Ceruletide for Schizophrenia:A Double-Blind Study
Jeffrey A. Mattes, William Hom, Joseph M. Rochford, and Michael Orlosky
Ceruletide,an analog of cholecystokinin (CCK), has been reported to have neuroleptic-like activity in mice, and, in three open studies,to benefit schizophrenic patients.This study evaluated ceruletide in schizophrenia using a double-blind design. Subjects were tics.Patients randomly received two injections, 1 week apart, of either ceruletide (0.6 μg/kg im) or placebo,while continuing neuroleptics;this regimen was found helpful in earlier studies.Evaluation included ratings of 29 variables related to prognosis in schizo-phrenia (e.g., age, number of previous hospitalizations), regular BPRSs and SCL-90s, and psychiatrist, patient, and relative ratings of global improvement. Results showed few significant differences between ceruletide and placebo,with exceptions as likely to favor placebo as ceruletide. Among the patients on ceruletide, no predictors of benefit were
Introduction
The majority of schizophrenics on an optimal regimen of neuroleptic medication have residual symptoms, with functional and social impairment. Additional treatment ap-proaches,therefore,are needed.A compound recently reported to improve schizophrenics already on neuroleptics is ceruletide. Ceruletide, found in amphibians, is a decapeptide that is nearly identical structurally to the C-terminal amino acid sequence of cholecys-tokinin (CCK), a 33 amino acid polypeptide secreted by the duodenum to initiate con-traction of the gallbladder during digestion.
As with other polypeptides(e.g.,adrenocorticotropic hormone,vasopressin),CCK has been found to have central nervous system (CNS) effects that are apparently unrelated to its originally discovered hormonal effect.There is evidence that CCK is present in the brain(Beinfeld 1981;Morley 1982), acting as a neurotransmitter or neuromodulator (Snyder 1980).Receptors for CCK have been found (Saito et al. 1980; Hays et al. 1982; Itoh et al.1982). In fact,CCK seems to be one of the most highly concentrated of the putative regulatory peptides in the brain (Noyer et al. 1980).
From the Carrier Foundation,Belle Mead,NJ.
Address reprint requests to Dr.Jeffrey A.Mattes,Carrier Foundation,Belle Mead,NJ 08502.
Received July 6, 1984.
©1985 Society of Biological Psychiatry

0006-3223/85/$03.30 
534

BIOL PSYCHIATRY
1985;20:533-538

J.A.Mattes et al.
Relevant to schizophrenia, Hokfelt et al. (1980) discovered the coexistence of CCK and dopamine in mesolimbic,but not nigrostriatal, dopamine neurons. In addition,CCK has been reported to reduce dopamine turnover in the caudate nucleus and nucleus ac-cumbens (Fuxe et al. 1980),to increase striatal dopamine release (Kovacs et al. 1981), to affect dopamine binding (Fuxe et al. 1981; Murphy and Schuster 1982), and to modify dopamine receptor sensitivity (Hommer and Skirbol 1983). Zetler(1981) found ceruletide to have neuroleptic-like activity in mice; also, Cohen et al. (1982) found that combining CCK-8 with haloperidol reduced signaled avoidance behavior significantly more than either drug did alone.
Moroji et al.(1982) evaluated ceruletide in an open study in 20 schizophrenic patients refractory to neuroleptics. Results indicated improvement in both positive (e.g.,hallu-cinations, delusions) and negative (e.g.,withdrawal, blunted affect) symptoms after a single im injection of 0.3 or 0.6 μg/kg to patients who were on stable dosages of neuroleptics for 3 weeks prior to the injection.The improvement was evident 2 hr after the dose,peaking 1 week later.Moroji et al.found the higher dose, 0.6 μg/kg,to be more beneficial than the lower one.Side effects,slight abdominal pains and borborygmus, were minimal. Itoh et al. (1982), using a similar open design and comparable dosage, also reported benefit in 23 of 58 chronic schizophrenics;improvement was primarily in mood and spontaneity rather than in psychotic symptoms.
Nair et al.conducted two open trials of cholecystokinin (1982, 1983b) and one open trial of the octapeptide CCK-8 (1983a) in a total of 35 chronic schizophrenics and also reported generally positive results. Nair et al. (1983a) subsequently conducted a placebo-controlled study of CCK in 17 schizophrenic inpatients and reported positive effects on Brief Psychiatric Rating Scales (BPRS) for anxiety-depression and thought-disturbance 7 days after a single infusion;however, the authors noted that the study was not truly double-blind,as patients fairly consistently noted epigastric discomfort from the dose of CCK used.Peselow et al.(1983), in the second double-blind trial using the octapeptide CCK-8,reported generally negative results. Nair et al. (1983b), however,found CCK-8 to show less striking results than CCK itself. Hommer et al. (1984), in a double-blind crossover study, found no benefit from ceruletide in eight chronic schizophrenics,and no benefit from CCK-8 in four patients; however, the numbers were small, and the use of a crossover design may have obscured drug-placebo differences,as the drug has been reported to act for up to 3 weeks (Moroji et al. 1982).
The purpose of this study was to more systematically evaluate ceruletide, using a double-blind design, in schizophrenics with less than an optimal response to neuroleptics.
Methods
Subjects
Subjects were inpatients at the Carrier Foundation who had a DSM-III diagnosis of schizophrenia,chronic or subchronic, and who had previously manifested a less than optimal response to neuroleptics.Patients who had never before received a good trial on neuroleptics (defined as at least 3 weeks on a dosage equivalent to 800 mg/day of chlorpromazine) were excluded,unless they had tried, but could not tolerate, that dosage. In addition,patients were excluded if, during their most recent adequate trial on neu-roleptics,psychotic symptoms were eliminated,and they were able to return to a relatively normal lifestyle.Patients with organic mental disorders were also excluded. 
Ceruletide for Schizophrenia

BIOL PSYCHIATRY
1985;20:533-538

535
Treatment
Prior to the administration of ceruletide,patients were given a 3-week trial of neuroleptics at optimal dosage (at least the equivalent of 800 mg/day of chlorpromazine, if tolerated). Unless improvement was approaching the optimal (i.e.,minimal symptomatology with adequate social and role functioning), patients were then randomized to receive either ceruletide or placebo.Each patient received two injections,1 week apart,either both ceruletide or both placebo.The dose of ceruletide was 0.6 μg/kg. Patients continued on neuroleptics throughout the study at constant dosage, except for clinical emergencies. Anticholinergics were administered as clinically indicated for neuroleptic side effects.
Evaluation
DSM-III diagnoses were made, recording the specific symptoms that justified the diag-nosis,and the DSM-III criteria for subtyping and classifying the course of illness were used.Data thought to predict prognosis in schizophrenia were recorded prior to admin-istering ceruletide,using a structured interview format.Sections from the Schedule for Affective Disorders and Schizophrenia,Lifetime version (the SADS-L),were included, covering age,education,marital status, age at first psychiatric hospitalization, number of hospitalizations, total duration of hospitalizations, and ratings of social,work,and overall functioning.Ratings were also made of degree of thought disorder, passivity, and blunted affect during remission,extent of recovery from previous psychotic episodes, and previous response to medication.
Brief Psychiatric Rating Scales and patient SCL-90s were completed at baseline.These were repeated with patient,psychiatrist, and relative ratings of global improvement after 1 week (just prior to the second injection) and after 2 weeks.Patient and psychiatrist global ratings were also made 2 hr and 2days after the first injection.
Results
Nine patients received ceruletide and eight received placebo.The two groups were com-pared(t-tests) on all baseline measures.There were no differences in regard to diagnosis, symptoms justifying the diagnosis, or on the other 17 variables recorded on the baseline structured interview form.The groups differed significantly on 1 of the 19 BPRS ratings, with the group receiving placebo being healthier at baseline than the group on ceruletide on the rating of hallucinations [mean 1.50 (SD 1.07) and 3.63 (SD 2.00),respectively; t=2.65,p<0.05]. In addition,the placebo group was healthier at baseline on 1 of the 12(the 9 factor and the 3 composite) SCL-90 scores: Factor 6, Hostility [mean 3.60 (SD 4.16) and 12.83 (SD 8.13),respectively;t=2.27,p<0.05].Side effects were minimal;only three patients noted mild abdominal distress, and one of these was on placebo.
There was no substantial evidence that ceruletide helped patients more than placebo. None of the improvement ratings, by patient,psychiatrist,or relative,at any of the four rating times, showed a significant difference (t-tests) between the groups. The BPRS and SCL-90 ratings after 1 week similarly yielded no significant differences. The ratings after 2 weeks yielded two significant (p <0.05)differences on the BPRS on the Anxiety and Depressed Mood ratings. In both cases, the group on placebo was doing better.Means for Anxiety were 2.20(SD 0.84) and 3.50 (SD 0.93) for the placebo and ceruletide groups, 
536

BIOL PSYCHIATRY
1985;20:533-538

J.A.Mattes et al.
respectively(t=2.55). Means for Depressed Mood were 1.60 (SD 0.55) and 2.75(SD 1.04),respectively(t=2.27).
As the placebo group had been healthier at baseline on two ratings (one on the BPRS and one on the SCL-90), the 2-week ratings on these variables were compared (ceruletide versus placebo),covarying out the baseline ratings;the results were still nonsignificant. The groups were similar on baseline total BPRS score (46.63 versus 51.00 for the placebo and ceruletide groups,respectively),but differed more on the SCL-90 General Symp-tomatic Index (GSI:the mean of all 90 items); means were 1.01 (SD 0.84) and 1.84 (SD 1.02),respectively,though this was still nonsignificant.We therefore compared global ratings after 2 weeks (ceruletide versus placebo), covarying out baseline SCL-90 GSI scores; again, results were not significant.
Despite the lack of group differences,it is possible that a subgroup of patients responded to ceruletide.To look for predictors of response,correlations were calculated between global improvement ratings (after 1 and 2 weeks) and all baseline variables. This was done separately for the ceruletide and placebo groups. Results did not indicate any clear predictors; for the ceruletide group,only 22 of 288 correlations were significant,which is only slightly more than the number that would be expected by chance.Some of the significant correlations,for both the ceruletide and placebo groups, indicated only that patients with more pathology at baseline improved the most; this was a “regression to the mean" effect that was apparently unrelated to ceruletide. There was a suggestion, though this was not statistically significant,that more BPRS rated hallucinations, sus-piciousness,and hostility predicted more improvement on ceruletide but not on placebo.
Discussion
This study did not find beneficial effects on ceruletide in chronic schizophrenics.Given the previous positive reports, the sample, dosage, and evaluation used herein would appear to be adequate; the most likely explanation for the discrepant results,therefore, is that this study was double-blind. In fact,most patients improved somewhat(the median global improvement rating, rated from much worse to much better, was minimal im-provement),and several patients,subjectively and objectively, improved quite dramati-cally and wanted to continue ceruletide after the study. However, as many improvers were on placebo as on ceruletide; a nonblind study would have suggested a beneficial effect.
Conceivably higher dosage will yield more positive results.Effects of the cholecys-tokinin octapeptide (CCK-8) in rats, in affecting sleep latency,were present at 10 mg/ kg but not at 1 mg/kg (Rojas-Ramirez et al. 1982); we used 0.06 μg/kg. Animal studies reporting CNS effects of CCK analogs generally used dosages about 10 times the dosage used in the studies to date with schizophrenics.
There is still a question as to whether or not ceruletide crosses the blood-brain barrier (Cornford et al. 1978; Itoh et al. 1982). Analogs do affect sleep and appetite,but this may be through a peripheral mechanism, as some of these effects have been eliminated by vagotomy (Smith et al. 1981). The mechanisms of an apparent analgesic effect (Piazza et al. 1981) and of effects on dopamine receptors (Hommer and Skirboll 1983) are unknown.
One would not expect that CCK analogs other than ceruletide would be more effective in schizophrenics; the CCK fragment found most in mammalian brains is the octapeptide CCK-8 (Rehfeld 1978),but ceruletide has been found to be far more potent than CCK- 
Ceruletide for Schizophrenia

BIOL PSYCHIATRY
1985:20:533-538

537
8 in respect to CNS effects (Itoh et al. 1982). Also, ceruletide and CCK bind to the same receptors in all known receptor systems (Piazza et al. 1981).
Given the wide distribution and high concentration of ceruletide in the CNS, and its apparent interaction with dopamine and other neurotransmitters, it would be expected that CCK analogs, and eventually CCK agonists and antagonists,will continue to be of interest to psychiatrists. However,clinical relevance may depend on the development of analogs that more easily cross the blood-brain barrier or on drugs structurally unrelated to CCK that nevertheless affect CCK receptors,synthesis, or release.
The authors thank Sarah Robin,Mary Anne Hinz,Marie Magee,and Al Abrams.
References
Beinfeld MC, Palkovits M (1981):Distribution of cholecystokinin (CCK) in the hypothalamus and limbic system of the rat.Neuropeptides 2:123-129.
Cohen SL,Knight M, Tamminga CA, et al (1982):Cholecystokinin-octapeptide effects on con-ditioned-avoidance behavior,stereotypy and catalepsy. Eur J Pharmacol 83:213-222.
Cornford EM, Braun LD,Crane PD,et al (1978): Blood-brain barrier restrictions of peptides and the lower uptake of enkephalin. Endocrinology 103:1297-1303.
Fuxe K,Andersson V,Locatelli LF,et al(1980): Cholecystokinin peptides produce marked re-duction of dopamine turnover in discrete areas in the rat brain following intraventricular injection. Eur J Pharmacol 67:329-331.
Fuxe K,Agnati L, Benfenati F,et al (1981): Modulation by cholecystokinins of ['H)spiroperidol binding in rat striatum:Evidence for increased affinity and reduction in number of binding sites. Acta Physiol Scand 113:567-569.
Hays SE,Beinfeld MC,Jensen RT,et al (1982):Demonstration of a putative receptor site for cholecystokinin in rat brain.Neuropeptides 1:53-62.
Hokfelt T, Skirboll L, Rehfeld JF, et al (1980): A subpopulation of mesencephalic dopamine neurons projecting to limbic areas contains a cholecystokinin-like peptide:Evidence from im-munohistochemistry combined with retrograde tracing. Neuroscience 5:2093-2124.
Hommer D,Skirboll L (1983):Cholecystokinin-like peptides potentiate opomorphine-induced in-hibition of dopamine neurons.Eur J Pharmacol 91:151-152.
Hommer DW,Pickar D,Roy A, et al (1984):The effects of ceruletide in schizophrenia.Arch Gen Psychiatry 41:617-619.
Itoh H,Tanoue S,Yagi G,et al (1982):Clinical study on the psychotropic effects of caerulein-An open clinical trial in chronic schizophrenic patients. Keio J Med 31:71-95.
Kovacs GL,Szabo G,Penke B,et al (1981):Effects of cholecystokinin octapeptide on the striatal dopamine metabolism and an apomorphine-induced stereotyped cageclimbing in mice. Eur J Pharmacol 69:313-319.
Morley JE (1982):The ascent of cholecystokinin(CCK)-From gut to brain. Life Sci 30:479-493.
Moroji T,Watanabe N,Aoki N, et al (1982): Antipsychotic effects of ceruletide (caerulein) on chronic schizophrenia.Arch Gen Psychiatry 39:485.
Murphy RB,Schuster DI(1982):Modulation of H3-dopamine binding by cholecystokinin.Peptides 3:539-543.
Nair NPV,Bloom DM,Nestoros JN (1982):Cholecystokinin appears to have antipsychotic prop-erties. Prog Neuro-Psychopharmacol Biol Psychiatry 6:509-512.
Nair NPV, Bloom DM, Schwartz G(1983a):Cholecystokinin and the Treatment of Chronic Schizo-phrenia.Paper presented at the 136th Annual Meeting of the American Psychiatric Association, New York. 
538

BIOL PSYCHIATRY
1985;20:533-538

J.A.Mattes et al.
Nair NPV, Bloom DM,Nestoros JN, et al (1983b):Therapeutic efficacy of cholecystokinin in neuroleptic resistent schizophrenic subjects. Psychopharmacol Bull 19:134-136.
Noyer M,Bui ND,Deschodt-Lanckman M, et al (1980): Postnatal development of the cholecys-tokinin-gastrin family of peptides in the brain and gut of rat. Life Sci 27:2197-2203.
Peselow ED,Angrist BM,Siekierski J, et al(1983):Cholecystokinin(CCK) in the Treatment of Chronic Schizophrenic Patients.Paper presented at NCDEU (New Clinical Drug Evaluation Unit) Annual Meeting, Key Biscayne,FL.
Piazza E,Brambilla M, Cattaneo MT, et al (1981): Analgesic activity of ceruletide in cancer patients. In Basso N, Lezocke E, Speronza V,Walsh JH (eds), International Symposium on Brain-Gut Axis,Abstracts. Florence, p 146.
Rehfeld JF (1978):Immunochemical studies on cholecystokinin. J Biol Chem 253:4022-4027.
Rojas-Ramirez JA,Crowley JM, Mendelson WB (1982): Electroencephalographic analysis of the sleep inducing actions of cholecystokinin. Neuropeptides 3:129-138.
Saito A, Sankaron H, Goldfine ID, et al (1980): Cholecystokinin receptors in the brain: Charac-terization and distribution. Science 208:1155-1156.
Smith GP,Jerome C,Cushin BJ,et al (1981):Abdominal vagotomy blocks the satiety effect of cholecystokinin in the rat. Science 213:1036-1037.
Snyder SH (1980):Brain peptides as neurotransmiters. Science 209:976-983.
Zetler G(1980):Effects of cholecystokinin-like peptides on rearing activity and hexobarbital-induced sleep. Eur J Pharmacol 66:137-139.
Zetler G(1981):Central effects of ceruletide analogues. Peptides (Suppl) 2:65-69. FI-6934