
either group. In addition, far fewer endocrine dis-
orders and effects on sexual function were observed
with amisulpride than with risperidone (0.7 % vs.
5.7 %). Of 152 amisulpride-treated patients, none re-
ported impotence, ejaculation failure, non-puerperal
lactation or breast pain (Figure 4). However, analy-
sis for a significant between-group difference was
not possible because of small numbers. Moreover,
amisulpride was clearly superior to risperidone with
S24 A. Mortimer

p = 0.015 40%
Increase ⭓7%

(baseline to end-point)
33%
30% *p < 0.05

Completers
93%
83%
80% 20% 18%
70%
12%
10%
6%

0%
2 months 6 months

2 months 6 months Figure 5. Gain in body weight after 2 and 6 months of

treatment. Responders were defined as patients gaining
Figure 3. Patients’ subjective response to treatment using
weight o7 % of that of baseline. %, Amisulpride (n=100) ;
the Van Putten Questionnaire. %, Amisulpride (n=121) ; &,
&, risperidone (n=96). (Adapted from Sechter et al., 2002.)
risperidone (n=123). (Adapted from Sechter et al., 2002.)

0.7 An interim analysis at day 56 demonstrated that

Global
5.7 amisulpride was at least as effective as olanzapine at
1.4 improving psychotic symptoms ; the total BPRS score
Amenorrhoea (F)
1.4
fell by 17.6 (S.D.=13.9) points in the amisulpride group
0
Ejaculation failure (M)
3.4 and by 16.3 (S.D.=13.4) points in the olanzapine group.

Downloaded from http://ijnp.oxfordjournals.org/ by guest on June 7, 2016

0 The two-sided 95 % confidence limits for the difference
Impotence (M)
3.4
Breast pain female (F)
0 confirming the hypothesis of non-inferiority of ami-
1.4
0
Lactation non-puerperal (F)
1.4
Figure 4. Proportion of patients experiencing endocrine
disorders and sexual dysfunction. %, Amisulpride
(n=152) ; &, risperidone (n=158). (Adapted from Sechter
et al., 2002.)
respect to weight gain, as only 18 % of amisulpride-
treated patients increased their weight by more than
7 % after 6 months of treatment compared with 33 % of
risperidone-treated patients (Figure 5).
The SOLIANOL study
The SOLIANOL study (Martin et al., 2002) was the first
randomized trial comparing the efficacy and toler-
ability of amisulpride with olanzapine. A total of 377
patients with predominantly positive symptoma-
tology were treated for 6 months with either ami-
sulpride (200–800 mg/d) or olanzapine (5–20 mg/d).
Randomization resulted in the demographic data for
both groups being very similar : patients were in their
late 30s, approximately two-thirds of the cohort was
male, and approximately half the patients had a BMI
index within the normal range. Baseline BPRS scores
of more than 50 showed that these patients were quite
ill, although both groups had similar distributions of
clinical symptoms. These patients were also depressed
as indicated by a baseline MADRS (Montgomery–
Asberg Depression Rating Scale) score of greater
than 16.
between the two treatment groups were (4.02x1.54)
sulpride treatment with respect to olanzapine treat-
ment. BPRS scores, as well as depressive symptoms,
improved to a similar extent with both treatments,
although some scores suggested a possible trend in
favour of amisulpride.
Positive and negative scores were also similar for
amisulpride and olanzapine and few EPS emerged
with either drug. As expected, weight gain was lower
in amisulpride-treated patients : by day 56, amisul-
pride recipients had gained 0.4 kg whereas olanzapine
recipients had gained 2.7 kg. This difference was
highly significant (p<0.0001). In addition, while hos-
pitalization status for both groups was identical at the
beginning of the trial, more patients were hospitalized
at day 56 if they were taking olanzapine rather than
amisulpride (10 % vs. 5 %). This difference was not
statistically significant.
Conclusion
Clinician choice of an atypical antipsychotic may
depend on several factors such as perceived efficacy,
tolerability and, in some countries such as the UK,
cost. These factors can vary considerably with each
atypical antipsychotic. Furthermore, treatment must
be chosen for the individual patient with particular
regard to their previous responses, experience of side-
effects, and personal clinical characteristics. These
issues are of more significance than cost alone.
Amisulpride has been shown to be effective and
well tolerated for the treatment of both positive

and negative symptoms of schizophrenia. Moreover,
amisulpride demonstrates clear advantages over some
other atypical antipsychotics with respect to negative
symptoms, depressive symptoms and weight gain.
Acknowledgements
Professor Mortimer has received unrestricted edu-
cational and research grants from Sanofi-Synthelabo,
Novartis, Astra Zeneca, Eli Lilly and Janssen-Cilag.
Statement of Interest
Professor Mortimer received an honorarium from
Sanofi-Synthelabo for presentation at the Symposium.
No honorarium was provided for the writing of this
paper.
References
Boyer P, Lecrubier Y, Puech AJ, Dewailly J, Aubin F (1995).
Treatment of negative symptoms in schizophrenia with
amisulpride. British Journal of Psychiatry 166, 68–72.
Danion JM, Rein W, Fleurot O (1999). Improvement of
schizophrenic patients with primary negative symptoms
treated with amisulpride. Amisulpride Study Group.
American Journal of Psychiatry 156, 610–616.
Davis R, Markham A (1997). Ziprasidone. CNS Drugs 8,
153–159.
Davis JM, Chen N, Glick ID (2003). A meta-analysis of the
efficacy of second-generation antipsychotics. Archives of
General Psychiatry 60(6), 553–564.
Kapur S, Seeman P (2001). Does fast dissociation from the
dopamine D(2) receptor explain the action of atypical
How do we choose between atypical antipsychotics ? S25
antipsychotics ? : a new hypothesis. American Journal of
Psychiatry 158, 360–369.
Loo H, Poirier-Littre MF, Theron M, Rein W, Fleurot O (1997).
Amisulpride versus placebo in the medium-term
treatment of the negative symptoms of schizophrenia.
British Journal of Psychiatry 170, 18–22.
Martin S, Ljo H, Peuskens J, Thirumalai S, Giudicelli A,
Fleurot O, Rein W, SOLIANOL Study Group (2002).
A double-blind, randomised comparative trial of
amisulpride versus olanzapine in the treatment of
schizophrenia : short-term results at two months.
Current Medical Research and Opinion 18, 355–362.
Munro J, O ’Sullivan D, Andrews C, Arana A, Mortimer A,
Kerwin R (1999). Active monitoring of 12,760 clozapine
recipients in the UK and Ireland. Beyond pharma-
covigilance. British Journal of Psychiatry 175, 576–580.
National Institute for Clinical Excellence (NICE) (2002).
Guidance on the use of newer (atypical) antipsychotic
drugs for the treatment of schizophrenia. Technology
Appraisal Guidance – No. 43.
Downloaded from http://ijnp.oxfordjournals.org/ by guest on June 7, 2016
Paillere-Martinot ML, Lecrubier Y, Martinot JL, Aubin F
(1995). Improvement of some schizophrenic deficit
symptoms with low doses of amisulpride. American Journal
of Psychiatry 152, 130–134.
Peretti CS, Danion JM, Kauffmann-Muller F, Grange D, Patat
A, Rosenzweig P (1997). Effects of haloperidol and
amisulpride on motor and cognitive skill learning in
healthy volunteers. Psychopharmacology (Berlin) 131,
329–338.
Sechter D, Peuskens J, Fleurot O, Rein W, Lecrubier Y,
Amisulpride Study Group (2002). Amisulpride vs.
risperidone in chronic schizophrenia : results of a 6-month
double-blind study. Neuropsychopharmacology 27,
1071–1081.
Tarsy D, Baldessarini RJ, Tarazi FI (2002). Effects of newer
antipsychotics on extrapyramidal function. CNS Drugs 16,
23–45.