1. Introduction

As a chronic psychiatric disorder, schizophrenia affects the
worldwide population with a lifetime prevalence of 1% and
considerable long-term mortality, morbidity, and burden (Esan
et al., 2012; Ro€ssler et al., 2005; Saha et al., 2007; S witaj et al.,
2012). Routine treatment regimens are accompanied by adverse
effects, and a significant portion of patients remain symptomatic

* Corresponding author. Psychiatric Research Center, Roozbeh Psychiatric Hos-
pital, Tehran University of Medical Sciences, South Kargar Street, Tehran 13337, Iran.

Best services for writing your paper according to Trustpilot

Premium Partner
From $18.00 per page
4,8 / 5
Writers Experience
Recommended Service
From $13.90 per page
4,6 / 5
Writers Experience
From $20.00 per page
4,5 / 5
Writers Experience
* All Partners were chosen among 50+ writing services by our Customer Satisfaction Team

E-mail address: [email protected] (S. Akhondzadeh).
1 The first four authors contributed equally in this study.


0022-3956/© 2017 Elsevier Ltd. All rights reserved.

© 2017 Elsevier Ltd. All rights reserved.

despite treatment, especially those with negative symptoms, which
are the major disabling factors in these patients (Akhondzadeh and
Moazen-Zadeh, 2017; Bobes et al., 2009; Buckley and Stahl, 2007;
Schooler et al., 2015).

Numerous lines of evidence indicated sex-specific characteris-
tics of schizophrenia (da Silva and Ravindran, 2015; Melcangi and
Garcia-Segura, 2010). Male patients are prone to higher risk,
lower age of onset, and more severe course of the disorder (Aleman
et al., 2003; McGrath et al., 2008). Compared to males, female pa-
tients are reported to respond better to treatment (Bloch et al.,
2005; Grigoriadis and Seeman, 2002; Riecher-Ro€ssler and H€afner,
2000), have lower remission and hospitalization rates (Desai
et al., 2013), and express more concomitant affective symptoms
such as depression, anxiety and irritability (Choi et al., 2001;


There have been few studies of pregnenolone therapy in schizophrenia and those that exist have been
subject to several critical limitations, thus yielding inconsistent results. We attempted to assess the
therapeutic effect of pregnenolone in a patient sample as homogeneous as possible. In this randomized
double-blind clinical trial, 82 female inpatients with chronic schizophrenia, who had discontinued their
antipsychotic medications for at least one week in case of any oral antipsychotic medication or a month
for any depot antipsychotic medication, received risperidone plus either pregnenolone (50 mg/day) or
placebo for 8 weeks. Inclusion criteria were acute illness with a baseline Positive and Negative Syndrome
Scale (PANSS) negative subscale score of !20. Exclusion criteria were the presence of severe depression
or other concomitant psychiatric disorders. Primary outcome was defined as the difference in the PANSS
total score change from baseline to week 8 in the pregnenolone group compared to the placebo group.
No significant difference was found in the PANSS total score changes between the two arms (mean
difference (CI 95%) 1?4 9.41 ( 20.24 to 1.41); p 1?4 0.087). Significant differences were initially found for
PANSS negative change scores (mean difference (CI 95%) 1?4 2.61 ( 5.03 to 0.19); p 1?4 0.035) and
general psychopathology change scores (mean difference (CI 95%) 1?4 5.93 ( 11.37 to 0.48); p 1?4 0.033).
However, these findings did not survive Bonferroni correction for multiple testing. While this trial may
suggest a potential effect of pregnenolone on schizophrenia symptoms, further studies are warranted.

Goldstein, 2006). Also, a second incidence peak of schizophrenia,
around the age of 40, is reported in female patients (van der Werf
et al., 2014). Further studies have suggested that neurosteroids,
including pregnenolone and its metabolites such as estrogens, play
an important role in pathophysiology of schizophrenia and can be
potential therapeutic targets (Halbreich and Kahn, 2003; Heringa
et al., 2015; Seeman, 2002).

Among neurosteroids, pregnenolone is the single primary pre-
cursor for all others and has provoked less concern about adverse
effects in case of long-term therapy compared to other neuro-
steroids including estrogens, progesterone, and dehydroepian-
drosterone (DHEA) (Scheffers et al., 2015; Shulman and Tibbo,
2005; Wentzensen and Trabert, 2015). Furthermore, pregneno-
lone improved response to stress, mood regulation, and cognitive
performance (Marx et al., 2011; Ritsner, 2010, 2011). Besides the
direct effects of pregnenolone, such as modulation of neurotrans-
mitters and hypothalamic-pituitary-adrenal (HPA) axis function
(Zheng, 2009), further effects of pregnenolone are postulated to be
mediated by its metabolites. For example, allopregnenolone de-
creases apoptosis and inflammation, increases neurogenesis, and
markedly increases gamma-aminobutyric acid A (GABAA) receptor
response. Pregnenolone sulfate, another metabolite, positively
modulates N-methyl-D-aspartate (NMDA) receptors. Improving
NMDA receptor function and GABA regulation have been proposed
as two potential mechanisms through which pregnenolone may
benefit patients with schizophrenia (MacKenzie et al., 2007; Marx
et al., 2011). As other metabolites of pregnenolone, estrogens are
widely studied in schizophrenia with promising effects according
to some trials (Heringa et al., 2015). Interestingly, lower serum
levels of pregnenolone was detected in schizophrenia patients
compared to the controls (Ritsner et al., 2007), and oral adminis-
tration of pregnenolone was associated with increased serum levels
of pregnenolone as well as its metabolites by several folds (Marx
et al., 2009, 2014). Furthermore, in animal models, pregnenolone
serum levels were correlated with pregnenolone levels in the
central nervous system (CNS) (Caruso et al., 2013).

Considering the neural effects of pregnenolone, and its metab-
olites, and the fact that oral administration of pregnenolone
potentially increases CNS levels of these neurosteroids, therapeutic
effects of oral administration of pregnenolone in schizophrenia
patients warranted further investigations. Meanwhile, the few
available clinical trials on the pregnenolone adjunctive treatment
in patients with schizophrenia had some limitations and yielded
inconsistent results (Marx et al., 2009, 2014; Ritsner et al., 2010,
2014). A recent meta-analysis on these few trials found no signifi-
cant difference between effects of pregnenolone adjunctive therapy
and placebo on total, positive, or negative symptoms of schizo-
phrenia measured by Positive and Negative Syndrome Scale
(PANSS) (Heringa et al., 2015); however, the high heterogeneity of
negative symptoms (Heringa et al., 2015) and inconsistent results of
previous trials necessitated further scrutiny.

We assessed the potential therapeutic effects of pregnenolone
as an adjunctive therapy to risperidone in women with schizo-
phrenia in an 8-week, randomized, double-blind, placebo-
controlled clinical trial with a sample size of enough power for
hypothesis testing. The primary outcome of interest was defined as
the mean difference in change of PANSS total symptoms score from
baseline to the end-point between the two treatment arms.

2. Methods

2.1. Trial design

In this parallel-group, placebo-controlled, double-blind clinical
trial, schizophrenia patients were equally randomized (1:1),

treated, and followed for 8 weeks. The trial protocol was approved
by the Institutional Review Board (IRB) of Tehran University of
Medical Sciences (Approval number: 27746) in accordance with the
World Medical Association (Declaration of Helsinki, as revised in
Brazil 2013) code of ethics, and registered at the Iranian Registry of
Clinical Trials (IRCT201504211556N76; http://www.irct.ir/). Writ-
ten informed consent was obtained from patients and their legally
authorized representatives in accordance with the procedures
defined by the local IRB, and participants were informed that they
were free to withdraw from the study at any time without any
negative effect on their relationship with the healthcare provider.

2.2. Participants

Participants were selected from female inpatients with chronic
schizophrenia who were 18e50 years old. The diagnosis was based
on the Structured Clinical Interview for Diagnostic and Statistical
Manual-IV-TR Axis I Disorders (SCID) (First et al., 1997), and
confirmed through an interview by a senior physician as well as a
chart review. It was required that the participants be in the active
phase of the illness, have a PANSS total score of !60, PANSS
negative subscale score of !20, and illness duration of !2 years.
Patients with any of the following conditions were excluded: his-
tory of comorbid DSM-IV Axis I disorders or alcohol/substance
(other than nicotine) dependence based on DSM-IV-TR, a score of
!14 on 17-item Hamilton Rating Scale for Depression (HRSD) or !4
on depression item of PANSS, a score of !2 on the suicide item of
HRSD or suicide ideation, mental retardation (intelligence quotient
<70) based on clinical judgment, inability to communicate, severe extrapyramidal symptoms, serious neurological or medical condi- tions, pregnancy, lactation, women in child-bearing ages without a reliable method of contraception, and any history of prior throm- boembolism, abnormal uterine bleeding, uterine or breast cancer, cerebrovascular accident, endocrine abnormality, hormone replacement therapy, or hypersensitivity to risperidone or preg- nenolone. Patients were also excluded in case of any oral antipsy- chotic medication during the last week, any depot antipsychotic medication during the last month, or electroconvulsive therapy during the last 2 weeks prior to their enrollment. Patients' prior medications were not changed or discontinued in this study, but rather the participants were selected from those who had dis- continued their medication due to some other reason including lack of supportive care or incompliance with their treatment. It was also required that participants do not use mood stabilizers, anti- depressants, sedating antihistamines, or a second antipsychotic, or receive behavior intervention therapy during the course of the trial. 2.3. Study settings This multicenter clinical trial was conducted at Roozbeh Hos- pital (Tehran University of Medical Sciences, Tehran, Iran), Razi Hospital (University of Social Welfare and Rehabilitation Sciences, Tehran, Iran), and Qods Hospital (Kurdistan University of Medical Sciences, Sanandaj, Iran) from May 2015 to November 2016. After the baseline visit, patients were evaluated in four consecutive visits with 2-week intervals. The trial was not subject to ethnical or regional restrictions as the patients were referred from different regions of Iran to these three large academic referral hospitals. 2.4. Interventions All the patients received risperidone (Risperdal®, Janssen Phar- maceuticals) in addition to the pregnenolone (Ray Kurzweil & Terry Grossman's Health Products) or placebo from the beginning of the study. The starting dose for risperidone was 2 mg/day and was L. Kashani et al. / Journal of Psychiatric Research 94 (2017) 70e77 71 72 L. Kashani et al. / Journal of Psychiatric Research 94 (2017) 70e77 subsequently increased with 2 mg increment on a weekly basis to a maximum dose of 6 mg/day (2 mg TID) based on the clinical response. Pregnenolone was administered at a constant dose of 50 mg/day from baseline to the endpoint. 2.5. Outcomes In this trial, the PANSS was used for the assessment of treatment efficacy at baseline, and weeks 2, 4, 6, and 8. As a 30-item rating scale, the PANSS consists of three validated subscales for mea- surement of negative (seven items), positive (seven items), and general psychopathological (16 items) symptoms of schizophrenia which are summed up in the PANSS total score (Kay et al., 1987). The PANSS was previously used in the Iranian population for clin- ical trials of schizophrenia (Iranpour et al., 2016; Nikbakhat et al., 2016; Tajik-Esmaeeli et al., 2017). Patients were rated on the PANSS scale by four trained raters who were third year residents of psychiatry and experienced in PANSS implementation in schizo- phrenia trials with an inter-rater reliability of >90% on PANSS total
score. The 17-item HRSD (Hamilton, 1960; Jafarinia et al., 2016) was
used for assessment of depressive symptoms at baseline and

The primary outcome of interest was defined as the mean dif-
ference in the PANSS total score change from baseline to week 8
between the pregnenolone and placebo arms. The secondary out-
comes were the mean difference in the PANSS subscale score
change between the two arms from baseline to the end-point, and

entities. An investigational drug pharmacist was responsible to
dispense the study medications in identical containers. Pregneno-
lone and placebo tablets were identical in size, shape, texture, color,
taste, and odor. The physicians, nurses, patients, and the statistician
were all blind to the treatment allocation.

2.9. Statistical methods

Data were analyzed and related plots were drawn using IBM
SPSS Statistic 19.0.0 (IBM Corporations) and SigmaPlot 12.2.0
(SYSTAT Software, Incorporated) respectively. Continuous and cat-
egorical variables were reported in the respective formats of mean
(±standard deviation, SD) and count (%).Chi-square test or Fisher’s
exact test were used where appropriate. Mean differences in
PANSS, HRSD, and ESRS scores between the two arms were re-
ported in the format of mean (95% confidence intervals (95%CI)).
Mean score changes from baseline to each time point were
compared between the two groups using the independent sample’s
t-test. Degrees of freedom and concordant p-values were corrected
whenever the assumption of equality of variances was violated
based on the Levene’s test results. Effect size was estimated for
between-group differences in mean score change from baseline to
each time-point using Cohen’s d (95% CI). Time treatment in-
teractions were assessed using two-way repeated measure analysis
of variance (ANOVA) in which the treatment groups and PANSS
measurements were considered as the between-subject and
within-subject factors respectively. Greenhouse-Geisser’s correc-
tion was applied whenever the assumption of sphericity was
violated based on the Mauchly’s test results. In order to adjust for
multiple testing effect, p-values of <0.012 were considered signif- icant in analysis of PANSS total and subscale score changes. P-value <0.05 was considered significant in analysis of baseline character- istics, time treatment interactions, and adverse events. Two- tailed p-values were reported. 3. Results 3.1. Participants As shown in the flow diagram (Fig. 1), out of 122 screened pa- tients, 90 were randomized to the pregnenolone and the placebo arms. Out of 45 patients in each group, 41 met the first post- baseline visit, continued to the end of the study, and were considered for analysis. Analysis of baseline characteristics of patients did not show any significant difference between the two treatment arms for socio- demographics, illness duration, schizophrenia subtype, prior anti- psychotic medications, baseline HRSD score, or baseline ESRS score (Table 1). Also, there was no significant difference in baseline PANSS total or subscale scores between the two arms. The mean dose of risperidone administered during this study was 4.35 (0.65) mg/day and 4.40 (0.70) mg/day in pregnenolone and placebo groups, respectively. 3.2. Outcomes 3.2.1. PANSS total score As the primary outcome of interest, there was no significant difference in mean total score changes between the pregnenolone and the placebo arms by the end of the study (Table 2, Fig. 2). During the study, there was no significant time treatment interaction between the two arms Two-way ANOVA with Greenhouse-Geisser correction: F (df, mean square) 1?4 2.53 (1.17, 922.98), p 1?4 0.110. 


I'm Niki!

Would you like to get a custom essay? How about receiving a customized one?

Check it out