Celecoxib as adjunctive therapy in schizophrenia: A double-blind, randomized and placebo-controlled trial

Schizophrenia Research 90 (2007) 179 – 185 www.elsevier.com/locate/schres

Celecoxib as adjunctive therapy in schizophrenia: A double-blind, randomized and placebo-controlled trial Shahin Akhondzadeh a,⁎, Maryam Tabatabaee a , Homayoun Amini a , Seyed Ali Ahmadi Abhari a , Seyed Hesamedin Abbasi b , Behnaz Behnam c a

Psychiatric Research Center, Roozbeh Psychiatric Hospital, Tehran University of Medical Sciences, South Kargar Street, Tehran 13337, Iran b Research Unit, Tehran Heart Center, Tehran University of Medical Sciences, Tehran, Iran c Department of Psychiatry, Semnan University of Medical Sciences, Semnan, Iran Received 31 August 2006; received in revised form 20 October 2006; accepted 25 November 2006 Available online 8 January 2007

Abstract Some evidence suggests that the pathophysiology of schizophrenia is associated with the abnormal immune system, and cytokines may be important in schizophrenia. Cyclooxygenase-2 (COX-2) inhibitors such as celecoxib reduce the production of proinflammatory cytokines including Th1-like cytokines. Indeed, COX-2 inhibitors rebalance type-1 and type-2 immune response. The purpose of the present investigation was to assess the efficacy of celecoxib as an adjuvant agent in the treatment of chronic schizophrenia in an eight-week, double-blind and placebo-controlled trial. Eligible participants in this study were 60 patients with chronic schizophrenia. All patients were inpatients and were in the active phase of the illness, and met DSM-IV criteria for schizophrenia. Patients were allocated in a random fashion, 30 to risperidone 6 mg/day plus celecoxib 400 mg/day (200 mg bid) (morning and evening) and 30 to risperidone 6 mg/day plus placebo. Although both protocols significantly decreased the score of the positive, negative and general psychopathological symptoms over the trial period, the combination of risperidone and celecoxib showed a significant superiority over risperidone alone in the treatment of positive symptoms, general psychopathology symptoms as well as PANSS total scores. The means Extrapyramidal Symptoms Rating Scale for the placebo group were higher than in the celecoxib group over the trial. However, the differences were not significant. The results of this study suggest that celecoxib may be an effective adjuvant agent in the management of patients with chronic schizophrenia and anti-inflammatory therapies should be further investigated. © 2006 Elsevier B.V. All rights reserved. Keywords: Celecoxib; COX-2 inhibitor; Chronic schizophrenia; Immune system

1. Introduction Although the etiology of schizophrenia is not well known, there is overall consensus that schizophrenia belongs to the complex disorders such as hypertension ⁎ Corresponding author. Tel.: +98 21 88281866; fax: +98 21 55419113. E-mail address: [email protected] (S. Akhondzadeh). 0920-9964/$ - see front matter © 2006 Elsevier B.V. All rights reserved. doi:10.1016/j.schres.2006.11.016

(Jablensky et al., 1991; Anderson, 2000; Akhondzadeh, 2006). Many hypotheses have been made to explain the pathophysiology of schizophrenia like dopamine, 5-HT, glutamate, and purinergic hypothesis of schizophrenia (Akhondzadeh, 1998, 2001; Lara and Souza, 2000; Mohammadi and Akhondzadeh, 2001). Evidence has accumulated that inflammatory processes in the brain contribute to the aetiopathogenesis of psychiatric

180

S. Akhondzadeh et al. / Schizophrenia Research 90 (2007) 179–185

disorders including schizophrenia (Körschenhausen et al., 1996; Müller and Ackenheil, 1998). Therefore, it is of interest to assess the role of anti-inflammatory agents in the treatment of schizophrenia (Müller et al., 2002, 2004; Riedel et al., 2005). The findings of a few recent studies leading to the hypothesis that the potential use of cyclooxygenase-2 (COX-2) inhibitors as an adjunctive treatment in the treatment of schizophrenia (Müller et al., 2002, 2004; Riedel et al., 2005). COX-2 inhibitors reduce the production of proinflammatory cytokines such as Th1-like cytokines (Nogawa et al., 1997; Riedel et al., 2005). Indeed, COX-2 inhibitors rebalance type-1 and type-2 immune response. Moreover, in schizophrenia a Th-1 to Th-2 shift was reported (Sirota et al., 1995; Müller and Ackenheil, 1998; Müller et al., 1999). COX-2 inhibitors prevent kainic acid induced neuronal death. Therefore, the therapeutic effect of COX-2 inhibitors is mediated by glutamatergic process as well (Nogawa et al., 1997; Baik et al., 1999; Yermakova and O'Banion, 2000). There are several lines of evidence that suggest an overactivation of NMDA receptors is involved in pathogenesis of schizophrenia (Akhondzadeh, 1998, 2006). All together, the use of COX-2 inhibitors such as celecoxib as adjunctive therapy in patients with schizophrenia is supported by the finding of recent studies (Müller et al., 2002, 2004; Riedel et al., 2005). Indeed, the second report of Müller et al. (2004) indicated that the beneficial effect of celecoxib is only in patients with recent onset of schizophrenia (Müller et al., 2004; Riedel et al., 2005). In addition, Rapaport et al. (2005) presented that celecoxib augmentation of continuously ill outpatient subjects with schizophrenia did not improve clinical symptoms or measure of disability. Nevertheless, Rapaport et al. (2005) chose patients that they felt had not achieved maximal benefit from prescribed medications (Rapaport et al., 2005). Together, the results so far are controversial and warrant further investigations (Rapaport et al., 2005; Riedel et al., 2005). The purpose of the present investigation was to assess the efficacy of celecoxib as an adjuvant agent in the treatment of chronic schizophrenia in an eight-week, double-blind and placebocontrolled trial. Compared to previous studies, the present investigation is relatively larger in terms of sample size, the follow-up of patients is longer and all patients are inpatients. Celecoxib is an NSAID that is believed to selectively inhibit prostaglandin synthesis, primarily via COX-2 (Riedel et al., 2005). It does not inhibit COX-1. Its metabolism is primarily mediated via the cytochrome P450 2C9 isoenzyme, potentially causing significant interaction with lithium.

2. Methods 2.1. Trial design This investigation was a prospective, eight-week, double-blind study of parallel groups of patients with chronic schizophrenia and was undertaken in Roozbeh Psychiatric Hospital in Tehran, Iran, from October 2005 to August of 2006. 2.2. Participants Eligible participants in the study were 60 patients with chronic schizophrenia (25 women and 35 men) with age ranging from 19 to 44 years. All participants were inpatients, in the active phase of illness, and met DSM-IV criteria for schizophrenia. All patients were free of any infection and autoimmune disease for at least two weeks, had normal complete blood counts, hepatic and renal panels and negative toxicological screens. The minimum score of 60 on Positive and Negative Syndrome Scale (PANSS) was required for entry into the study (Kay et al., 1987). The patients did not receive neuroleptics from a week prior to entering the trial or depot neuroleptic at least two months before the study. Patients were excluded from the study if they had a clinically significant organic and neurological disorder, current abuse or dependence on drugs within 6 months, serious psychotic disorders other than schizophrenia, peptic ulcer disease or a history of gastrointestinal bleeding, use of any medications identified as contradicted with COX-2 inhibitors. Pregnant or lactating women and those of reproductive age without adequate contraception were also excluded. The trial was performed in accordance with the Declaration of Helsinki and subsequent revisions (World Medical Association. Declaration of Helsinki, 2000) and approved by ethics committee at Tehran University of Medical Sciences. Written informed consents were obtained before entering into the study. 2.3. Intervention Patients were randomly allocated, 30 to risperidone 6 mg/day plus celecoxib 400 mg/day (200 mg bid) (morning and evening) and 30 to risperidone 6 mg/day plus placebo for an eight week, double-blind, placebocontrolled study. Starting dosage of risperidone was 2 mg/day and was increased to 6 mg/day with 2 mg increments in daily dosage for the first two days. Patients in the placebo group received two identical capsules (morning and evening). During the washout

S. Akhondzadeh et al. / Schizophrenia Research 90 (2007) 179–185

period, the patients received benzodiazepines if necessary. Lorazepam was the drug of choice. Five patients dropped out over the trial (two from celecoxib group and three from placebo group). Patients also received biperiden if they had faced extrapyramidal symptoms. Patients were assessed by a psychiatrist at baseline and after 2, 4, 6 and 8 weeks after the medication started. 2.4. Outcome The principal measure of the outcome was the PANSS. The rater used standardized instructions in the use of PANSS. The mean decrease in PANSS score from baseline was used as the main outcome measure of response of schizophrenia to treatment. The extrapyramidal symptoms were assessed using the Extrapyramidal Symptoms Rating Scale (ESRS) (part one: parkinsonism, dystonia, dyskinesia and sum of 11 items) (Chouinard et al., 1980). Side effects were systematically recorded throughout the study and were assessed using a checklist administered by a resident of psychiatry on day 7, 14, 28, 42 and 56 (Table 3). Patients were randomized to receive celecoxib or placebo in a 1:1 ratio using a computer-generated code. Throughout the study, the person who administrated the medications, the rater and the patients were blind to assignments. 2.5. Statistical analysis A two-way repeated measures analysis of variance (time-treatment interaction) was used. The two groups as a between-subjects factor (group) and the five measurements during treatment as the within-subjects factor (time) were considered. This was done for positive, negative, general psychopathology subscale and PANSS total scores. A Greenhouse–Geisser correction was used for sphericity. In addition, a one-way repeated measures analysis of variance with a two-tailed post-hoc Tukey mean comparison tests were performed in the change from baseline in each group. The Tukey test is a popular post-hoc test that compares pairs of group means. To compare the two groups at baseline and the outcome of two groups at the end of the trial, an unpaired Student's t-test with a two-sided P value was used. To compare the demographic data and frequency of side effects between the protocols, Fisher's exact test was performed. Results are presented as mean ± SEM. Differences were considered significant with P b 0.05. To consider, α = 0.05, β = 0.2, the final difference between the two groups at least score of 5 on the PANSS rating scale, S = 5 and power = 0.8 (according to the pilot study of this research), the sample size was calculated at least 15 in

181

each group. Intention to treat (ITT) analysis with last observation carried forward (LOCF) procedure was performed. 3. Results Seventy two patients were screened for the study and 60 were randomized to trial medication (30 patients in each group). No significant differences were identified between patients randomly assigned to the group 1 or 2 condition with regard to basic demographic data including age, age of first onset of illness, gender, marital status, level of education, mean duration of illness and number of life-time hospitalization (Table 1). Although the number of dropout in the placebo group was higher than the celecoxib group (2 in the celecoxib group and 3 in the placebo group), no significant difference was observed in the two groups in terms of dropout (Fig. 1). 3.1. Positive symptoms The mean ± SEM scores of two groups of patients are shown in Fig. 2. There were no significant differences between the two groups at week 0 (baseline) on the PANSS (t = 0.17, df = 58, P = 0.86). The difference between the two treatments was significant as indicated by the effect of group, the between-subjects factor (Greenhouse–Geisser corrected: F = 4.00, df =1, P = 0.05, d (effect size) = 0.066). The behavior of the two treatment groups was homogeneous across time (groups-by-time interaction, Greenhouse–Geisser corrected: F = 3.06, df = 1.56, P = 0.06). In addition, a one-way repeated measures analysis of variance showed a significant effect Table 1 Baseline data

Gender

Celecoxib group

Placebo group

P

Male: 18, female: 12 33.10 ± 7.29 (year) 25.83 ± 5.90

Male: 17, female: 13 34.30 ± 7.21 (year) 24.56 ± 5.98

ns

Age (mean ± SD) Age of first onset of illness (mean ± SD) Level of education Under diploma: 23, diploma: 6, higher diploma: 1 Duration of illness 7.79 ± 5.87 (year) (mean ± SD) Marital status Single: 20, married: 8, divorced: 2 Number of life-time 4.06 ± 2.03 hospitalization (mean ± SD)

ns ns

Under diploma: 21, ns diploma: 7, higher diploma: 2 7.98 ± 5.87 (year) ns Single: 21, married: 7, divorced: 2 4.26 ± 1.98

ns

ns

182

S. Akhondzadeh et al. / Schizophrenia Research 90 (2007) 179–185

Fig. 1. Trial profile.

The mean ± SEM scores of two groups of patients are shown in Fig. 3. There were no significant differences

between the two groups at week 0 (baseline) on the PANSS (t = 0.54, df = 58, P = 0.58). The difference between the two treatments was not significant as indicated by the effect of group, the between-subjects factor (Greenhouse–Geisser corrected: F = 0.004, df = 1, P = 0.95). The behavior of the two treatment groups was homogeneous across time (groups-by-time interaction, Greenhouse–Geisser corrected: F = 1.12, df = 1.48, P = 0.31). In addition, a one-way repeated measures analysis of variance showed a significant effect of both treatments on the negative subscale scores of PANSS

Fig. 2. Mean ± SEM of the two protocols on the positive subtotal scores of the PANSS. ns = non-significant, ⁎P b 0.05 and ⁎⁎⁎P b 0.001.

Fig. 3. Mean ± SEM of the two protocols on the negative subtotal scores of the PANSS. ns = non-significant, ⁎⁎ P b 0.01 and ⁎⁎⁎P b 0.001.

of both treatments on the positive subscale scores of PANSS rating scale (P b 0.0001). In both groups posthoc comparisons showed a significant change from week 2. The difference between the two treatments was significant at the endpoint (week 8) (t = 2.73, df = 44, P = 0.03). 3.2. Negative symptoms

S. Akhondzadeh et al. / Schizophrenia Research 90 (2007) 179–185

183

rating scale (P b 0.0001). The difference between the two treatments was not significant at the endpoint (week 8) (t = 0.56, df = 58, P = 0.57). 3.3. General psychopathological symptoms The mean ± SEM scores of two groups of patients are shown in Fig. 4. There were no significant differences between the two groups at week 0 (baseline) on the PANSS (t = 0.40, df = 58, P = 0.68). The difference between the two treatments was significant as indicated by the effect of group, the between subjects factor (Greenhouse–Geisser corrected: F = 4.58, df = 1, P = 0.03, d (effect size) = 0.073). The behavior of the two treatment groups was not homogeneous across time (groups-by-time interaction, Greenhouse–Geisser corrected: F = 10.94, df = 1.78, P b 0.001). In addition, a one-way repeated measures analysis of variance showed a significant effect of both treatments on the general psychopathological symptoms subscale scores of PANSS rating scale (P b 0.0001). In both groups posthoc comparisons showed a significant change from week 2. The difference between the two treatments was significant at the endpoint (week 8) (t = 4.50, df = 58, P b 0.001). 3.4. PANSS total scores The mean ± SEM scores of two groups are shown in Fig. 5. There were no significant differences between the two groups at week 0 (baseline) on the PANSS (t = 0.76, df = 58, P = 0.44). The difference between the two treatments was significant as indicated by the effect of group, the between-subjects factor (Green-

Fig. 5. Mean ± SEM of the two protocols on the total scores of the PANSS. ns = non-significant, ⁎⁎P b 0.01 and ⁎⁎⁎P b 0.001.

house–Geisser corrected: F = 5.16, df = 1, P = 0.02 d (effect size) = 0.082). The behavior of the two treatment groups was not homogeneous across time (groups-bytime interaction, Greenhouse–Geisser corrected: F = 9.95, df = 1.71; P b 0.001). In addition, a one-way repeated measures analysis of variance showed a significant effect of both treatments on the total scores of PANSS rating scale (P b 0.0001). In both groups post-hoc comparisons showed a significant change from week 2. The difference between the two treatments was significant at the endpoint (week 8) (t = 3.17, df = 58, P = 0.002). The changes at the endpoint compared to baseline were: − 43.70 ± 15.91 (mean ± SD) and − 29.16 ± 14.84 for celecoxib and placebo, respectively. A significant difference was observed on the PANSS total at week 8 compared to baseline in the two groups (t = 3.65, df = 58, P b 0.001). 3.5. Extrapyramidal Symptoms Rating Scale Although the means ESRS for the placebo group were higher than celecoxib group over the trial, the differences were not statistically significant (Table 2). No significant difference was observed between the overall mean Table 2 Extrapyramidal symptoms based on extrapyramidal symptoms rating scale Mean ± SEM

Fig. 4. Mean ± SEM of the two protocols on the general psychopathology subtotal scores of the PANSS. ns = non-significant and ⁎⁎⁎P b 0.001.

Week 0 Week 1 Week 2 Week 4 Week 6 Week 8

P

Celecoxib group

Placebo group

1.23 ± 0.45 4.40 ± 0.91 7.16 ± 1.66 5.26 ± 0.61 4.20 ± 0.65 3.20 ± 0.54

1.20 ± 0.50 7.73 ± 1.10 7.86 ± 1.89 5.86 ± 1.03 4.88 ± 0.87 3.63 ± 0.66

0.96 0.81 0.78 0.61 0.54 0.61

184

S. Akhondzadeh et al. / Schizophrenia Research 90 (2007) 179–185

Table 3 Number of patients with side effects Side Effects Daytime drowsiness Morning drowsiness Constipation Dizziness Stiffness Slowed movement Tremor Increased appetite Loss of appetite Nervousness Restlessness Skin rash Urinary retention Diarrhea Dry mouth Fatigue Sexual dysfunction

Risperidone + celecoxib

Risperidone + placebo

P

4 (13%)

3 (10%)

ns

8 (26%)

5 (16%)

ns

2 (6%) 6 (20%) 6 (20%) 8 (26%) 6 (20%) 2 (6%) 6 (20%) 4 (13%) 8 (26%) 2 (6%) 2 (6%) 2 (6. %) 10 (33%) 2 (6%) 1 (3%)

3 (10%) 4 (13%) 5 (16%) 6 (30%) 8 (26%) 4 (13%) 5 (16%) 5 (16%) 5 (16%) 3 (10%) 3 (10%) 3 (10%) 9 (30%) 1 (3%) 3 (10%)

ns ns ns ns ns ns ns ns ns ns ns ns ns ns ns

biperiden dosages in two groups. (114.66 ± 104.25 and 136.30 ± 74.98 for celecoxib and placebo group respectively (mean ± SD) (P = 0.36; t = 0.92; df = 58). Moreover, the difference between the two treatments in terms of the number of days of biperiden treatment was not significant (19.09 ± 17.73 and 22.69 ± 12.49 for celecoxib and placebo group respectively (mean ± SD) (P = 0.35; t = 0.92; df = 58). 3.6. Clinical complications and side effects Seventeen side effects were observed over the trial. The difference between the celecoxib and placebo in the frequency of side effects was not significant (Table 3). 4. Discussion The syndrome of schizophrenia is likely to be etiologically heterogeneous and a multitude of CNS disorders can give rise to schizophrenia-like psychoses (Akhondzadeh, 2006). Accumulating evidence suggests that aberrant immune function may be implicated in the pathogenesis of this severe disease (Ganguli et al., 1993; Licinio et al., 1993; Müller et al., 1999; Kronfol and Remick, 2000). Chronically activated macrophages and T-lymphocytes, along with excessive interleukin-2 (IL2) and other cytokine secretions has been hypothesized as the cause of schizophrenia (Bessler et al., 1995; Sirota et al., 1995; Kronfol and Remick, 2000; Cazzullo et al., 2001). There are now a number of indications that

alteration in both the cellular and the humoral arm of immune system may be involved in the pathogenesis of schizophrenia, suggesting suppression of some immune functions and activation of others (Müller et al., 1999). The most consistent findings were decreased levels of the T-cell cytokines IL-2 (Licinio et al., 1993; Sirota et al., 1995) as well as increased levels of the soluble IL-2 receptor in a whole blood assay of schizophrenic patient samples after stimulation with phytohemagglutinin (PHA) (Müller et al., 1997). Different mechanisms of action have been reported for direct effects of COX-2 inhibitors on an altered immune state of patients with schizophrenia including regulation of cytokine production, inhibition of synthesis of proinflammatory prostaglandins, rebalancing Th1 and Th2 responses and a non-immunological mechanisms, such as direct neuroprotection via downregluation of NMDA receptormediated neurotoxicity (Riedel et al., 2005). Previous reports have demonstrated contradictory results regarding the beneficial effect of celecoxib adjunctive therapy for patients with schizophrenia. Our hypothesis was that celecoxib augmentation would cause improvement in patients with schizophrenia (Müller et al., 2002; Rapaport et al., 2005; Riedel et al., 2005). In this study as expected, both groups of patients showed significant improvement on the Positive and Negative Syndrome Scale and on its positive and general psychopathological symptoms subscales during the 8 weeks of treatment with risperidone. In agreement with our hypothesis, the celecoxib group had significantly greater improvement in the positive symptoms, general psychopathological symptoms and PANSS total scores over 8 weeks trial. No significant differences were observed between the means of the two groups on the negative scores. Clinical characteristics of the patients, such as sex, age and duration of illness, did not differ between groups and cannot explain differences in the therapeutic outcome. Extrapyramidal side effects measured by the ESRS did not show any differences between the two groups over the trial. The use of biperiden was greater in the group receiving risperidone + placebo but the difference was not statistically significant. Therapy with 400 mg/day of celecoxib was well tolerated, and no clinically important side effects were observed. The therapeutic benefit of the combined therapy has to be attributed to effects of celecoxib. This finding might reflect a complex interplay between the anti-inflammatory effects and the modulation of glutamatergic and dopaminergic systems by the COX-2 inhibitor (Yermakova and O'Banion, 2000; Riedel et al., 2005) However, the limitations of the present study, including the relatively

S. Akhondzadeh et al. / Schizophrenia Research 90 (2007) 179–185

short period of study and only a fixed dose of celecoxib should be taken into account and this indicates the need for further research. In addition, from a scientific viewpoint, the therapeutic effects of celecoxib without an additional neuroleptic drug would be more interesting. However, since neuroleptics are effective in antipsychotic treatment, ethic committees would not approve a study with celecoxib as the only drug for acutely ill schizophrenic patients. 5. Conclusion The present study indicates celecoxib as a potential adjunctive treatment strategy for chronic schizophrenia and anti-inflammatory therapies should be further investigated. Acknowledgments This study was Dr. Maryam Tabatabaee's postgraduate thesis. This study was supported by a grant (grant number: 1741) from Tehran University of Medical Sciences to Dr. Shahin Akhondzadeh. References Anderson, N.C., 2000. Schizophrenia: the fundamental questions. Brain Res. Rev. 31, 106–112. Akhondzadeh, S., 1998. The glutamate hypothesis of schizophrenia. J. Clin. Pharm. Ther. 23, 243–246. Akhondzadeh, S., 2001. The 5-HT hypothesis of schizophrenia. IDrugs 4, 295–300. Akhondzadeh, S., 2006. Pharmacotherapy of schizophrenia: the past, present and future. Curr. Drug Ther. 1, 1–7. Baik, E.J., Kim, E.J., Lee, S.H., Moon, C., 1999. Cyclooxygenase-2 selective inhibitors aggravate kainic acid induced seizure and neuronal cell death in the hippocampus. Brain Res. 843, 118–129. Bessler, H., Levental, Z., Karp, L., Modai, I., Djaldetti, M., Weizman, A., 1995. Cytokine production in drug-free and neuroleptic-treated schizophrenic patients. Biol. Psychiatry 38, 297–302. Cazzullo, C.L., Sacchetti, E., Galluzzo, A., Panariello, A., Colombo, F., Zagliani, A., Clerici, M., 2001. Cytokine profiles in drug-naive schizophrenic patients. Schizophr. Res. 47, 293–298. Chouinard, G., Ross-Chouinard, A., Annables, L., Jones, B.D., 1980. Extrapyramidal Symptoms Rating Scale (abstract). Can. J. Neurol. Sci. 7, 233. Ganguli, R., Brar, J.S., Chengappa, K.N.R., Yang, Z.W., Nimgaonkar, V.L., Rabin, B.S., 1993. Autoimmunity in schizophrenia: a review of recent findings. Ann. Med. 25, 489–496. Jablensky, A., Sartorius, N., Ernberg, G., 1991. Schizophrenia: manifestation, incidence and course in different cultures. A World Health Organization ten-country study. Psychol. Med. 20, 1–97 (suppl).

185

Kay, S.R., Fiszbein, A., Opler, L.A., 1987. The Positive and Negative Syndrome Scale for schizophrenia. Schizophr. Bull. 13, 261–276. Körschenhausen, D., Hampel, H., Ackenheil, M., Penning, R., Müller, N., 1996. Fibrin degradation products in post mortem brain tissue of schizophrenics: a possible marker for underlying inflammatory processes. Schizophr. Res. 19, 103–109. Kronfol, Z., Remick, D.G., 2000. Cytokines and the brain: implications for clinical psychiatry. Am. J. Psychiatry 157, 683–694. Lara, D.R., Souza, D.O., 2000. Schizophrenia: a purinergic hypothesis. Med. Hypotheses 54, 157–166. Licinio, J., Seibyl, J.P., Altemus, M., Charney, D.S., Krystal, J.H., 1993. Elevated CSF levels of interleukin-2 in neuroleptic-free schizophrenic patients. Am. J. Psychiatry 150, 1408–1410. Mohammadi, M.R., Akhondzadeh, S., 2001. Schizophrenia: etiology and pharmacotherapy. IDrugs 4, 1167–1172. Müller, N., Ackenheil, M., 1998. Psychoneuroimmunology and the cytokine action in the CNS: implications for psychiatric disorders. Prog. Neuro-Psychopharmacol. Biol. Psychiatry 22, 1–33. Müller, N., Empel, M., Riedel, M., Schwarz, M.J., Ackenheil, M., 1997. Neuroleptic treatment increases soluble IL-2 receptors and decreases soluble IL-6 receptors in schizophrenia. Eur. Arch. Psychiatry Clin. Neurosci. 247, 308–313. Müller, N., Riedel, M., Ackenheil, M., Schwarz, M.J., 1999. The role of immune function in schizophrenia: an overview. Eur. Arch. Psychiatry Clin. Neurosci. 249 (Suppl. 4), 62–68. Müller, N., Riedel, M., Scheppach, C., Branstätter, B., Sokullu, S., Krampe, K., Ulmschneider, M., Engel, R.R., Müller, H.J., Scwarz, J., 2002. Beneficial antipsychotic effects of celecoxib add-on therapy compared to risperidone alone in schizophrenia. Am. J. Psychiatry 159, 1029–1034. Müller, N., Ulmschneider, M., Scheppach, C., Schwarz, M.J., Ackenheil, M., Müller, H.J., Gruber, R., Riedel, M., 2004. COX-2 inhibition as a treatment approach in schizophrenia: immunological considerations and clinical effects of celecoxib add-on therapy. Eur. Arch. Psychiatry Clin. Neurosci. 254, 14–22. Nogawa, S., Zhang, F., Ross, M.E., Iadecola, C., 1997. Cyclooxygenase-2 gene expression in neurons contributes to ischemic brain damage. J. Neurosci. 17, 2746–2755. Rapaport, M.H., Delrahim, K.K., Bresee, C.J., Maddux, R.E., Ahmadpour, O., Dolnak, D., 2005. Celecoxib augmentation of continuously ill patients with schizophrenia. Biol. Psychiatry 57, 1594–1596. Riedel, M., Strassing, M., Scwarz, J., Müller, N., 2005. COX-2 inhibition as adjunctive therapy in schizophrenia. CNS Drugs 19, 806–819. Sirota, P., Schild, K., Elizur, A., Djaldetti, M., Fishman, P., 1995. Increased interleukin-1 and interleukin-3 like activity in schizophrenic patients. Prog. Neuro-Psychopharmacol. Biol. Psychiatry 19, 75–83. World Medical Association. Declaration of Helsinki, 2000. Ethical Principles for Medical Research Involving Human Subjects. Available at: http://www.wma.net. Yermakova, A., O'Banion, M.K., 2000. Cyclooxygenases in the central nervous system: implications for treatment of neurological disorders. Curr. Pharm. Des. 6, 1755–1776.