Background information of the condition:

Bipolar disorder Type I is characterized by at least one manic episode, with or without major depression. Patients are often symptomatically ill for nearly half of the time and their most frequent symptoms are depression, followed distantly by mania/hypomania, and then mixed episodes. 1 Drugs used for treatment of acute mania are: lithium, valproate (valproic acid or valproate divalproex sodium), haloperidol, carbamazepine and antipsychotics (risperidone, quetiapine, olanzapine, ziprasidone).

In clinical trials, the efficacy of anti-manic drugs is measured using the Young Mania Rating Scale (YMRS). The score ranges from 0-60.2 Efficacy is measured either as a mean change in YMRS score from baseline; or as clinical response, defined as 50% reduction from baseline; or as symptomatic remission, defined as an endpoint score < 12. The YMRS score has good reliability (r = 0.93) and validity, associated with the number of days of continued stay in hospital (r = 0.66). Limitations of this score include subjectivity, lack of assessment of depressed mood, and double rating of 4 items (irritability, speech, content and disruptive-aggressive behaviour).

Other outcome measures used in clinical trials are - 21-item Hamilton Depression Rating Scale (HAM-D-21), the Positive and Negative Syndrome Scale (PANSS), Clinical Global Impression bipolar version (CGI-BP), Brief Psychiatric rating Scale (BPRS), Global Assessment Functioning Scale (GAF), and Health Related Quality of Life Assessment SF- 36 (HRQOL).

Drug (Product Monograph)3

Category: Antipsychotic and an anti-manic drug

Mechanism of Action: Olanzapine, a thienobenzodiazepine, works across a number of receptor systems: dopamine D1, D3, D4, 5-HT2A/C, 5-HT3, 5-HT6, muscarinic M1-M5, adrenergic α1 and histamine H1.

Indications for Product Monograph: “for the acute treatment of manic or mixed episodes in bipolar I disorder, either as monotherapy or cotherapy with other agents (e.g., lithium or divalproex sodium)”.

Dose: The recommended starting dose in adults is 15mg once a day as monotherapy and 10 mg once a day as combination therapy. Daily dose (5-20 mg) should be adjusted in response to clinical assessment.

Duration: Not specified. If the patient responds to acute treatment and continues on with maintenance therapy, the dose should initially remain unchanged.

Methodology of Systematic review

Research questions:

1. In double blind randomized control trials (DBRCTs), does oral olanzapine monotherapy provide a therapeutic advantage compared to placebo or other anti-manic drugs, in treating acute manic or mixed episodes in patients with bipolar disorder type I?
2. In DBRCTs, does oral olanzapine in combination with other anti-manic drugs provide a therapeutic advantage over the comparator drugs mentioned below, in treating acute manic or mixed episodes in patients with bipolar disorder type I?
   • Olanzapine combination therapy with other drug vs. olanzapine monotherapy
   • Olanzapine combination with other drug vs. the other drug monotherapy
   • Olanzapine combination therapy with one drug vs. olanzapine combination therapy with a different drug.

Assessment principles: DBRCTs that compare olanzapine mono-therapy or in combination with other drugs to appropriate comparators, in adult patients with bipolar I disorder (manic or mixed episode) will be critically appraised. Therapeutic effect will be assessed according to the following hierarchy of health outcome measures: all-cause mortality, non-fatal serious adverse events, general health and quality of life, remission or response of manic symptoms, decrease in depression symptoms, withdrawals due to adverse events, and other adverse events such as movement disorders, cardiovascular effects, weight gain, sedation and abnormal laboratory values.

Search strategy: Databases searched include MEDLINE (1966 to February 2007), EMBASE (1988-February 2007), and the Cochrane database (issue 4, 2006).

Search findings: Eight DBRCTs were identified comparing olanzapine monotherapy - two versus lithium4 ,5, two versus valproate6 ,7, one versus haloperidol8 , one versus risperidone9 and 2 versus placebo10 ,11. One combination therapy trial compared (olanzapine + lithium) versus (olanzapine + valproate) versus (lithium + placebo) versus (valproate + placebo)12.

Results:

Olanzapine vs. Lithium monotherapy

Two trials, of 4 weeks duration compared olanzapine to lithium in treatment of acute mania. The recommended duration of therapy for an acute episode is 8-12 weeks. Berk et al study4 randomized 30 patients to olanzapine at a fixed dose of 10mg/day (N = 15) or lithium 400mg b.i.d. (N = 15) and study F1D-GH-LOBV5 randomized 69 Chinese patients to olanzapine (flexible dosing 5-20mg/day) and 71 patients to lithium (600 to 1800mg/day).

In Berk et al study4, the dose and blood level of lithium were lower than those recommended in the product monograph. Mortality and serious adverse events was not reported. Quality of life scores, total withdrawals and withdrawals due to adverse events were not significantly different between treatment groups. The study demonstrated no significant difference in efficacy, as measured by the total mania rating score; CGI improvement score and BPRS score. YMRS was not reported. Adverse events were not reported.

Study F1D-GH-LOBV5 reported no deaths or serious adverse events. Quality of life score was not reported. Total withdrawals and withdrawals due to adverse events were not significantly different between treatment groups. This study showed statistically significant improvement in Clinical Global Impression Bipolar version (CGI-BP) overall severity score, favoring the olanzapine group by -0.61 points, the clinical significance of this statistically significant difference is not known. The mean change from baseline in YMRS was significantly improved in the olanzapine group by 4.5 points; response rate (>50% in YMRS score from baseline) was significantly improved in the olanzapine group with an ARR =13.7% and NNT =7. The Brief Psychiatry Rating scale (BPRS) score was significantly improved in the olanzapine group by -2.1 points.

Olanzapine treatment was associated with a higher rate of weight gain [> 7% from baseline ARI =16.2%, NNH = 8]. No difference was found regarding extra pyramidal symptoms.

Olanzapine versus valproate, haloperidol, risperidone and combination therapy

The withdrawal rates (ranging from 27% to 69%) of randomized patients in the trials comparing olanzapine monotherapy to valproate6 ,7, haloperidol8 and risperidone9, as well as the combination therapy trial (69%)12, invalidates study findings as reported. Reporting results with last observation carried forward (LOCF) does not compensate for the incomplete patient follow-up. A modified study design is required to produce scientifically valid results whereby patients are followed after withdrawal from active therapy, for the full trial period.
Significantly lower total withdrawal rates from the olanzapine treatment group (ranging from 6.7% to 66.7%) as compared to active control group (ranging from 20% to 71.4%) was found (olanzapine 213/665 (32%), comparators 257/655 (39.2%), p=0.005 ARR=6.8%, NNT = 15). No statistically significant differences were found in withdrawal due to lack of efficacy or due to adverse events. However, this outcome could not be interpreted as a benefit or harm, due to lack of reporting of additional information required for interpretation.

Olanzapine vs. placebo

In 2 DBRCTs (Tohen et al 1999, N=139, 3 weeks duration and Tohen et al 2000 N=115, 4 weeks duration), the withdrawal rates ranging from 36% to 64% of randomized patients invalidate study findings as reported. Reporting results with last observation carried forward (LOCF) does not compensate for the incomplete patient follow-up.

Significantly fewer patients withdrew from the olanzapine group (ranging from 36% to 38.6%) as compared to placebo group (ranging from 58% to 64%)10 ,11. This was mainly due to lack of efficacy in the placebo group. No statistically significant differences were found in withdrawal due to adverse events. One trial11 demonstrated that patients who stayed in trial were patients who improved. Therefore, significantly lower total withdrawal rate from the olanzapine arm can be considered as an advantage of treatment. However, absolute risk increased for several adverse events for olanzapine compared to placebo: somnolence (ARI=15.5%, NNH = 7); dry mouth (ARI=7.0%, NNH = 14); dizziness (ARI=17.1%, NNH = 6); and weight gain (ARI=10%, NNH = 10).

Critical appraisal issues:

Most studies were conducted in a tertiary center where patients were more severely ill (high YMRS score at baseline) than in most clinical settings. Duration of the acute treatment in these trials was limited to less than 4 weeks. According to guidelines and product monographs, the first 8-12 weeks of treatment should be considered “acute”. Blinding should have been assessed at the end of the study due to a significant increase in adverse events in the olanzapine group. Significant differences in withdrawal rates from treatment arms (for example placebo 65.2% compared to olanzapine 38.6%) may lead to significant bias in effect size. 10

The most important limitation of interpreting study findings is due to an intention-to-treat analysis with last observation carried forward with a high drop out rate (31%6, 67-71%7, 40%8, 67-79%9, 39-65%10 38-58%11). Withdrawals are more likely when patients’ scores are worsening13-16. This bias increases, as the study gets longer and could lead to an underestimation of the treatment effects. With unequal dropouts, bias can be larger and in any direction. In psychiatric data there is a strong time trend: the improvement in the placebo group from baseline is often greater than the difference between the treatment arms13-16.

Conclusions

Monotherapy

There is insufficient evidence that olanzapine provides a therapeutic advantage versus lithium, valproate, haloperidol, or risperidone. Olanzapine was associated with lower total withdrawal rates compared to active comparators. However, this outcome could not be interpreted as a benefit or harm, due to lack of reporting of additional information required for interpretation.

Olanzapine was associated with lower withdrawal rate compared to placebo. Additional data from one trial validates this finding as an advantage (more people were improving on olanzapine than with placebo).

Combination therapy

There is insufficient evidence that olanzapine combination therapy provides a therapeutic advantage versus valproate or lithium alone.

References:

1. Judd LL, Akiskal HS, Schettler PJ, et al. The long-term natural history of the weekly symptomatic status of bipolar I disorder. Archives of General Psychiatry June 2002; 59(6):530-537.
2. Young RC, Biggs JT, Ziegler VE, et al. A rating scale for mania: reliability, validity and sensitivity. The British Journal of Psychiatry November 1978; 133:429-435.
3. Olanzapine product monograph at eCPS (May 2007).
4. Berk M, Ichim L and Brook S. Olanzapine compared to lithium in mania: a double-blind randomized controlled trial. International clinical psychopharmacology November 1999; 14(6):339-343.
5. Study F1D-GH-LOBV in Lillytrials.com published February 2006. Link: http://www.lillytrials.com/results_files/zyprexa/zyprexa_summary_7598.pdf
6. Tohen M, Baker RW, Altshuler LL, et al. Olanzapine versus divalproex in the treatment of acute mania. The American Journal of Psychiatry June 2002; 159(6):1011-1017.
7. Zajecka JM, Weisler R, Sachs G, et al. A comparison of the efficacy, safety, and tolerability of divalproex sodium and olanzapine in the treatment of bipolar disorder. The Journal of clinical psychiatry December 2002; 63(12):1148-1155.
8. Tohen M, Goldberg JF, Gonzalez-Pinto Arrillaga AM,et al. A 12-week, double-blind comparison of olanzapine vs haloperidol in the treatment of acute mania. Archives of General Psychiatry December 2003; 60(12):1218-1226.
9. Perlis RH, Baker RW, Zarate Jr CA, et al. Olanzapine versus risperidone in the treatment of manic or mixed states in bipolar I disorder: A randomized, double-blind trial. The Journal of Clinical Psychiatry November 2006; 67(11):1747-1753.
10. Tohen M, Sanger TM, McElroy SL, et al. Olanzapine versus placebo in the treatment of acute mania. American Journal of Psychiatry 1999; 156(5):702-709.
11. Tohen M, Jacobs TG, Grundy SL, et al. Efficacy of olanzapine in acute bipolar mania: A double-blind, placebo-controlled study. Archives of General Psychiatry September 2000; 57(9):841-849.
12. Tohen M, Chengappa KN, Suppes T, et al. Efficacy of olanzapine in combination with valproate or lithium in the treatment of mania in patients partially nonresponsive to valproate or lithium monotherapy. Archives of General Psychiatry January 2002; 59(1):62-69.
13. Lavori PW. Clinical trials in psychiatry: should protocol deviation censor patient data? Neuropsychopharmacology. January 1992; 6(1):39-48.
14. Hennen J. Statistical methods for longitudinal research on bipolar disorders. Bipolar Disord. 2003 Jun; 5(3):156-68.
15. Unnebrink K, Windeler J. Intention-to-treat: methods for dealing with missing values in clinical trials of progressively deteriorating diseases. Stat Med. 2001 Dec 30; 20(24):3931-46.
16. Lane P. Handling drop-out in longitudinal clinical trials: a comparison of the LOCF and MMRM approaches. Pharmaceutical statistics 2007 Mar 9.