Efficacy of olanzapine for the prophylaxis of chemotherapy‐induced nausea and vomiting: a meta‐analysis

Abstract

Aim

The aim of the present study was to evaluate the efficacy of olanzapine for the prevention of chemotherapy‐induced nausea and vomiting (CINV).

Methods

The literature was searched for randomized controlled trials (RCTs) evaluating the efficacy of olanzapine for the prophylaxis of CINV using PubMed, Embase, Central, as well as clinicaltrials.gov for unpublished studies. The endpoints of the study were the number of patients who achieved a complete response (CR; no emesis and no rescue) and no nausea in the acute, delayed and overall phases. Two authors independently selected studies, assessed the risk of bias and extracted data. The included RCTs were analysed using RevMan 5.3 provided by the Cochrane Collaboration.

Results

Ten RCTs were identified for the meta‐analysis. Compared with other antiemetic agents, olanzapine significantly improved the CR in the delayed and overall phases, but did not enhance the CR in the acute phase. For the control of CINV, olanzapine was better than and comparable with aprepitant in the acute phase and delayed phase, respectively. Compared with placebo, treatment with 5 mg and 10 mg olanzapine exhibited similar efficacy in terms of the CR in the delayed and overall phases.

Conclusions

Olanzapine is an excellent alternative for the prophylaxis of CINV. Olanzapine 5 mg per day should be recommended as the initial dose because of equivalent efficacy to a 10 mg dose but a lower potential risk of side effects. Further studies are needed to explore the optimal combination of medicines.

Tables of Links

TARGETS
G protein‐coupled receptors 2	D4 receptor http://www.guidetopharmacology.org/GRAC/ObjectDisplayForward?objectId=595
Acetylcholine receptors (muscarinic) http://www.guidetopharmacology.org/GRAC/ObjectDisplayForward?objectId=2635	5‐HT2A receptor
α1‐adrenoceptors	5‐HT2C receptor
D1 receptor	5‐HT6 receptor
D2 receptor http://www.guidetopharmacology.org/GRAC/ObjectDisplayForward?objectId=249	H1 receptor
D3 receptor	NK1 receptor http://www.guidetopharmacology.org/GRAC/ObjectDisplayForward?objectId=2534

LIGANDS
5‐hydroxytryptamine	Gemcitabine
Aprepitant http://www.guidetopharmacology.org/GRAC/LigandDisplayForward?ligandId=3726	Granisetron
Azasetron	Olanzapine
Cisplatin	Ondansetron
Dexamethasone	Palonosetron http://www.guidetopharmacology.org/GRAC/LigandDisplayForward?ligandId=2285
Diphenhydramine	Tropisetron

These Tables list key protein targets and ligands in this article that are hyperlinked to corresponding entries in http://www.guidetopharmacology.org, the common portal for data from the IUPHAR/BPS Guide to PHARMACOLOGY 1, and are permanently archived in the Concise Guide to PHARMACOLOGY 2015/16 2.

Introduction

Chemotherapy‐induced nausea and vomiting (CINV) is a common side effect of chemotherapy. The prevalence of CINV has been estimated to be as high as 70–80% without appropriate antiemetic prophylaxis 3. Patients' quality of life can be decreased by CINV, which may even result in malnutrition, electrolyte disturbance and dehydration 4. CINV can also lead to poor compliance with chemotherapy, which could severely affect the therapeutic efficacy 5. Based on the occurrence time, CINV can be divided into acute (<24 h postchemotherapy) and delayed (24–120 h postchemotherapy) events. To some degree, CINV can be controlled by prophylactic antiemetic regimens. The previous recommendation for the prevention of CINV was concurrent use of a glucocorticoid, a 5‐hydroxytryptamine‐3 (5‐HT₃) antagonist and a neurokinin‐1 (NK₁) antagonist in patients receiving highly emetogenic chemotherapy (HEC) 6. In the moderately emetogenic chemotherapy (MEC) setting, a combination of a glucocorticoid and 5‐HT₃ antagonist was recommended, with an NK₁ antagonist added to patients with additional risk factors 6, 7. However, refractory CINV, especially in the delayed phase, may still occur despite the proper use of the prophylactic medicines 8. With the use of aprepitant, ondansetron and dexamethasone, the rate of complete response (CR) in patients receiving high‐dose cisplatin during the acute phase was 89.2%, while CR during the delayed phase was only 75.4% 9. The mechanism concerning the difference between the acute and delayed phases is poorly understood, and CINV in the delayed phase remains an intractable issue to be solved.

Olanzapine, an atypical antipsychotic drug, was approved by the US Food and Drug Administration (FDA) in 1996. In the supraspinal region, olanzapine inhibits multiple neurotransmitter receptors, including dopaminergic D₁, D₂, D₃ and D₄ brain receptors; serotonergic 5‐HT_2A, 5‐HT_2C, 5‐HT₃, 5‐HT₆ receptors; catecholamine alpha₁ adrenergic receptors; acetylcholine muscarinic receptors and histamine H₁ receptors 10. Its involvement in a variety of neuromodulative mechanisms affecting 5‐HT₂, 5‐HT₃ and D₂ receptors provides a pharmacological rationale for its efficacy in the prevention of CINV 11.

Initially, some case reports indicated that olanzapine was associated with remission of CINV 12, 13. Thereafter, several phase I and phase II trials verified the efficacy of olanzapine for the prevention of CINV 14, 15. In recent years, more evidence about the potential of olanzapine to prevent CINV has accumulated in randomized controlled trials (RCTs) 16, 17, 18, 19, 20, 21, 22, 23, 24, 25. According to the present National Comprehensive Cancer Network (NCCN) guidelines, olanzapine is considered as one of the options for antiemetic therapy 7. Comparisons between olanzapine‐containing antiemetic therapies and non‐olanzapine treatments have been meta‐analysed previously 26, 27. However, in the meta‐analysis conducted by Wang et al. 27, some mistakes were made in the data extraction from two RCTs 18, 21, which may have resulted in problematic conclusions. There were also some flaws with another meta‐analysis, by Chiu et al. 26, as outlined below. First, it included a study by Navari et al. 28, in which the subjects received not only chemotherapy, but also radiotherapy. Secondly, as one of the included studies 22 lacked data on the CR for the overall phase, the data for the acute phase was assumed as the overall data, which may significantly have overestimated the CR in the overall phase. Thirdly, the included studies were not stratified based on whether olanzapine was compared with other antiemetic drugs or with placebo. Finally, data collection took place only up to June 2015, but a further three RCTs 16, 19, 24 were published in the subsequent year, providing more data about the efficacy of olanzapine in preventing CINV. Therefore, by excluding the study by Navari et al. 28 and including the three newly published RCTs 16, 19, 24, we conducted the present meta‐analysis to update the conclusions about the efficacy of olanzapine in the prophylaxis of CINV.

Methods

Search strategy

Literature searches of all publication years (up to July 2016) were carried out using PubMed, Embase and Central. The clinicaltrials.gov website was also searched for unpublished studies. Major search terms were ‘olanzapine’, ‘chemotherapy‐induced nausea and vomiting’, ‘CINV’, ‘vomiting’, ‘emesis’ and ‘nausea’.

Study selection

RCTs that evaluated the efficacy of olanzapine for CINV in a prophylaxis setting were included in the meta‐analysis. Olanzapine was compared with other active drugs or with placebo in combination with basic antiemetic regimens. The trials reported at least one of the two outcome measures: the proportion of CRs (no emesis, no rescue) and of no nausea. Preliminary screening was completed to exclude obviously irrelevant studies. Secondary screening, by reviewing titles and/or abstracts, was conducted by two independent reviewers to identify potentially relevant studies. The full texts of the remaining studies were reviewed by two authors independently to identify the final studies for meta‐analysis. When opinions differed, divergence was resolved by discussion, with the involvement of a third reviewer. If there was more than one publication from the same study, the latest publication with the most complete data was adopted for meta‐analysis.

Quality assessment and data extraction

The quality of each study was assessed by the method referred to in the Cochrane Handbook for Systematic Reviews of Interventions 29, for the aspects of random sequence generation, allocation concealment, blinding, incomplete outcome data, selective reporting and other bias. Two investigators independently reviewed all included studies and extracted relevant data with a predesigned table. Information extracted included first author, year of publication, sample size, subjects' disease characteristics and emesis/nausea distribution in experimental and control groups. When necessary, the number of events was calculated from percentile data.

Outcomes

The outcomes were the incidence rates of CR and no nausea in the acute, delayed and overall phases. CR was defined to be no emetic episodes and no rescue medication. The acute, delayed and overall phase was defined as 0–24 h, 24–120 h and 0–120 h postchemotherapy, respectively.

Statistical analysis

Meta‐analysis was performed using RevMan 5.3 software provided by the Cochrane Collaboration. We stratified the studies according to whether olanzapine was compared with other antiemetic medicines or with placebo. Between‐study heterogeneity was tested using the I ² index. When I ² > 50%, which was considered as a substantial heterogeneity, a random‐effects model was used for the meta‐analysis to solve the heterogeneity 29. If I ² < 50%, the fixed‐effects model was adopted. The results were demonstrated in the form of risk ratios (RRs) and 95% confidence intervals (CIs). A P‐value of <0.05 and the 95% CIs of RRs not crossing 1.00 were regarded as statistically significant. Sensitivity analysis was executed to determine the impact of each single study on the overall RR, by excluding individual studies, one at a time.

Results

Study selection

A total of 541 citations were identified from the literature search, of which 84 duplications were removed. In the preliminary screening, 250 apparently irrelevant records were excluded, and 155 were discarded based on title and/or abstract review. For the remaining 52 studies, the full text was retrieved, after which six were excluded for including treatment of breakthrough CINV, five because they included concurrent chemo‐radiation, eight because they were reviews, nine for containing duplicate data, seven because there were no reports of outcome measures, and a further seven for other reasons. Therefore, 10 RCTs 16, 17, 18, 19, 20, 21, 22, 23, 24, 25 were included in the meta‐analysis. A detailed flowchart of the selection process is shown in Figure 1.

Figure 1.

Flow diagram of study screening and selection processCINV, chemotherapy‐induced nausea and vomiting

Characteristics of included studies

The characteristics of the included studies are shown in Table 1. The number of patients in these studies ranged from 18 to 380. Six studies 17, 19, 22, 23, 24, 25 compared 5 mg 17, 23, 24, 25 or 10 mg 19, 22 olanzapine with placebo. These six studies differed in the drugs used in the concurrent antiemetic regimens: a corticosteroid, 5‐HT₃ receptor antagonist and NK₁ receptor antagonist in two studies 17, 19; a corticosteroid and 5‐HT₃ receptor antagonist in two studies 24, 25; a 5‐HT₃ receptor antagonist alone in one study 22; and a corticosteroid, 5‐HT₃ receptor antagonist and diphenhydramine in one study 23. Elsewhere, four trials 16, 18, 20, 21 compared 10 mg olanzapine with other antiemetic medicines, including aprepitant 16, 18, 20 and dexamethasone 21. The same concurrent drugs (a corticosteroid and a 5‐HT₃ receptor antagonist) were used in the four trials. One study 22 enrolled only nonsmall cell lung cancer patients receiving chemotherapy with cisplatin combined with gemcitabine. Subjects in six studies 16, 18, 19, 20, 22, 25 received only HEC, while patients in the four other studies 17, 21, 23, 24 underwent MEC or HEC. Of the four studies involving both HEC and MEC, only two studies 21, 24 reported the results of HEC and MEC separately, and two studies were not enough for subgroup analysis. All 10 trials reported a CR to antiemetic therapy and defined CR as no vomiting and no rescue, while only six studies reported the rate of no nausea. The CR was evaluated in the acute, delayed and overall phases in seven studies 16, 17, 18, 19, 20, 21, 25. One study 23 provided the CR only in the overall phase, and two studies 22, 24 lacked overall data. One study 20 defined the delayed phase as 24–96 h postchemotherapy, in contrast to 24–120 h in the rest of the studies. The methodological quality of the included studies was generally high; but seven studies 16, 18, 21, 22, 23, 24, 25 had high performance bias and eight 16, 17, 18, 20, 21, 22, 23, 24 had unclear detection bias (Figure 2).

Table 1.

Characteristics of the randomized controlled trials included in the meta‐analysis

Studies	Sample size	Age (years)	Intervention	Comparison	Cancer types	Chemotherapy
Tan et al. 2009 21	121/108	Average 51.7/average 52.5	OLA 10 mg qd p.o. days 1–5Azasetron 10 mg qd i.v. day 1DEX 10 mg qd i.v. day 1	Azasetron 10 mg qd i.v. day 1DEX 10 mg qd i.v. days 1–5	Lung: 32%/14%Breast: 19%/29%Colorectal: 13%/13%Lymphoma: 10%/9%Stomach: 7%/11%Others: 18%/24%	HEC or MEC
Navari et al., 2011 18	121/120	Median 63, range 39–77/median 61, range 42–81	OLA 10 mg qd p.o. days 1–4Palonosetron 0.25 mg qd i.v. day 1DEX 20 mg qd i.v. day 1	APR 125 mg qd p.o. day 1APR 80 mg qd p.o. days 2–3Palonosetron 0.25 mg qd i.v. day 1DEX 12 mg qd i.v. day 1DEX 4 mg bid p.o. days 2–4	Breast: 50%/55%Lung (nonsmall cell): 35%/33%Malignant lymphoma: 10%/8%Bladder: 7%/3%	HEC
Lv et al., 2013 23	30/30	Median 45, range 31–72	OLA 2.5 mg bid p.o. or 5 mg qn p.o. day 1Diphenhydramine 20 mg i.m. day 1DEX 5–10 mg i.v. day 1Tropisetron 5 mg i.v. day 1	Diphenhydramine 20 mg i.m. day 1DEX 5–10 mg i.v. day 1Tropisetron 5 mg i.v. day 1	Stomach: 30%Breast: 28%Colorectal: 25%Lung: 16%	HEC or MEC
Mizukami et al., 2014 17	22/22	Median 63, range 22–78/median 55, range 33–75	OLA 5 mg qd p.o. days 0–5DEX 9.9 mg i.v. day 1DEX 6.6 mg i.v. days 2–4APR 125 mg p.o. day 1APR 80 mg p.o. days 2–35‐HT3 receptor antagonist	Placebo p.o. days 0–5DEX 9.9 mg i.v. day 1DEX 6.6 mg i.v. days 2–4APR 125 mg p.o. day 1APR 80 mg p.o. days 2–35‐HT3 receptor antagonist	Bladder: 27%/14%Lymphoma: 18%/18%Pharyngeal: 14%/18%Breast: 14%/32%Others: 27%/18%	HEC or MEC
Mukhopadhyay et al., 2015 25	50/50	N/A	OLA 5 mg p.o. days 1–5Palonosetron day 1DEX day 1	Palonosetron day 1DEX day 1	N/A	HEC
Shumway et al., 2015 20	9/10	Median 54/median 61	OLA 5 mg p.o. days −1 to −2OLA 10 mg p.o. days 1–4DEX 12 mg i.v. day 1DEX 4 mg bid p.o. days 2–4Palonosetron 0.25 mg i.v. day 1	Placebo p.o. days −1 to −2 and day 4APR 125 mg p.o. day 1APR 80 mg p.o. days 2–3DEX 12 mg i.v. day 1DEX 4 mg bid p.o. days 2–4Palonosetron 0.25 mg i.v. day 1	Breast: 44%/50%Lung: 22%/40%Hodgkin's: 11%/10%Others: 22%/0%	HEC
Wang et al., 2015 22	42/42	Median59, range 39–74/median 60, range 40–76	OLA 10 mg qd p.o. days 1–8 Ondansetron 8 mg i.v. day 1	Ondansetron 8 mg i.v. day 1	Nonsmall cell lung cancer	Cisplatin–gemcitabine
Babu et al., 2016 16	50/50	Average 43.3/average 44.7	OLA 10 mg p.o. day 1OLA 5 mg bid p.o. days 2–4Palonosetron 0.25 mg i.v. day 1DEX 20 mg i.v. day 1DEX 4 mg bid p.o. days 2–4	APR 125 mg p.o. day 1APR 80 mg p.o. days 2–3Palonosetron 0.25 mg i.v. day 1DEX 12 mg i.v. day 1DEX 4 mg bid p.o. days 2–4	Breast: 52%/50%Lymphoma: 14%/18%Head and neck: 20%/18%Osteosarcoma: 10%/10%Stomach: 4%/4%	HEC
Meng et al., 2016 24	60/60	<60 years vs. ≥60 years: 27:33/29:31	OLA 2.5 mg bid days 1–5Granisetron 3 mg qd day 1DEX 5 mg qd i.v. day 1	Granisetron 3 mg qd day 1DEX 5 mg qd i.v. day 1	Solid tumour except breast cancer	HEC or MEC
Navari et al., 2016 19	192/188	Median 58, range 29–86/median 56, range 28–89	OLA 10 mg qd days 1–45‐HT3 receptor antagonist day 1DEX 12 mg p.o. day 1DEX 8 mg p.o. days 2–4NK1 receptor antagonist	Placebo days 1–45‐HT3 receptor antagonist day 1DEX 12 mg p.o. day 1DEX 8 mg p.o. days 2–4NK1 receptor antagonist	Breast: 63%/65%Lung: 14%/12%Others: 23%/23%	HEC

5‐HT₃, 5‐hydroxytryptamine‐3; APR, aprepitant; bid, twice a day; DEX, dexamethasone; HEC, highly emetogenic chemotherapy; i.m., intramuscular; i.v., intravenous; MEC, moderately emetogenic chemotherapy; N/A, not available; NK1, neurokinin‐1; OLA, olanzapine; p.o., orally; qd, once a day; qn, every night.

Figure 2.

Risk of bias summary for included studies

Efficacy evaluation – CR

Compared with active agents 16, 18, 20, 21, olanzapine significantly improved CR in the delayed (RR = 1.12, 95% CI = 1.02, 1.22, P = 0.02) (Figure 3B) and overall (RR = 1.12, 95% CI = 1.02, 1.23, P = 0.02) (Figure 3C) phases, but did not enhance CR in the acute (RR = 1.05, 95% CI = 0.96, 1.15, P = 0.32) (Figure 3A) phase. After excluding the study by Tan et al. 21, olanzapine was compared with aprepitant 16, 18, 20. The incidence of CR significantly increased in the olanzapine‐containing group in the acute phase (RR = 1.10, 95% CI = 1.02, 1.18, P = 0.02). However, there was no significant difference in CR between olanzapine and aprepitant in the delayed (RR = 1.04, 95% CI = 0.93, 1.16, P = 0.45) and overall (RR = 1.04, 95% CI = 0.93, 1.18, P = 0.49) phases.

Figure 3.

Complete response with olanzapine compared with active agents: (A) acute phase; (B) delayed phase; (C) overall phase. CI, confidence interval; M‐H, Mantel‐Haenszel

Compared with placebo, 5 mg olanzapine treatment achieved a statistically superior CR in the delayed (RR = 1.73, 95% CI = 1.17, 2.55, P = 0.006) (Figure 5) and overall (RR = 1.87, 95% CI = 1.32, 2.64, P = 0.0005) (Figure 6) phases, but not in the acute phase (RR = 1.12, 95% CI = 0.94, 1.34, P = 0.21) (Figure 4). As the dose increased to 10 mg, the significant advantage of olanzapine was shown in the acute (RR = 1.26, 95% CI = 1.10, 1.44, P = 0.001) (Figure 4), delayed (RR = 1.22, 95% CI = 1.08, 1.38, P = 0.001) (Figure 5) and overall (RR = 1.57, 95% CI = 1.26, 1.94, P < 0.0001) (Figure 6) phases.

Figure 4.

Complete response with olanzapine compared with placebo in the acute phase. CI, confidence interval; M‐H, Mantel‐Haenszel

Figure 5.

Complete response with olanzapine compared with placebo in the delayed phase. CI, confidence interval; M‐H, Mantel‐Haenszel

Figure 6.

Complete response with olanzapine compared with placebo in the overall phase. CI, confidence interval; M‐H, Mantel‐Haenszel

Efficacy evaluation – no nausea

Compared with active agents 16, 18, 20, 21, olanzapine showed no superiority in antinausea effect in the acute (RR = 1.01, 95% CI = 0.96, 1.07, P = 0.66) (Figure 7a), delayed (RR = 1.45, 95% CI = 0.99, 2.14, P = 0.06) (Figure 7b) or overall (RR = 1.39, 95% CI = 0.90, 2.13, P = 0.14) (Figure 7c) phases. When olanzapine was compared with aprepitant by discarding the trial by Tan et al. 21, there was no significant difference between the experiment and control groups in the acute (RR = 0.98, 95% CI = 0.90, 1.07, P = 0.61), delayed (RR = 1.38, 95% CI = 0.78, 2.42, P = 0.27) or overall (RR = 1.24, 95% CI = 0.68, 2.27, P = 0.48) phases. As only two studies 19, 22 reported the antinausea effects of olanzapine compared with placebo, the ‘no nausea’ endpoint was not meta‐analysed; both of these studies revealed statistical superiority to standard antiemetic regimens in this endpoint.

Figure 7.

Incidence of no nausea with olanzapine compared with active agents: (A) acute phase; (B) delayed phase; (C) overall phaseCI, confidence interval; M‐H, Mantel‐Haeszel

Discussion

The present meta‐analysis investigated the efficacy of olanzapine in the prophylaxis of CINV. Compared with active agents, the rate of CR was significantly higher in the olanzapine‐treated group in the delayed and overall phases but not in the acute phase, which was consistent with the results of a previous meta‐analysis 27. However, when olanzapine was compared with dexamethasone and aprepitant separately, the results were very different. Compared with dexamethasone, olanzapine improved CR from 67.6% to 84.3% in the delayed phase but did not increase CR in the acute phase (94.2% vs. 93.5%). Compared with aprepitant, olanzapine achieved better (91.7% vs. 83.3%) and equivalent (79.4% vs. 76.1%) control of emesis in the acute and delayed phases, respectively. In terms of the antinausea effect, the meta‐analysis showed no advantage of olanzapine in any phase compared with active agents or aprepitant. There was 87% and 99% heterogeneity when comparing the antinausea effect between active agents and olanzapine in the delayed and overall phases. By excluding the study by Babu et al. 16, the heterogeneity in both phases was reduced to 0%. The differences in tumour category and the type of chemotherapy might have resulted in the heterogeneity. The trial by Navari et al. 18 provided the strongest evidence for the efficacy of olanzapine in the prophylaxis of CINV as dexamethasone was given only on day 1 in the olanzapine group in contrast to days 1–4 in the control group.

In the prophylaxis of CINV, olanzapine is more efficacious than, and equally efficacious to aprepitant in the acute and delayed phases, respectively. In addition, the cost per cycle of chemotherapy is much lower for olanzapine than for aprepitant 16, which will enable more patients, especially those in less developed countries, to afford medical cost and avail themselves of adequate treatment and benefit. Therefore, olanzapine scores over aprepitant in respect of the cost–benefit ratio.

When comparing olanzapine with placebo, either 5 mg (n = 324) 17, 23, 24, 25 or 10 mg (n = 464) 19, 22 of drug was administered. Subgroup analysis was conducted based on the dose of olanzapine. Both the 5 mg and 10 mg olanzapine groups achieved a higher rate of CR compared with the non‐olanzapine group in the delayed and overall phases. Unexpectedly, the effect size of 5 mg olanzapine was larger than that of the 10 mg dose. The difference in concomitant therapies may explain the unexpected results. A previous phase I trial explored the maximum tolerated dose of olanzapine as an antiemetic, and the upper limit was set at 5 mg and 10 mg for days −2 to −1 and days 0–7, respectively 15. As the efficacy of 5 mg olanzapine in emesis control significantly exceeded that of the placebo, and taking the possible increased risk of adverse drug reactions (ADRs) with the dose increase into consideration, olanzapine at a dose of 5 mg day^–1 should be recommended for the prophylaxis of CINV. If patients fail to achieve CR, the dose can be increased to 10 mg day^–1 in the next chemotherapy cycle.

Common ADRs of olanzapine as an antipsychotic drug include fatigue, sedation, dizziness, dry mouth, constipation, body weight gain, a rise in blood glucose, and dyslipidaemia. When this agent has been used for the prevention of CINV, sedation and dizziness were the most common treatment‐related adverse events (AEs), but these AEs did not lead to level 3 or 4 toxicity and could be tolerated by patients 16, 18, 24. In the study by Navari et al. 19, patients receiving olanzapine were more drowsy on day 2 than before chemotherapy, but this symptom was alleviated on days 3–5, in spite of continual use of olanzapine on days 3–4, suggesting that patients adapted to the sedative effect of olanzapine. Body weight, blood lipids and blood glucose did not change significantly after chemotherapy 21, 24. The severity of some AEs significantly increased over the course of a few days in some single cycles but did not increase further with subsequent cycles 18. This might be attributed to the low dose and short treatment course of olanzapine.

Mood swings are common after a cancer diagnosis and can adversely affect medical treatment, diet and sleep. It was observed that olanzapine not only enhanced the CR of CINV, but also improved the sleep and appetite of patients compared with those on the non‐olanzapine regimen 21. In another study, the food intake of patients in the control group was significantly decreased on days 5 and 6, but those in the olanzapine group maintained a similar food intake throughout the chemotherapy process 17. Ninety‐one per cent of patients in the olanzapine group wanted to continue taking this drug in the next cycle of chemotherapy 17. Another study found better chemotherapy compliance in the olanzapine group 23.

There have also been several studies evaluating the efficacy of olanzapine in the rescue of breakthrough CINV 30, 31, 32. The results showed that olanzapine was statistically and clinically better than prochlorperazine and metoclopramide in controlling breakthrough CINV, which provided the supporting evidence for the recommendation of olanzapine as an option for breakthrough CINV in the 2016 NCCN guidelines 7.

It should be noted that there were several limitations to the present meta‐analysis. First, the combinations of medicines taken by participants were not uniform, with one RCT not including dexamethasone 22, which might have contributed to the heterogeneity. Second, one of the included RCTs was available only as a conference abstract, and lacked the full details of methodology, patient population and toxicity 25. Third, subgroup analysis based on the emetogenic potential of chemotherapies was impossible because of data deficiency. Finally, the safety of olanzapine for the prophylaxis of CINV was not meta‐analysed owing to the lack of data.

Despite these limitations, the overall weight of evidence clearly demonstrates the efficacy of olanzapine for the prevention of CINV. The results of the present meta‐analysis could guide the optimal clinical application of olanzapine for this indication.

Conclusions

In view of the efficacy and safety associated with its use for the prevention of CINV, and also its higher cost‐effectiveness compared with aprepitant, olanzapine is a good choice for prophylaxis in patients receiving HEC or MEC. The use of 5 mg day^–1 olanzapine should be recommended as the initial dose for preventing CINV, rather than a 10 mg dose, because of its comparable efficacy but lower potential risk of ADRs. Further studies are needed to explore the optimal combination of medicines and assess the ADRs systematically.

Competing Interests

There are no competing interests to declare.

Yang, T. , Liu, Q. , Lu, M. , Ma, L. , Zhou, Y. , and Cui, Y. (2017) Efficacy of olanzapine for the prophylaxis of chemotherapy‐induced nausea and vomiting: a meta‐analysis. Br J Clin Pharmacol, 83: 1369–1379. doi: 10.1111/bcp.13242.