Академический Документы
Профессиональный Документы
Культура Документы
Journal of Clinical Psychopharmacology • Volume 37, Number 5, October 2017 www.psychopharmacology.com 595
one of the antipsychotics was clozapine, and when treatment was psychosis cluster, BPRS anergia (negative symptoms), BPRS to-
initiated simultaneously with both antipsychotics in the combi- tal score, and Scale of Functioning20 overall level of function.
nation, rather than a second antipsychotic being added owing Comparisons were made in terms of these outcomes among the
to lack of response. In summary, although there is evidence 3 groups. Because of the extreme skewness in the clinical mea-
of increased risk of adverse effects and consistent association sures data, nonparametric methods based on ranks were used for
with inpatient admission, the practice of combining antipsy- analysis. For each outcome, the 3 groups were compared at base-
chotics is broadly used and appears to be effective in a group line and at study endpoint using the Kruskal-Wallis test, followed
of patients who are sicker, but not clearly recognized as having by nonparametric pairwise comparisons comparing LAI versus
“treatment refractory” schizophrenia. OA, LAI versus CA, and OA versus CA. In addition, to account
To elucidate the long-term outcome of switching from CA to for any baseline differences among groups, “change scores” were
monotherapy, we performed an exploratory analysis of the National calculated for each outcome for each subject. These were defined
Institutes of Health data set from the Preventing Relapse in Schizo- in terms of improvement in the clinical measure between baseline
phrenia: Oral Antipsychotics Compared with Injectables: Evaluat- and study endpoint, so that a positive value indicated that the
ing Efficacy (PROACTIVE) study,18 which followed 305 patients patient improved in terms of that clinical measure, whereas a neg-
with schizophrenia and schizoaffective disorder for up to 30 months ative value indicated that the patient worsened. The Kruskal-
after randomization to LAI risperidone or any second-generation Wallis test, together with nonparametric pairwise comparisons,
oral antipsychotic (OA) chosen by their treating physician to assess was then used to compare the 3 groups in terms of improvement
relapse prevention and course of clinical outcome. The primary out- in the clinical measures over the course of the study. The χ2 test
come of the study was time to first relapse, defined as substantial was used to compare the 3 groups in terms of the percentage in
clinical deterioration, increase in level of care or psychiatric hospi- each group who experienced a relapse. The Kaplan-Meier method
talization. PROACTIVE's pragmatic elements, that is, acceptance was used to construct survival curves for time until first relapse in
of patients on LAI antipsychotics, OA, or CAs at baseline, flexibil- the 3 groups, and the log-rank test was used to perform a compar-
ity in allowing clinician's expertise in implementing the intervention ison of the groups in terms of overall “survival.” The log-rank test
(type and dose of OA, dose of LAI risperidone), inclusion of new with Bonferroni adjustment was used to perform pairwise com-
antipsychotics as they entered the market, inclusion of all patients parisons among the 3 groups in terms of time to relapse. The
in the analysis through intent to treat and survival analysis, and Kruskal-Wallis test was used to compare the 3 groups in terms
the long duration of the study (up to 30 months), allowed for an of baseline patient characteristics measured as continuous or ordi-
analysis of the patients who entered the study on CA.18 The goal nal variables (eg, age, education), and the χ2 test was used to
of our exploratory analysis was to examine the effects of medication compare the 3 groups in terms of patient characteristics measured
status (CA, OA, and LAI) at study entry on outcomes and to eval- as nominal variables (eg, race). Any characteristics that differed
uate whether that status moderated the effect of being randomized significantly between groups at baseline were considered for pos-
to second-generation OA or LAI risperidone microspheres mono- sible inclusion as confounders in the group comparisons of the
therapy. This secondary analysis study is hypothesis-generating clinical measures and time until first relapse. Unless otherwise
and has limitations because the patient assignment to these groups specified, 2-tailed tests with a significance level of 0.05 were
was not randomized or blinded. Based on clinical experience (A. used. Summary statistics for continuous and ordinal variables
F.), we expected that patients receiving CA would be the most im- are given as mean ± SD and as percentages for nominal variables.
paired clinically and those receiving LAI would be the least im- All statistical analyses were performed using SAS 9.4 (SAS Insti-
paired. Furthermore, we expected that, when treated with LAI tute, Inc., Cary, NC, 2012) and IBM SPSS Statistics, version 21
risperidone microspheres monotherapy, patients on CA would do (IBM Corp, 2012). A power calculation indicated that the observed
better than those who entered the study on antipsychotic monother- sample sizes (LAI, n = 20; OA, n = 206; CA, n = 50) would yield
apy, because nonadherence might have contributed to their clinical 80% power for detecting an effect size of 0.19 when comparing
presentation at study entry, whereas LAI patients switched to oral the 3 groups in terms of a continuous outcome (eg, any of the BPRS
monotherapy would do worse. subscales) using the Kruskal-Wallis test, or when comparing the
groups in terms of a nominal outcome (eg, relapse) using the χ2
test, with a significance level of 0.05. An effect size of 0.19 is clas-
MATERIALS AND METHODS sified as “small to medium” according to the Cohen criteria.21
This secondary analysis of the limited access data set distrib-
uted from the National Institutes of Health–supported PROAC-
TIVE study was approved by the Florida International University RESULTS
Institutional Review Board as exempt. Details about the inclusion As shown in Table 1, 50 study participants (18%) were pre-
and exclusion criteria and the methodology used to establish relapse scribed CA, 206 were prescribed OA, and 20 were prescribed
in the PROACTIVE study were described elsewhere.18 Based on the LAI, for a total of 276 included in the present analysis. Patients
antipsychotic medication prescribed at study entry, the 305 patients who were not taking an antipsychotic at study entry (13) or those
with schizophrenia and schizoaffective disorder in the PROAC- with missing medication data (16) were not included in the analy-
TIVE study were divided into the following groups: (1) LAI, sis. Risperidone was the most frequently prescribed drug in all 3
first- or second-generation LAI (n = 20); (2) OA, any single OA groups (CA, OA, and LAI) at study entry. The most frequent an-
(n = 206); and (3) CA, combination of 2 or more antipsychotics tipsychotic combinations were risperidone and quetiapine (10)
(n = 50). Consistent with the PROACTIVE study medication and risperidone and olanzapine (7). Four patients were taking 3
manual, which allowed quetiapine up to 200 mg daily to be added antipsychotics at study entry (all antipsychotic combinations are
to the randomization antipsychotic to facilitate sleep, patients who shown in Supplemental Digital Content 1, Supplementary Table 1,
were taking quetiapine up to 200 mg daily in addition to another http://links.lww.com/JCP/A462). The patient characteristics at
antipsychotic at study entry were included in the OA or LAI, baseline are summarized in Supplemental Digital Content 2,
rather than the CA group. Data were analyzed separately for each Supplementary (http://links.lww.com/JCP/A463). The only sig-
of the following clinical measures: Brief Psychiatric Rating Scale nificant differences when comparing the 3 groups of patients were
(BPRS)19 anxiety depression, BPRS activation excitement, BPRS for age at study entry (P = 0.039) and number of hospitalizations
596 www.psychopharmacology.com © 2017 Wolters Kluwer Health, Inc. All rights reserved.
TABLE 2. Improvement in Clinical Measures from Baseline to Endpoint as Measured by Change Scores
© 2017 Wolters Kluwer Health, Inc. All rights reserved. www.psychopharmacology.com 597
FIGURE 1. A comparison of time to first relapse between patients on OA, patients on LAI antipsychotic, and patients on 2 or more
antipsychotics (CA).
relapse among the groups (mean time to first relapse for people on whereas the patients in the LAI group could only have dose adjust-
CA at baseline was 409.5 days vs those on OA, 562.8 days, ments or leave treatment if LAI risperidone needed to be dis-
and those on LAI, 594 days). However, after adjusting for the continued. Therefore, the differences between groups could have
number of prior hospitalizations, the difference between the OA been confounded by these factors.
and the CA groups in terms of time to first relapse was no longer In conclusion, there was a significant difference in subjects'
significant. The pragmatic nature of PROACTIVE, which included number of prior hospitalizations at baseline, with the highest num-
patients on monotherapy, those on CA, and patients on no antipsy- ber in the group taking CAs. People on CAs at baseline relapsed
chotics at study baseline, allowed us to perform a secondary analysis considerably sooner than those on oral and those on LAI antipsy-
to create a “switch trial” of antipsychotic combinations to monother- chotics when switched to antipsychotic monotherapy, although
apy. Among CA patients, who entered the 30-month PROACTIVE this difference was not statistically significant when we adjusted
study with a history of significantly more hospitalizations, randomi- for the number of hospitalizations at baseline. Within the limita-
zation to oral or LAI monotherapy did not have a significant effect. tions described, our analysis confirms the clinical experience that
Based on our exploratory analysis, caution must be taken when being on 2 or more antipsychotics indicates greater risk of relapse
switching from antipsychotic polypharmacy to monotherapy. This and predicts earlier relapse in schizophrenia. Treatment guidance
study confirms the findings of Essock et al16 in a sample of patients for this group of patients still needs to be established.
followed for up to 30 months, and it appears to validate the clinical
decision-making of the treating physicians, although our symptom
ACKNOWLEDGMENT
rating scales did not reflect greater severity of illness in the patients
Data used in the preparation of this article were obtained
on CA. Beyond the general acceptance of the fact that patients who
from the limited access datasets distributed from the NIH-
are treated with antipsychotic combinations represent a vulnerable
supported “Preventing Relapse in Schizophrenia: Oral Anti-
group, clinical guidance for treating this group of patients is lack-
psychotics Compared with Injectables: Evaluating Efficacy”
ing. Studies are warranted to explore whether psychosocial inter-
(PROACTIVE). This is a multi-site, clinical trial of persons with
vention to help patients recognize relapse, or early introduction of
Schizophrenia comparing effectiveness of randomly assigned med-
clozapine, could be effective in breaking the cycle of relapse in this
ication treatment. The study was supported by NIMH grant
patient group. In addition, systematic investigation of combinations
#s: U01 MH070007-01, U01 MH070023, U01 MH070011, U01
of antipsychotics carefully selected to augment each other's phar-
MH070009, U01 MH070008, U01 MH070017, U01 MH070010,
macodynamic profile and avoid exacerbating adverse effects in pa-
U01 MH070016, U01 MH070012. The ClinicalTrials.gov identifier
tients who do not respond to one antipsychotic appears necessary to
is NCT00330863. This manuscript reflects the views of the authors
further validate and guide clinical practice.
and may not reflect the opinions or views of the PROACTIVE Study
Our study had substantial limitations, as expected in any sec-
Investigators or the NIH.
ondary data analysis. The groups that we compared were defined
by medication status at PROACTIVE study entry, rather than by
randomization, and the CA and LAI groups were relatively small. AUTHOR DISCLOSURE INFORMATION
Further, once patients entered the study, clinicians treating those Adriana Foster has nothing to disclose. Peter Buckley has
randomized to second-generation OAs were allowed to make received grant and research support from Ameritox, Auspex Phar-
medication changes as dictated by the patient's clinical status, maceuticals, Inc, Alkermes, Inc, Avanir Pharmaceuticals, Inc,
598 www.psychopharmacology.com © 2017 Wolters Kluwer Health, Inc. All rights reserved.
Otsuka Pharmaceuticals, and National Institutes of Mental Health 9. Galletly C, Castle D, Dark F, et al. Royal Australian and New Zealand
and is a consultant for National Institutes of Mental Health. John College of Psychiatrists clinical practice guidelines for the management of
Lauriello participated in clinical research at a site on a study schizophrenia and related disorders. Aust N Z J Psychiatry. 2016;50:
headed and paid by Columbia University, sponsored by Sunovion, 410–472.
as well as in a clinical research site for a study headed and paid 10. Sneider B, Pristed SG, Correll C, et al. Frequency and correlates of
by Florida Atlantic University, sponsored by Otsuka, and is a mem- antipsychotic polypharmacy among patients with schizophrenia in
ber of the Event Monitoring Board for Janssen (services funded Denmark: A nation-wide pharmacoepidemiological study.
through a contract with the University of Missouri), Event Monitor- Eur Neuropsychopharmacol. 2015;25:1669–1676.
ing Board for Alkermes (services funded through a contract with the 11. Bolstad A, Andreassen OA, Røssberg JI, et al. Previous hospital admissions
University of Missouri), and Advisory Panel for Sunovion, Janssen, and disease severity predict the use of antipsychotic combination treatment
Otsuka, Reckitt Benckiser, and Teva. Stephen Looney has nothing to in patients with schizophrenia. BMC Psychiatry. 2011;11:126.
disclose. Nina Schooler has received grant support from Genentech 12. Gallego JA, Bonetti J, Zhang J, et al. Prevalence and correlates of
and Otsuka; she has served on advisory boards or provided consul- antipsychotic polypharmacy: a systematic review and meta-regression of
tation to Alkermes, Allergan, Forum, Roche, and Sunovion. global and regional trends from the 1970s to 2009. Schizophr Res. 2012;
138:18–28.
13. Gallego JA, Nielsen J, De Hert M, et al. Safety and tolerability of
REFERENCES antipsychotic polypharmacy. Expert Opin Drug Saf. 2012;11:527–542.
1. Freudenreich O, Goff DC. Antipsychotic combination therapy in 14. Tsutsumi C, Uchida H, Suzuki T, et al. The evolution of antipsychotic
schizophrenia. A review of efficacy and risks of current combinations. Acta switch and polypharmacy in natural practice–a longitudinal perspective.
Psychiatr Scand. 2002;106:323–330. Schizophr Res. 2011;130:40–46.
2. Schumacher JE, Makela EH, Griffin HR. Multiple antipsychotic 15. Suzuki T, Uchida H, Watanabe K, et al. A clinical case series of switching
medication prescribing patterns. Ann Pharmacother. 2003;37: from antipsychotic polypharmacy to monotherapy with a
951–955. second-generation agent on patients with chronic schizophrenia. Prog
3. Kreyenbuhl JA, Valenstein M, McCarthy JF, et al. Long-term antipsychotic Neuropsychopharmacol Biol Psychiatry. 2004;28:361–369.
polypharmacy in the VA health system: patient characteristics and 16. Essock SM, Schooler NR, Stroup TS, et al. Effectiveness of switching from
treatment patterns. Psychiatr Serv. 2007;58:489–495. antipsychotic polypharmacy to monotherapy. Am J Psychiatry. 2011;168:
4. Lehman AF, Lieberman JA, Dixon LB, et al. Practice guideline for the 702–708.
treatment of patients with schizophrenia. Am J Psychiatry. 2004;161 17. Correll CU, Rummel-Kluge C, Corves C, et al. Antipsychotic
(suppl 2):40. combinations vs monotherapy in schizophrenia: a meta-analysis of
5. Kreyenbuhl J, Buchanan RW, Dickerson FB, et al. The schizophrenia randomized controlled trials. Schizophr Bull. 2009;35:443–457.
patient outcomes research team (PORT): updated treatment 18. Buckley PF, Schooler NR, Goff DC, et al. Comparison of SGA oral
recommendations 2009. Schizophr Bull. 2009;36:94–103. medications and a long-acting injectable SGA: The PROACTIVE study.
6. Psychosis and Schizophrenia in Adults: Prevention and Management. Schizophr Bull. 2015;41:449–459.
February 2014. Available at: https://www.nice.org.uk/guidance/cg178. 19. Overall JE, Gorham DR. The brief psychiatric rating scale. Psychol Rep.
Accessed November 26, 2016. 1962;10:799–812.
7. Moore T. Schizophrenia treatment guidelines in the United States. Clin 20. Rapaport MH, Bazzetta J, McAdams LA, et al. Validation of the scale of
Schizophr Relat Psychoses. 2011;5:40–49. functioning in older outpatients with schizophrenia. Am J Geriatr
8. Moore TA, Buchanan RW, Buckley PF, et al. The Texas Medication Psychiatry. 1996;4:218–228.
Algorithm Project antipsychotic algorithm for schizophrenia: 2006 update 21. Cohen J. Statistical Power Analysis for the Behavioral Sciences. 2nd ed.
J Clin Psychiatry. 2007;68:1751–1762. Hillsdale, NJ: Lawrence Erlbaum Associates; 1988.
© 2017 Wolters Kluwer Health, Inc. All rights reserved. www.psychopharmacology.com 599