Clozapine Augmented With Risperidone in the Treatment of

Document Sample
Clozapine Augmented With Risperidone in the Treatment of Powered By Docstoc

        Clozapine Augmented With Risperidone
          in the Treatment of Schizophrenia:
  A Randomized, Double-Blind, Placebo-Controlled Trial

Richard C. Josiassen, Ph.D.                 Objective: The authors evaluated the ef-       with clozapine/risperidone treatment than
                                            ficacy and safety of augmenting clozapine      with clozapine/placebo. The adverse event
                                            with risperidone in patients with treat-       profile for clozapine/risperidone treat-
Ashok Joseph, M.D.
                                            ment-resistant schizophrenia.                  ment was similar to that for clozapine/pla-
                                                                                           cebo. Simpson-Angus Rating Scale scores
Eva Kohegyi, M.D.                           Method: In a randomized, double-blind,
                                            placebo-controlled 12-week trial, 40 pa-       were lower with clozapine/risperidone
                                            tients unresponsive or partially responsive    treatment throughout the trial but in-
Sudhir Stokes, M.D.
                                            to clozapine monotherapy received a            creased to approach those of clozapine/
                                            steady dose of clozapine combined with         placebo treatment at week 12. Clozapine/
Mahmood Dadvand, M.D.                       either placebo (N=20) or up to 6 mg/day of     risperidone treatment did not induce addi-
                                            risperidone (N=20). Patient psychopathol-      tional weight gain, agranulocytosis, or sei-
Wynn Wynn Paing, M.D.                       ogy was assessed at 2-week intervals with      zures compared with clozapine/placebo
                                            the Brief Psychiatric Rating Scale (BPRS)      treatment.
Rita A. Shaughnessy, M.D., Ph.D.            and the Scale for the Assessment of Nega-
                                            tive Symptoms (SANS), among other mea-         Conclusions: In patients with a subopti-
                                            sures. Movement disorders were assessed        mal response to clozapine, the addition of
                                            with the Simpson-Angus Rating Scale.           risperidone improved overall symptoms
                                            Results: From baseline to week 6 and           and positive and negative symptoms of
                                            week 12, mean BPRS total and positive          schizophrenia. The combination appears
                                            symptom subscale scores were reduced           to be safe and well tolerated. Augmenta-
                                            significantly in both groups, but the reduc-   tion of clozapine with risperidone may
                                            tions were significantly greater with cloza-   provide additional clinical benefit for pa-
                                            pine/risperidone treatment. Reductions in      tients who are nonresponsive or only par-
                                            SANS scores were also significantly greater    tially responsive to clozapine alone.

                                                                                                 (Am J Psychiatry 2005; 162:130–136)

T    he landmark study by Kane et al. (1) established
clozapine as the treatment of choice for severely ill pa-
                                                                    week randomized crossover phases. Although higher
                                                                    clozapine doses were associated with a greater reduction
tients with schizophrenia whose psychotic symptoms are              in symptom severity, overall efficacy was disappointing:
refractory to conventional antipsychotic treatment. In the          only five of 50 patients were judged to be responsive. In a
Kane study, 30% of patients with refractory symptoms re-            meta-analysis, Wahlbeck et al. (5) found that clozapine
sponded to treatment with clozapine. That finding led to            was superior to conventional antipsychotics in controlling
renewed interest in clozapine and an optimistic outlook             symptoms in patients with refractory schizophrenia, but
regarding improved treatment for severely ill psychotic             only one-third of patients met the criteria for a clinically
patients. Meltzer (2) reported in an uncontrolled study that        meaningful treatment response. According to a review of
6 months of clozapine treatment yielded a good clinical re-         data from clozapine clinical trials by Buckley et al. (6),
sponse in 50% of patients whose symptoms had previ-                 about half of patients with symptoms refractory to other
ously been refractory.                                              antipsychotic agents are also nonresponsive to clozapine.
  However, a considerable number of patients treated                These authors estimate that some 325,000 patients in the
with clozapine are still nonresponsive or only partially re-        United States with refractory schizophrenia spectrum dis-
sponsive. For example, in a double-blind comparative                orders are also nonresponsive to clozapine or would be if
study, Rosenheck et al. (3) reported a lower response rate          offered this treatment.
with clozapine than that reported by either Kane et al. (1)           There is little evidence to guide the next treatment steps
or Meltzer (2) and that the drug was only slightly better           for patients who do not respond to clozapine monother-
than the conventional antipsychotic haloperidol. A dou-             apy at recommended doses. To improve response rates,
ble-blind study by Simpson et al. (4) compared three clo-           U.S. psychiatrists began to administer clozapine in doses
zapine doses (100, 300, and 600 mg/day) in sequential 16-           higher than those administered by European psychia-

130                                                   Am J Psychiatry 162:1, January 2005
                                                                                              JOSIASSEN, JOSEPH, KOHEGYI, ET AL.

trists, but higher doses did not appear to improve efficacy         sulting from higher overall drug exposure, leading to
(7). Others reasoned that the efficacy of clozapine, an             greater liability for tardive dyskinesia. Complex drug regi-
agent with broad-spectrum receptor activity but relatively          mens can also compromise patient compliance with treat-
weak dopamine D2-antagonist properties, might be en-                ment, and their benefit may not justify their costs. Given
hanced by augmentation with an antipsychotic that pro-              the substantial public health burden of clozapine-refrac-
vides high-potency D2 blockade (8).                                 tory schizophrenia and the prevalent use of augmentation
   A limited number of published case reports and small             strategies, rigorous investigation of these strategies is of
studies have reported the effects of clozapine augmenta-            utmost importance. To that end, we conducted a 12-week,
tion with other antipsychotics. Sulpiride, a relatively selec-      randomized, double-blind, placebo-controlled study of
tive D2 antagonist, showed promise in a placebo-con-                the efficacy and safety of clozapine augmented with ris-
trolled clozapine augmentation trial (9). In this study,            peridone in a well-defined group of patients with schizo-
eight of 16 patients treated with clozapine and sulpiride,          phrenia refractory to clozapine monotherapy.
compared with only one of 12 patients given clozapine
and placebo, demonstrated a score reduction of 20% or               Method
more on the Brief Psychiatric Rating Scale (BPRS). How-
ever, in a controlled trial conducted in China, clozapine           Patients
augmentation with chlorpromazine was not superior to                   Inpatients or outpatients who continued to experience signifi-
clozapine monotherapy (10). Case reports and a small pa-            cant psychotic symptoms despite adequate treatment with cloza-
                                                                    pine were eligible to participate in this trial. Study participants
tient series have suggested promise for clozapine aug-              were recruited from a large population of state hospital inpatients
mentation with loxapine (11), pimozide (12), and olanzap-           and recently discharged patients living in highly structured com-
ine (13), but rigorous trials of these agents have not been         munity settings. Thorough records of psychiatric history and
conducted.                                                          prior antipsychotic drug therapy were available for all patients.
                                                                    Patients were included if they 1) had a DSM-IV diagnosis of
   Risperidone is, to date, the most extensively docu-
                                                                    schizophrenia or schizoaffective disorder; 2) were 20–65 years of
mented clozapine augmentation agent. Compared with                  age; 3) had, before treatment with clozapine, documented treat-
clozapine, risperidone has greater affinity for D2 and sero-        ment failure after two antipsychotics approved by the U.S. Food
tonin 5-HT2 receptors. Multiple case reports (14–18) have           and Drug Administration were administered for an adequate du-
described a benefit from clozapine augmentation with ris-           ration in a sufficient dose (6 or more weeks of 1000 mg/day of
                                                                    chlorpromazine equivalents); 4) demonstrated a documented
peridone, but several others (reviewed by Chong and
                                                                    failure to show a satisfactory clinical response to an adequate trial
Remington [19]) have reported negative results. Of greater          of clozapine (3 or more months of at least 600 mg/day of oral cloz-
interest are three uncontrolled prospective treatment               apine or a plasma drug level of 350 ng/ml or higher); and 5) had
studies in larger series of patients. In the first study (20), 12   persistent psychotic symptoms, as evidenced by either a total
patients with psychotic symptoms previously refractory to           score of at least 45 on the BPRS (on which each of 18 items is
                                                                    scored from 1 to 7) or a rating of moderately ill (4 or more) on at
clozapine monotherapy received open-label clozapine
                                                                    least two of the four BPRS positive symptom items (hallucinatory
augmented with risperidone for 4 weeks. By the end of the           behavior, conceptual disorganization, unusual thought content,
trial, 10 of the patients exhibited a 20% or greater reduc-         and suspiciousness).
tion in BPRS total scores. In the second study (21), a 12-
                                                                    Study Procedures
week trial, seven of 13 patients with symptoms refractory
to clozapine demonstrated a 20% or greater reduction in                All patients were fully informed about the benefits, risks, and
                                                                    potential adverse effects involved in participating in this study
total scores on the Positive and Negative Syndrome Scale,
                                                                    and signed an informed consent document. They continued to be
and four were rated as “much improved” on the Clinical              treated by their primary psychiatrist, who also signed the in-
Global Impression (CGI) improvement scale. In contrast,             formed consent. Patients remained in their current living ar-
the results of the third trial showed that none of the 12 pa-       rangements without any study-related modifications to their
tients whose symptoms were refractory to clozapine ex-              daily routines beyond regularly scheduled clinical rating sessions.
                                                                       A 4-week baseline evaluation and clozapine run-in phase was
hibited a response (20% reduction in the Positive and Neg-
                                                                    followed by 12 weeks of placebo-controlled augmentation with
ative Syndrome Scale total score) after they had received 4         risperidone. To initiate the baseline observation period, all pa-
weeks of combined treatment with clozapine and risperi-             tients had to have remained on a stable dose of clozapine for at
done (22).                                                          least 4 weeks. Baseline doses of clozapine were established by
   Although augmentation strategies are poorly docu-                treating psychiatrists and remained stable throughout the study.
                                                                    After the run-in period, patients were randomly assigned in a 1:1
mented, they are fairly common in clinical practice, with           ratio to augmentation with risperidone or matching placebo. Ris-
perhaps one-sixth of patients with psychosis in the United          peridone was started at 1 mg/day, with planned increases to 1 or
States receiving more than one antipsychotic drug (23, 24).         2 mg/day on day 4, to 2 or 3 mg/day on day 8, to 4 mg/day on day
Understandably, this situation has caused some concern              21, and to 6 mg/day on day 22. Each patient’s study medication
(25–28). In addition to their poorly documented efficacy,           dose was managed by a nonblinded research fellow not involved
                                                                    in any aspect of patient care who acted as intermediary between
polypharmacy regimens pose certain risks, including                 the study investigators, treating psychiatrist, and pharmacy. The
pharmacokinetic interactions, increased risk of adverse             raters, treating psychiatrist, and patient remained blinded
events, and, at least theoretically, loss of “atypicality” re-      throughout the study. At each weekly clinic visit, the treating psy-

Am J Psychiatry 162:1, January 2005                                                       131

TABLE 1. Demographic and Clinical Characteristics of 40                with placebo and clozapine with risperidone. The a priori
Pa tien ts With Refra ctory Sc hizophrenia R andomly                   planned comparisons used between-group repeated measures
Assigned to 12 Weeks of Double-Blind Treatment With                    analysis of variance (ANOVA) (SPSS 9.0 for Windows, general lin-
Clozapine Augmented With Either Risperidone or Placebo                 ear model, SPSS, Inc., Chicago) with group (the two treatment
                                          Treatment Group              groups) and time (baseline, week 6, and week 12) as main effects.
                                   Clozapine/                          The ANOVA model also included a group-by-time interaction
                                  Risperidone      Clozapine/Placebo   term. The specific two-group efficacy comparisons across time
Characteristic                       (N=20)             (N=20)         were BPRS total score, BPRS positive symptom score (sum of
                                  N         %         N         %      BPRS hallucinatory behavior, conceptual disorganization, un-
Sexa                                                                   usual thought content, and suspiciousness), and negative symp-
  Male                           19        95          16       80     tom score (sum of global SANS ratings).
  Female                          1         5           4       20        Between-group differences on the safety evaluation were de-
                                                                       termined by using a repeated-measures ANOVA model when ap-
                                Mean        SD        Mean      SD     propriate; otherwise, the unmatched t test and chi-square com-
                                                                       parisons were performed. Exploratory analyses were performed
Age (years)                      40.8       6.9        39.9     10.8   by using an analysis of covariance model on the efficacy and
Initial clozapine dose                                                 safety endpoints to examine the influence of age, duration of ill-
  (mg/day)b                     528.8     166.7       402.5    102.9
                                                                       ness, and dose and duration of clozapine therapy used through-
Age at symptom onset
  (years)                        18.4        3.1       17.7      5.3   out the study.
Age at first hospitalization                                              All significance tests were performed by using two-tailed prob-
  (years)                        20.4        3.8       19.7      5.3   abilities with an alpha level of 0.05.
Duration of illness (years)      21.8        7.0       22.4     11.6
Brief Psychiatric Rating
  Scale score                                                          Results
  Total                          48.8       9.2        47.1     13.3
  Positive symptom                                                       Forty patients with persistent psychotic symptoms en-
    subscale                     15.4       4.1        15.7      5.6   tered the trial and were randomly assigned in equal num-
CGI improvement rating            5.2       1.1         5.2      1.7   bers to receive clozapine combined with either risperi-
SANS score                       68.4      27.5        71.5     30.9
Simpson-Angus Rating                                                   done or placebo. Background characteristics of the two
  Scale score                     0.15      0.7         0.75     1.3   groups were similar (Table 1). The initial clozapine dose
a Significant   between-group difference (χ2 =23.4, df=1, p<0.005).    was significantly higher and the initial SANS score some-
b Significant   between-group difference (t=2.79 df=38, p<0.005).      what lower, although not reaching significance, in the clo-
                                                                       zapine/risperidone group than the clozapine/placebo
chiatrist could instruct the research fellow to either continue or     group. In general, all patients had severe refractory symp-
increase the current dose of blinded study medication. Patients        toms that had affected their lives for an average of more
judged by their treating psychiatrist to be unable to tolerate the
dose escalation schedule because of adverse effects (other than
                                                                       than 20 years. They had not responded to at least two and
clinical symptoms of schizophrenia) were maintained at their           in some cases six or more different antipsychotic drugs.
maximum tolerated dose for the remainder of the study.                 Nine had not responded to a trial of risperidone. More-
                                                                       over, as required for study entry, they had continued to
Patient Assessments
                                                                       manifest substantial psychotic symptoms despite having
   Biweekly systematic evaluations of psychopathology and ad-          received optimal treatment with clozapine as monother-
verse events were initiated during the 4-week baseline observa-
tion period and continued throughout the 12-week randomized
                                                                       apy, and their BPRS, CGI, and SANS scores indicated con-
treatment phase. These assessments were performed primarily            siderable psychopathology. The duration of prior clozap-
by two senior research fellows, blinded to treatment assignment        ine treatment ranged from 30 to 584 weeks (mean=396.9,
and not directly involved in patient care. Raters evaluated the        SD=174.4), and two patients had received doses as high as
same patients throughout the study period; however, a third            900 mg/day.
“back-up” rater was available for times of planned vacations and
illness. All three raters took part in regularly scheduled (at least     All patients completed 12 weeks of treatment in their
monthly) practice ratings with other patients or videotaped inter-     randomly assigned group. Mean doses of risperidone were
views to maintain consistency and to identify any “idiosyncratic”      4.1 mg/day (SD=1.4) at week 6 and 4.43 mg/day (SD=1.5)
approaches that developed over time.                                   at week 12. Only eight patients received the maximum 6
   Efficacy evaluations included the BPRS (29), CGI (30), and          mg/day dose of risperidone. Of those, two experienced
Scale for the Assessment of Negative Symptoms (SANS) (31).
Movement disorders were assessed at 2-week intervals with the
                                                                       acute akathisia and required dose reduction.
Simpson-Angus Rating Scale (32). A full biochemistry test panel,
urinalysis, and hematologic studies were obtained before ran-
domization and at the end of the study. Weekly safety and tolera-        By the end of the 12-week treatment period, a treatment
bility evaluations included white blood cell counts, temperature,      response (20% or greater reduction in BPRS total score)
blood pressure, heart rate, and respiration as well as subjective      was achieved by seven (35%) of the 20 patients in the cloz-
                                                                       apine/risperidone group and by two (10%) of the 20 pa-
Statistical Analysis                                                   tients in the clozapine/placebo group (χ 2 =25, df=1,
   The primary objective was to test the hypothesis of differential    p<0.01). As seen in Figure 1, the BPRS total scores de-
clinical efficacy between the two treatment groups: clozapine          creased significantly from baseline to week 12 in both

132                                                      Am J Psychiatry 162:1, January 2005
                                                                                                                                                     JOSIASSEN, JOSEPH, KOHEGYI, ET AL.

FIGURE 1. BPRS Total Scores Over the Study Period for 40                          FIGURE 2. BPRS Positive Symptom Subscale Scores Over
Pa tien ts With Refra ctory Sc hizophrenia R andomly                              the Study Period for 40 Patients With Refractory Schizo-
Assigned to 12 Weeks of Double-Blind Treatment With                               phrenia Randomly Assigned to 12 Weeks of Double-Blind
Clozapine Augmented With Either Risperidone or Placebo                            Treatment With Clozapine Augmented With Either Risperi-
                                                                                  done or Placebo

                                                                                  Mean Score on BPRS Positive Symptoms Subscale
                                                 Clozapine/risperidone (N=20)
                                                 Clozapine/placebo (N=20)                                                         15.5
Mean Total Score on BPRS

                                                                                                                                  13.0                                          a
                                                                                                                                  12.5       Clozapine/risperidone (N=20)
                           42                                                                                                                Clozapine/placebo (N=20)
                                0               6                    12                                                           12.0

                                    Time From Baseline (weeks)                                                                           0                       6             12
                                                                                                                                                  Time From Baseline (weeks)
a         Significantly different from the score at 12 weeks for clozapine/pla-
          cebo treatment per ANCOVA with baseline BPRS total score as the         a                  Significantly greater score reduction at 6 and 12 weeks relative to
          covariate (F=3.15, df=2, 74, p<0.05).                                                      clozapine/placebo treatment per ANOVA (group-by-time interac-
                                                                                                     tion: F=3.18, df=2, 76, p<0.05).

treatment groups (main effect for time: F=7.8, df=2, 76,
p<0.0001), with a significant group-by-time interaction re-                       Simpson-Angus Rating Scale scores below 1.0, indicating
flecting a greater score reduction with clozapine/risperi-                        minimal severity of movement disorders. The Simpson-
done treatment (F=3.73, df=2, 76, p<0.04). Score reductions                       Angus Rating Scale scores were lower with clozapine/
from baseline to week 6 were similar in the two groups but                        risperidone treatment throughout the trial, although they
were greater with clozapine/risperidone treatment from                            increased to approach those of clozapine/placebo treat-
week 6 to week 12. To determine whether the initial base-                         ment by week 12 (Figure 4). The frequency of other adverse
line differences contributed to the between-group differ-                         events such as weight gain, agranulocytosis, and seizures
ence, an analysis of covariance was performed using the                           did not differ between groups.
baseline BPRS total score as the covariate. With the effects                         Plasma clozapine levels did not increase significantly
of the baseline differences removed (main covariate be-                           during augmentation treatment, and no changes were ob-
tween patients effect: F=258.8, df=1, 37, p<0.0001), the                          served in white blood cell counts. Absolute neutrophil
main effect of time was no longer significant (F=0.338, df=                       counts were comparable in the two groups at baseline but
2, 74, p>0.05). However, the between-group difference at                          significantly higher with clozapine/risperidone treatment
                                                                                  at both treatment time points.
endpoint remained significant (Figure 1).
   As seen in Figure 2, scores on the BPRS positive symp-
tom subscale decreased significantly from baseline to week                        Discussion
12 in both groups (main effect for time: F=8.3, df=2, 76,                           The efficacy of clozapine augmented with risperidone
p<0.001), but the group-by-time interaction reflected a sig-                      was superior to that of clozapine combined with placebo
nificantly greater score reduction with clozapine/risperi-                        in this randomized, double-blind trial in 40 patients with
done treatment than with clozapine/placebo (Figure 2).                            schizophrenia unresponsive to clozapine monotherapy.
Negative symptoms (SANS scores) decreased significantly                           The beneficial effect of clozapine/risperidone treatment
from baseline to week 12 with clozapine/risperidone treat-                        was observed in the mean BPRS total scores and even
ment (main effect for time: F=4.46, df=2, 76, p<0.04), and                        more prominently in the BPRS positive symptom subscale
the between-group difference was significant (Figure 3).                          scores. In addition, negative symptom scores on the SANS
                                                                                  improved significantly with clozapine/risperidone treat-
Safety and Tolerability                                                           ment relative to clozapine/placebo treatment. Except for
 In general, double-blind augmentation treatment was                              mild akathisia, which was observed in two patients and
well tolerated. Both treatment groups had mean initial                            corrected through dose adjustment, risperidone augmen-

Am J Psychiatry 162:1, January 2005                                                                                                        133

FIGURE 3. SANS Scores Over the Study Period for 40                          FIGURE 4. Simpson-Angus Scale Scores Over the Study Period
Pa tien ts With Refra ctory Sc hizophrenia R andomly                        for 40 Patients With Refractory Schizophrenia Randomly
Assigned to 12 Weeks of Double-Blind Treatment With                         Assigned to 12 Weeks of Double-Blind Treatment With Cloza-
Clozapine Augmented With Either Risperidone or Placebo                      pine Augmented With Either Risperidone or Placebo

                  80                                                                                            1.0
                                                                                                                          Clozapine/risperidone (N=20)
                                                                                                                          Clozapine/placebo (N=20)

                                                                            Mean Score on Simpson-Angus Scale

Mean SANS Score



                  60                            a

                           Clozapine/risperidone (N=20)
                           Clozapine/placebo (N=20)
                  50                                                                                            0.0
                       0                       6               12                                                     0                      6                 12
                                Time From Baseline (weeks)                                                                    Time From Baseline (weeks)
a     Significantly greater score reduction at 6 and 12 weeks relative to
      clozapine/placebo treatment per ANOVA (main effect for group: F=      and others must be kept in mind. It is uncertain whether
      4.08, df=2, 76, p<0.05).
                                                                            our study is sufficiently powered to identify treatment ef-
                                                                            fects of as yet uncertain magnitude. In one cautionary re-
tation was not associated with any additional adverse ef-                   port, Stern et al. (33) reviewed 13 published trials of antip-
fects and was well tolerated.                                               sychotic augmentation and reported that the likelihood of
   To date, only a few controlled trials have evaluated cloz-               finding a positive result by chance was 63%. These authors
apine augmented with an antipsychotic agent of any type,                    concluded that augmentation trials require 40 to 100 pa-
and to our knowledge, this is the first randomized con-                     tients, depending on the number and variability of out-
trolled trial involving clozapine augmented with an atypi-                  come measures and effect size, and our investigation is at
cal antipsychotic. Previous trials of this combination in-                  the lowest range of acceptable limits. Our study avoided
clude case reports and three open-label, uncontrolled                       the pitfall of evaluating too many treatment outcomes in
prospective studies that used standardized rating criteria,                 relation to the sample size, which could lead to chance
all involving an approximate total of 50 patients. Inconsis-                findings of statistically significant differences. Neverthe-
tencies in the results of these earlier trials may be attrib-               less, our findings must be regarded as preliminary. Al-
uted to several factors beyond the tendency of small, un-                   though random assignment and blinding are important
controlled studies to produce unconfirmed results. The                      strengths of our study, definitive conclusions about the ef-
studies differed in the patients’ baseline severity of schizo-              ficacy of clozapine combined with risperidone must await
phrenia (two studies focused on outpatients and one on                      future studies with larger sample sizes.
inpatients), prior exposure to risperidone monotherapy (a                      A second important issue relates to the cause of the im-
requirement in one study, undocumented in two studies),                     provement we observed. If a patient is observed to im-
duration of augmentation treatment (4 weeks versus 12                       prove after risperidone is added to his or her clozapine
weeks), and measures of treatment outcome (BPRS in one                      therapy, it cannot be assumed that the effect of clozapine
study, Positive and Negative Syndrome Scale in two stud-                    is augmented by risperidone. A substantial improvement
ies). Taken together, the results of these studies suggest                  occurred over time in both treatment groups, possibly due
that perhaps half of patients whose symptoms are refrac-                    to patients’ participation in the study and the benefits of
tory to clozapine respond to risperidone augmentation.                      weekly clinic visits. On the other hand, this improvement
Our finding of a significant treatment-by-time interaction                  could have been an artifact of the fluctuating course of
suggests that studies based on a short (i.e., 4-week) period                schizophrenia and the tendency of extreme observations
of observation are likely to produce unreliable results.                    to regress toward the mean over time. However, our pa-
   Although the double-blind design of our investigation is                 tients did seem stable in their lack of clozapine responsiv-
stronger than that of the others, several methodologic is-                  ity, as evidenced by the fact that all had received optimal
sues raised by Freudenreich and Goff (8), Stern et al. (33),                clozapine monotherapy for a considerable time (mean=

134                                                                                        Am J Psychiatry 162:1, January 2005
                                                                                           JOSIASSEN, JOSEPH, KOHEGYI, ET AL.

396.9 weeks, SD=174.4). The improvement associated with          studied in prospective trials. Additional larger controlled
risperidone was significantly greater than that seen with        trials are required before firm clinical recommendations
placebo, but the observed treatment effect could be due to       can be made about clozapine augmentation strategies.
risperidone alone rather than to the combination. This           Moreover, in an age in which neurocognitive and func-
cannot be completely ruled out with our patient sample;          tional state measures are becoming central to the discus-
however, nine of the 20 patients in the risperidone aug-         sion of clinical efficacy and effectiveness, a comprehen-
mentation group were unresponsive to prior risperidone           sive assessment that extends beyond psychopathology
monotherapy, and of these nine unresponsive patients,            and side effects would be a welcome addition to the exist-
four responded to the combination of clozapine and ris-          ing literature.
peridone. The improvement associated with risperidone
could also have been due to the fact that this randomly as-        Presented in part at the 42nd annual meeting of the New Clinical
                                                                 Drug Evaluation Unit, Boca Raton, Fla., June 10–13, 2002; and at the
signed group entered the study receiving a higher dose of        ninth International Congress on Schizophrenia Research, Colorado
clozapine (528.8 mg versus 402.5 mg). These higher doses         Springs, Colo., March 29–April 2, 2003. Received Sept. 5, 2003;
of clozapine, acting throughout the course of the study,         revision received Jan. 13, 2004; accepted Feb. 4, 2004. From the
                                                                 Arthur P. Noyes Research Foundation; the Department of Psychiatry,
might account for the significant pattern of effects, al-        Drexel University College of Medicine, Philadelphia; the Department
though the group mean duration of clozapine treatment            of Psychiatry, University of Pennsylvania, Philadelphia; and the
was more than 7 years and raises the question of what            Department of Psychiatry, Norristown State Hospital, Norristown, Pa.
                                                                 Address correspondence and reprint requests to Dr. Josiassen, The
might cause a treatment effect after such an extensive           Arthur P. Noyes Research Foundation, Norristown State Hospital,
treatment duration.                                              1001 Sterigere St., Norristown, PA 19401; richardjosiassen@
   The risks of antipsychotic polypharmacy have not been           This study was funded by Johnson & Johnson Pharmaceutical Re-
well studied, and information about long-term risks is           search & Development.
particularly sparse. Isolated case reports of agranulocyto-
sis associated with the combination of clozapine and ris-
peridone have been reported (34), as well as a pharmaco-         References
kinetic interaction between the two agents (35). In one            1. Kane J, Honigfeld G, Singer J, Meltzer H (Clozaril Collaborative
study, 20 patients who received clozapine augmented with              Study Group): Clozapine for the treatment-resistant schizo-
risperidone had modestly higher serum prolactin levels                phrenic: a double-blind comparison with chlorpromazine.
                                                                      Arch Gen Psychiatry 1988; 45:789–796
than did a group receiving clozapine monotherapy (36).
                                                                   2. Meltzer HY: Duration of a clozapine trial in neuroleptic resis-
  We did not observe an effect of risperidone on serum                tant schizophrenia (letter). Arch Gen Psychiatry 1989; 46:672
clozapine levels, nor was such an effect seen in the two           3. Rosenheck R, Cramer J, Xu W, Thomas J, Henderson W, Frisman
open-label studies in which clozapine levels were mea-                L, Fye C, Charney D: A comparison of clozapine and haloperi-
                                                                      dol in hospitalized patients with refractory schizophrenia. N
sured (20, 22). A theoretical concern is that the addition of
                                                                      Engl J Med 1997; 337:809–815
a potent dopamine D2 receptor blocker to an atypical
                                                                   4. Simpson GM, Josiassen RC, Stanilla JK, de Leon J, Nair C, Abra-
agent may jeopardize the atypicality of the agent, increas-           ham G, Odom-White A, Turner RM: A double-blind study of
ing the risk of movement disorders and tardive dyskinesia.            clozapine dose response in chronic schizophrenia. Am J Psychi-
This phenomenon has been observed, although infre-                    atry 1999; 156:1744–1750; correction, 2001; 158:834
quently, with the combination of an atypical and a con-            5. Wahlbeck K, Cheine M, Essali A, Adams C: Evidence of cloza-
                                                                      pine’s effectiveness in schizophrenia: a systematic review and
ventional antipsychotic (8); the implications for combin-
                                                                      meta-analysis of randomized trials. Am J Psychiatry 1999; 156:
ing with atypical agents are unknown. Future studies of               990–999
antipsychotic combinations should include long-term ob-            6. Buckley P, Miller A, Olsen J, Garver D, Miller DD, Csernansky J:
servation of patients for potential toxic effects, particu-           When symptoms persist: clozapine augmentation strategies.
larly hematologic effects, movement disorders, and tar-               Schizophr Bull 2001; 27:615–628
                                                                   7. Pollack S, Lieberman JA, Fleischhacker WW, Borenstein M, Saf-
dive dyskinesia.
                                                                      ferman AZ, Hummer M, Kurz M: A comparison of European
   Our observations, although preliminary, suggest that               and American dosing regimens of schizophrenic patients on
augmentation with risperidone may benefit patients who                clozapine: efficacy and side effects. Psychopharmacol Bull
are partially responsive or nonresponsive to clozapine                1995; 31:315–320
                                                                   8. Freudenreich O, Goff DC: Antipsychotic combination therapy in
monotherapy. Our study population represents both inpa-
                                                                      schizophrenia: a review of efficacy and risks of current combi-
tients and those who have recovered sufficiently to live in           nations. Acta Psychiatr Scand 2002; 106:323–330
sheltered community settings, but it is not clear whether          9. Shiloh R, Zemishlany Z, Aizenberg D, Radwan M, Schwartz B,
our results can be equally generalized to both types of pa-           Dorfman-Etrog P, Modai I, Khaikin M, Weizman A: Sulpiride
tients. Clozapine augmentation with risperidone appears               augmentation in people with schizophrenia partially respon-
                                                                      sive to clozapine: a double-blind, placebo-controlled study. Br
to be well tolerated and safe, at least for 12 weeks of treat-
                                                                      J Psychiatry 1997; 171:569–573
ment, but clinicians should approach this augmentation
                                                                  10. Potter WZ, Ko GN, Zhang LD, Yan W: Clozapine in China: a re-
strategy with caution because polypharmacy is associated              view and preview of US/PRC collaboration. Psychopharmacol-
with potential risks that have not been systematically                ogy (Berl) 1989; 99:S87–S97

Am J Psychiatry 162:1, January 2005                                                    135

11. Mowerman S, Siris SG: Adjunctive loxapine in a clozapine-resis-            schizophrenia and other psychotic disorders. Schizophr Bull
    tant cohort of schizophrenic patients. Ann Clin Psychiatry                 2000; 26:451–457
    1996; 8:193–197                                                      24.   Weissman EM: Antipsychotic prescribing practices in the Veter-
12. Friedman J, Ault K, Powchik P: Pimozide augmentation for the               ans Healthcare Administration—New York Metropolitan re-
    treatment of schizophrenic patients who are partial respond-               gion. Schizophr Bull 2002; 28:31–42
    ers to clozapine. Biol Psychiatry 1997; 42:522–523                   25.   Marder SR: Can clinical practice guide a research agenda?
13. Gupta S, Sonnenberg SJ, Frank B: Olanzapine augmentation of                Schizophr Bull 2002; 28:127–129
    clozapine. Ann Clin Psychiatry 1998; 10:113–115                      26.   Miller AL, Craig CS: Combination antipsychotics: pros, cons,
14. Adesanya A, Pantelis C: Adjunctive risperidone treatment in                and questions. Schizophr Bull 2002; 28:105–109
    patients with “clozapine-resistant schizophrenia.” Aust NZ J         27.   Stahl SM: Antipsychotic polypharmacy, part 1: therapeutic op-
    Psychiatry 2000; 34:533–534                                                tion or dirty little secret? J Clin Psychiatry 1999; 60:425–426
15. McCarthy RH, Terkelsen KG: Risperidone augmentation of clo-
                                                                         28.   Stahl SM: Antipsychotic polypharmacy: squandering precious
    zapine. Pharmacology 1995; 28:61–63
                                                                               resources? J Clin Psychiatry 2002; 63:93–94
16. Morera AL, Barreiro P, Cano-Munoz JL: Risperidone and cloza-
                                                                         29.   Woerner MG, Mennuzza S, Kane JM: Anchoring the BPRS-A: an
    pine combination for the treatment of refractory schizophre-
                                                                               aid to improved reliability. Psychopharmacol Bull 1988; 24:
    nia. Acta Psychiatr Scand 1999; 99:305–307
17. Raju Kumar R, Khanna S: Clozapine-risperidone combination
                                                                         30.   Guy W (ed): ECDEU Assessment Manual for Psychopharmacol-
    in treatment-resistant schizophrenia (letter). Aust NZ J Psychia-
                                                                               ogy: Publication ADM 76-338. Washington, DC, US Department
    try 2001; 35:543
                                                                               of Health, Education, and Welfare, 1976, pp 218–222
18. Raskin S, Katz G, Zislin Z, Knobler HY, Durst R: Clozapine and
                                                                         31.   Andreasen NC: Scale for the Assessment of Negative Symptoms
    risperidone: combination/augmentation treatment of refrac-
                                                                               (SANS). Iowa City, University of Iowa, 1981
    tory schizophrenia: a preliminary observation. Acta Psychiatr
    Scand 2000; 101:334–336                                              32.   Simpson GM, Angus JWS: A rating scale for extrapyramidal side
19. Chong SA, Remington G: Clozapine augmentation: safety and                  effects. Acta Psychiatr Scand Suppl 1970; 212:11–19
    efficacy. Schizophr Bull 2000; 26:421–440                            33.   Stern RG, Schmeidler J, Davidson M: Limitations of controlled
20. Henderson DC, Goff DC: Risperidone as an adjunct to clozapine              augmentation trials in schizophrenia. Biol Psychiatry 1997; 42:
    therapy in chronic schizophrenics. J Clin Psychiatry 1996; 57:             138–143
    395–397                                                              34.   Godleski LS, Sernyak MJ: Agranulocytosis after addition of ris-
21. Taylor CG, Flynn SW, Altman S, Ehmann T, MacEwan GW, Honer                 peridone to clozapine treatment (letter). Am J Psychiatry 1996;
    WG: An open trial of risperidone augmentation of partial re-               153:735–736
    sponse to clozapine. Schizophr Res 2001; 48:155–158                  35.   Tyson SC, Devane CL, Risch SC: Pharmacokinetic interaction be-
22. De Groot IW, Heck AH, van Harten PN: Addition of risperidone               tween risperidone and clozapine (letter). Am J Psychiatry 1995;
    to clozapine therapy in chronically psychotic inpatients (letter).         152:1401–1402
    J Clin Psychiatry 2001; 62:129–130                                   36.   Henderson DC, Goff DC, Connolly CE, Borba CP, Hayden D: Ris-
23. Wang PS, West JC, Tanielian T, Pincus HA: Recent patterns and              peridone added to clozapine: impact on serum prolactin lev-
    predictors of antipsychotic medication regimens used to treat              els. J Clin Psychiatry 2001; 62:605–608

136                                                         Am J Psychiatry 162:1, January 2005