Mepolizumab versus placebo for asthma
Abstract
Background
Mepolizumab is a human monoclonal antibody against interleukin‐5 (IL‐5), the main cytokine involved in the activation of eosinophils, which in turn causes airway inflammation. Recent studies have suggested these agents may have a role in reducing exacerbations and improving health‐related quality of life (HRQoL). There are no recommendations for the use of mepolizumab in adults or children in the recent update of the BTS/SIGN guidelines (BTS/SIGN 2014).
Objectives
To compare the effects of mepolizumab with placebo on exacerbations and HRQoL in adults and children with chronic asthma.
Search methods
We searched the Cochrane Airways Group Register (CAGR) of trials, clinical trial registries, manufacturers' websites and the reference lists of included studies. Searches were conducted in November 2013 and updated in November 2014.
Selection criteria
We included randomised controlled trials comparing mepolizumab versus placebo in adults and children with asthma.
Data collection and analysis
Two authors independently extracted data and analysed outcomes using a random‐effects model. We used standard methods expected by The Cochrane Collaboration.
Main results
Eight studies on 1707 participants met the inclusion criteria. Only two studies included children (over 12 years of age), but they did not report separate findings for the adolescents. Seven studies involved intravenous mepolizumab alone; one included a subcutaneous arm. There was heterogeneity in the severity and clinical pattern of asthma among the participants in the eight studies, varying from mild to moderate atopic asthma, to persistent asthma and eosinophilic asthma with recurrent exacerbations. Selection bias was a concern in several of the studies included in this review.
Four trials compared intravenous mepolizumab to placebo in relation to HRQoL. Two studies measured scores from the Asthma Quality of Life Questionnaire (AQLQ), which showed a non‐significant difference between mepolizumab and placebo (mean difference (MD) 0.21, 95% confidence interval (CI) − 0.01 to 0.44; participants = 682), in the direction favouring mepolizumab. The third study used the St. George's Respiratory Questionnaire (SGRQ) and found a significant difference between mepolizumab and placebo (MD 6.40, 95% CI 3.15 to 9.65; participants = 576), which indicated a clinically important benefit favouring mepolizumab. A fourth study noted that there was no significant difference but did not provide any data. The two studies in people with eosinophilic asthma showed a reduction in clinically significant exacerbation rates (Risk Ratio 0.52, 95% CI 0.43 to 0.64; participants = 690). However, an analysis of four studies that were not confined to people with eosinophilic asthma indicated considerable heterogeneity and no significant difference in people with one or more exacerbations between mepolizumab and placebo using a random‐effects model (Risk Ratio 0.67, 95% CI 0.34 to 1.31; participants = 468; I2 = 59%).The analysis of serious adverse events indicated a significant difference favouring mepolizumab (Risk ratio 0.49, 95% CI 0.30 to 0.80; participants = 1441; studies = 5; I2 = 0%). It was not possible to combine the results for adverse events, and we deemed the quality of this evidence to be low.
A single study compared subcutaneous mepolizumab to placebo in 385 adults with severe eosinophilic asthma and found an improvement in HRQoL scores and a reduction in asthma exacerbations, including exacerbations requiring admission to hospital.
Authors' conclusions
It is not possible to draw firm conclusions from this review with respect to the role of mepolizumab in patients with asthma. Our confidence in the results of this review are limited by the fact that the intravenous route is not currently licensed for mepolizumab, and the evidence for the currently licenced subcutaneous route is limited to a single study in participants with severe eosinophilic asthma.
The currently available studies provide evidence that mepolizumab can lead to an improvement in health‐related quality of life scores and reduce asthma exacerbations in people with severe eosinophilic asthma.
Further research is needed to clarify which subgroups of patients with asthma could potentially benefit from this treatment. Dosage, ideal dosing regimens and duration of treatment need to be clarified, as the studies included in this review differed in their protocols. There are no studies reporting results from children, so we cannot comment on treatment for this age group. At the present time, larger studies using licenced treatment regimens are required to establish the role of mepolizumab in the treatment of severe asthma.
PICOs
Plain language summary
Mepolizumab as opposed to placebo for asthma
Review question
We considered in this review whether taking mepolizumab is better than a placebo for people with asthma.
Background
Asthma is an inflammatory lung condition characterised by the narrowing of the airways, breathlessness, a tight chest and reduced quality of life. By the year 2025, there may be up to 400 million people with asthma worldwide. Mepolizumab is one treatment that may help to reduce the symptoms.
Study characteristics
Eight studies compared mepolizumab treatment to a placebo in 1707 patients with asthma. Six studies only included adults. We summarised the results as they relate to quality of life, occurrence of asthma attacks needing hospital admission and side effects of mepolizumab.
Key results
We found that patients with severe asthma who had high levels of eosinophils (inflammatory cells in the blood stream) benefited from taking mepolizumab through improved quality of life and reduced asthma attacks. There was no benefit in terms of lung function. We have avoided making recommendations because we think that further research is needed to clarify aspects such as dosage and length of treatment as well as which patients might benefit the most.
Authors' conclusions
Summary of findings
| IV mepolizumab compared to placebo for asthma | ||||||
| Patient or population: adults with asthma of varying degrees of severity | ||||||
| Outcomes | Illustrative comparative risks* (95% CI) | Relative effect | No. of participants | Quality of the evidence | Comments | |
| Assumed risk | Corresponding risk | |||||
| Placebo | IV mepolizumab | |||||
| Change in HRQoL assessed with AQLQ. Scale from 1 to 7 (higher is better) Follow‐up: 52 weeks | The mean change in HRQoL ranged from 0.18 to 0.71 units | The mean change in HRQoL in the intervention group was 0.21 units more (0.01 fewer to 0.44 more) | ‐ | 682 | ⊕⊕⊝⊝ | Trial participants had severe eosinophilic asthma |
| Change in HRQoL assessed with SGRQ. Scale from: 0 to 100 (lower is better) Follow‐up: 32 weeks | The mean change in HRQoL was − 9.0 units | The mean change in HRQoL in the intervention group was 6.4 units lower (3.15 lower to 9.65 lower) | ‐ | 382 | ⊕⊕⊕⊝ | Trial participants had severe eosinophilic asthma |
| Rate of clinically significant exacerbations ‐ 75 mg mepolizumab versus placebo Follow‐up: range 32 to 52 weeks | The mean rate of clinically significant exacerbations on placebo was 1 per patient per yearc | The mean rate of clinically significant exacerbations in the intervention group was 0.48 less per patient per year (0.57 less to 0.46 less) | Rate ratio 0.52 (0.43 to 0.64) | 690 | ⊕⊕⊕⊝ | Trial participants had severe eosinophilic asthma |
| Rate of clinically significant exacerbations ‐ 250 mg mepolizumab versus placebo Follow‐up: 52 weeks | The mean rate of clinically significant exacerbations on placebo was 0.43 per patient per year | The mean rate of clinically significant exacerbations in the intervention group was 0.17 less per patient per year (0.08 less to 0.23 less) | Rate ratio 0.61 (0.46 to 0.81) | 307 | ⊕⊕⊕⊝ | Trial participants had severe eosinophilic asthma |
| Rate of clinically significant exacerbations ‐ 750 mg mepolizumab versus placebo Follow‐up: 52 weeks | The mean rate of clinically significant exacerbations on placebo was 0.43 per patient per year | The mean rate of clinically significant exacerbations in the intervention group was 0.22 less per patient per year (0.17 less to 0.25 less) | Rate ratio 0.48 (0.36 to 0.64) | 311 | ⊕⊕⊕⊝ | Trial participants had severe eosinophilic asthma |
| People with one or more exacerbations Follow‐up: 20 to 50 weeks | 264 per 1000 | 177 per 1000 | Risk ratio 0.67 | 467 | ⊕⊕⊝⊝ | Variety of asthma severity in the trials |
| Serious adverse events Follow‐up: 20 to 52 weeks | 82 per 1000 | 40 per 1000 | Risk ratio 0.49 | 1441 | ⊕⊕⊕⊝ | Variety of asthma severity in the trials |
| *The basis for the assumed risk is the mean control group risk across studies. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). | ||||||
| GRADE Working Group grades of evidence | ||||||
| aThe intravenous route is not currently licenced for mepolizumab; one point deducted for indirectness. | ||||||
| Subcutaneous mepolizumab compared to placebo for asthma | ||||||
| Patient or population: adults with severe eosinophilic asthma | ||||||
| Outcomes | Illustrative comparative risks* (95% CI) | Relative effect | No. of participants | Quality of the evidence | Comments | |
| Assumed risk | Corresponding risk | |||||
| Placebo | SC mepolizumab | |||||
| Change in HRQoL assessed with SGRQ. Scale from: 0 to 100 (lower is better) Follow‐up: 32 weeks | The mean HRQoL was − 9.0 units | The mean HRQoL ‐ SGRQ in the intervention group was 7 units fewer (10.19 fewer to 3.81 fewer) | ‐ | 385 | ⊕⊕⊕⊝ | |
| Rate of exacerbations requiring admission Follow‐up: 32 weeks | The mean rate of exacerbations requiring admission on placebo was 0.10 per patient per year | The mean rate of exacerbations requiring ED visit or admission in the intervention group was 0.07 less per patient per year (0.01 less to 0.09 less) | Rate ratio 0.31 | 385 | ⊕⊕⊕⊝ | |
| Rate of exacerbations requiring ED or admission Follow‐up: 32 weeks | The mean rate of exacerbations requiring ED or admission on placebo was 0.20 per patient per year | The mean rate of exacerbations requiring ED or admission in the intervention group was 0.12 less per patient per year (0.03 less to 0.16 less) | Rate ratio 0.39 | 385 | ⊕⊕⊕⊝ | |
| Rate of clinically significant exacerbations Follow‐up: 32 weeks | The mean rate of clinically significant exacerbations on placebo was 1.75 per patient per year | The mean rate of clinically significant exacerbations in the intervention group was 0.93 less per patient per year (0.65 less to 1.14 less) | Rate ratio 0.47 | 385 | ⊕⊕⊕⊝ | |
| Asthma symptoms measured on Asthma Control Questionnaire Scale from: 0 to 6 (lower is better)b Follow‐up: 32 weeks | The mean change in asthma symptoms was − 0.5 units | The mean asthma symptoms in the intervention group was 0.44 units fewer (0.64 fewer to 0.24 fewer) | ‐ | 385 | ⊕⊕⊕⊝ | |
| *The basis for the assumed risk was the event rate in the placebo arm of the single included study. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). | ||||||
| GRADE Working Group grades of evidence | ||||||
| aThis finding is from a single study so we do not know how well this will match further research; one point deducted. cThe mean difference is less than the clinical minimally important difference (0.5 units), and no responder analysis is available; one point deducted. | ||||||
Background
Description of the condition
A recent global estimate of the number of people currently suffering from asthma is in the region of 300 million, and it is expected that by 2025 the number will increase to 400 million (WHO 2007). The subsequent burden of disease is likely to continue to impose additional pressures on patients, their families and healthcare systems (Masoli 2004). The increased incidence in morbidity has been associated with suboptimal delivery of care, including under‐treatment with corticosteroids and a limited awareness of the condition amongst patients (Gibson 1993; Kandane‐Rathayake 2009).
In the USA, the number of people with asthma increased from 20 million in 2001 to 25 million in 2009 (CDC 2011). Prevalence rates are slightly higher among children (10%) than among adults (8%) (CDC 2011; CDC 2012), with considerable variation among different ethnic groups. Between 2008 and 2010, asthma prevalence rates in the USA were 14.1% among multiracial individuals, 11.2% among blacks, 9.4% among Alaska Natives, 9.4% among other Native Americans, 7.7% among whites and 5.2% among those of Asian descent (CDC 2011). Globally, the prevalence of wheezing symptoms in children varies geographically, with the UK having the highest recorded prevalence of current wheezing at 32.3% and Ethiopia the lowest at 1.7% (Patel 2008).
For many people, asthma has an important impact on quality of life (Clayton 2005) and on financial considerations (Wu 2007). In the USA, approximately 10 million people experience asthma exacerbations each year (Krishnan 2006), and in the UK, over 65,000 hospital admissions for asthma were recorded in 2005 and 2006 (NHS 2011).
In recent years, clinical guidelines have been produced for the management of asthma at national (e.g. BTS/SIGN 2014; NIH 2007) and international (GINA 2012) levels. Several risk factors for asthma have been identified, including triggers such as allergens, chemical irritants and tobacco smoke, but asthma‐related mortality and morbidity remain a major health concern (Braman 2006). On the other hand, the condition can also be controlled and health‐related quality of life (HRQoL) maintained for considerable periods (WHO 2011).
Description of the intervention
One of the core pathological features of asthma is considered to be eosinophilic infiltration of the bronchial mucosa, which triggers an inflammatory response. Mepolizumab is a humanised monoclonal antibody against interleukin‐5 (IL‐5) that has been shown to inhibit eosinophilic airway inflammation. A number of studies have been conducted in young adults (> 12 years old) and adults with recurrent severe asthma exacerbations and signs of eosinophilic inflammation (Haldar 2009; Nair 2009; Pavord 2012). The results of these studies suggest that inhibiting eosinophilic inflammation by monoclonal antibodies may be associated with a reduced risk of acute exacerbations of asthma and a reduction in eosinophil count.
How the intervention might work
Proteins secreted by eosinophils cause damage to the epithelium, initiating vasodilatation, smooth muscle contraction and increased mucous secretion, which in turn is associated with increased airway hyperresponsiveness, asthma symptoms and airway narrowing (Liu 2013).
Mepolizumab is a key monoclonal antibody inhibiting IL‐5, which is the main cytokine involved in eosinophil activation and recruitment. This intervention might work by preventing the initiation of the inflammatory response. Mepolizumab is administered intravenously as either a one‐off dose of 2.5 to 10 mg/kg or monthly doses of 75 mg, 250 mg or 750 mg given for a period ranging from 16 to 52 weeks. Mepolizumab can also be given subcutaneously.
Why it is important to do this review
In a recently published meta‐analysis of seven randomised placebo‐controlled trials on 1131 adults, mepolizumab was shown to reduce the risk of exacerbations and improve quality of life in people with eosinophilic asthma, but did not lead to a significant improvement in lung function (Liu 2013).
It is important to do this review so that the evidence presented and the judgements made in Liu 2013 are available and placed in context within The Cochrane Library. Our review will also set the stage for future updates as more evidence becomes available.
Objectives
To compare the effects of mepolizumab with placebo on exacerbations and HRQoL in adults and children with chronic asthma.
Methods
Criteria for considering studies for this review
Types of studies
We included randomised controlled trials (RCTs). We included studies reported as full text, those published as abstracts only and unpublished data. Included trials were a minimum of 16 weeks in duration.
Types of participants
We included both adults and children with a diagnosis of asthma. We focused on collating data from people who have been reported as having eosinophilic asthma to analyse these individuals as a subgroup. We examined individual articles in order to determine how this group should be defined.
Individuals with congential heart disease and respiratory comorbidities such as cystic fibrosis were excluded, as were current smokers.
Types of interventions
We included trials comparing mepolizumab with placebo. We planned to include the following cointerventions provided they were not part of the randomised treatment: leukotriene antagonists, inhaled bronchodilators (including long‐acting beta2‐agonists), systemic and inhaled steroids, oral aminophylline and macrolide antibiotics.
Studies that initiated a reduction in standard asthma management as part of the protocol were excluded. Nair 2009 included a reduction in the dose of prednisolone in the second phase of the trial. Therefore, only phase one of this trial was included as patients remained on their standard asthma treatment during this four‐week period.
Types of outcome measures
Primary outcomes
-
HRQoL (as measured by a validated questionnaire)
-
Asthma exacerbation as defined by a hospital admission or treatment with a course of oral corticosteroids
-
Serious adverse events
Secondary outcomes
-
Measures of lung function: forced expiratory flow in one second (FEV1), peak expiratory flow rate (PEFR)
-
Asthma symptoms
-
Adverse events/side effects
-
Eosinophil counts in peripheral blood, sputum or bronchioalveolar lavage fluid
Reporting one or more of the outcomes listed here in the trial was not an inclusion criterion for the review.
Search methods for identification of studies
Electronic searches
We identified trials from the Cochrane Airways Group Specialised Register (CAGR), which is maintained by the Trials Search Co‐ordinator for the Group. The Register contains trial reports identified through systematic searches of bibliographic databases, including the Cochrane Central Register of Controlled Trials (CENTRAL), MEDLINE, EMBASE, CINAHL, AMED and PsycINFO. We also handsearched respiratory journals and meeting abstracts (please see Appendix 1 for further details). We searched all records in the CAGR using the search strategy in Appendix 2.
We also conducted a search of ClinicalTrials.gov (www.clinicaltrials.gov) and the World Health Organization (WHO) trials portal (www.who.int/ictrp/en/).
We searched all databases from their inception to the present and imposed no restriction on language of publication. The search was first conducted in November 2013 and was updated in November 2014.
Searching other resources
We checked the bibliographies of all primary studies and review articles for additional references. We searched relevant manufacturers' websites for trial information.
We searched for errata and retractions relevant to the included studies published in full text on PubMed (www.ncbi.nlm.nih.gov/pubmed) and planned to report the date this was done within the review if this was an issue.
Data collection and analysis
Selection of studies
Two review authors (NW, CP) independently screened titles and abstracts of all the potential studies identified in the search and coded them as 'retrieve' (eligible or potentially eligible/unclear) or 'do not retrieve'. We retrieved the full‐text study reports/publications, and two review authors (NW, CP) independently screened the full text and identified studies for inclusion, identifying and recording reasons for excluding the ineligible studies. We planned to resolve any disagreement through discussion or, if required, by consulting a third author (SJM); however, this was not necessary. We identified and excluded duplicates and collated multiple reports of the same study so that each study rather than each report was the unit of interest in the review. We recorded the selection process in sufficient detail to complete a PRISMA flow diagram and a 'Characteristics of excluded studies' table.
Data extraction and management
We used a data collection form to record study characteristics and outcome data that had been piloted on at least one study in the review. Two review authors (LB, NW) extracted the following study characteristics from included studies.
-
Methods: study design, total duration of study, details of any run‐in period, number of study centres and location, study setting, withdrawals and date of study.
-
Participants: number, mean age, age range, gender, severity of condition, diagnostic criteria, baseline lung function, smoking history, inclusion criteria and exclusion criteria.
-
Interventions: intervention, comparator, concomitant medications and excluded medications.
-
Outcomes: primary and secondary outcomes specified and collected, and time points reported.
-
Notes: funding for trial and notable conflicts of interest of trial authors.
Two review authors (LB, NW) independently extracted outcome data from included studies. We noted in the 'Characteristics of included studies' table if outcome data were not reported in a usable way. We planned to resolve disagreements by consensus or by involving a third author (CP), but this was not necessary. One review author (KD) transferred data into Review Manager (RevMan). We double‐checked that data were entered correctly by comparing the data presented in the systematic review with the study reports. A second review author (SJM) spot‐checked study characteristics for accuracy against the trial report.
Assessment of risk of bias in included studies
Two review authors (LB, NW) independently assessed risk of bias for each study using the criteria outlined in the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011). We planned to resolve any disagreements by discussion or by involving another author (SJM), but this was not necessary. We assessed the risk of bias according to the domains:
-
random sequence generation;
-
allocation concealment;
-
blinding of participants and personnel;
-
blinding of outcome assessment;
-
incomplete outcome data;
-
selective outcome reporting;
-
other bias.
We graded each potential source of bias as high, low or unclear, and provided a quotation from the study report together with a justification for this judgement in the 'Risk of bias' table. We summarised the risk of bias judgements across different studies for each of the domains listed. We considered blinding separately for different key outcomes where necessary (e.g. for an unblinded outcome assessment, risk of bias for all‐cause mortality may be very different than that for a patient‐reported pain scale). Where information on risk of bias related to unpublished data or correspondence with a trialist, we noted this in the 'Risk of bias' table.
When considering treatment effects, we took into account the risk of bias for the studies that contributed to that outcome.
We conducted the review according to this published protocol and have reported any deviations from it in the 'Differences between protocol and review' section of the systematic review.
Measures of treatment effect
We analysed dichotomous data as risk ratios and rate ratios and continuous data as mean differences or standardised mean differences, which are presented with 95% confidence intervals. We entered data presented on a scale with a consistent direction of effect. However, on one occasion we had to use the risk ratio as one study had reported this (Haldar 2009).
We have undertaken meta‐analyses only where this was meaningful (i.e. if the treatments, participants and underlying clinical question were sufficiently similar for pooling to make sense).
Where multiple trial arms were reported in a single trial (Flood‐Page 2007; Pavord 2012), we combined the relevant arms (750 mg, 250 mg, 75 mg in Pavord 2012 and 750 mg, 250 mg in Flood‐Page 2007) when appropriate.
In future updates of this review, we will narratively describe skewed data reported as medians and interquartile ranges. Where multiple trial arms are reported in a single trial, we will include only the relevant arms. If two comparisons (e.g. drug A versus placebo and drug B versus placebo) are combined in the same meta‐analysis, we will halve the control group to avoid double‐counting.
Unit of analysis issues
No cross‐over studies or cluster randomised trials were identified for inclusion in this version of the review. If cross‐over trials are identified in the future, data from a paired analysis will be sought from the trial report or authors in order to appropriately include data in the review using the inverse variance method. If cluster randomised trials are identified in the future, then analyses will be at the level of the individual while allowing for the clustering in the data by using the intracluster correlation coefficient. If this is not reported in the trial, then it will be imputed from similar studies.
Dealing with missing data
Although unnecessary for this version of the review, we may contact investigators or study sponsors for future versions in order to verify key study characteristics and obtain missing numerical outcome data where possible (e.g. when a study is identified as an abstract only). Where this is not possible, and the missing data are thought to introduce serious bias, we will explore the impact of including such studies in the overall assessment of results by a sensitivity analysis.
Assessment of heterogeneity
Statistical heterogeneity between studies was assessed visually by inspection of the forest plots and using the Chi2 test (a P value < 0.10 was considered significant due to the low power of the test). The I2 statistic was also calculated; this describes the percentage of the variability in effect estimates that is due to heterogeneity rather than sampling error (chance). Values of I2 range from 0% to 100%, with 0% representing no heterogeneity and 100% representing considerable heterogeneity.
For this review, heterogeneity as reported using the I2 statistic was defined as follows.
-
0% to 40%: heterogeneity might not be important.
-
30% to 60%: may represent moderate heterogeneity.
-
50% to 90%: may represent substantial heterogeneity.
-
75% to 100%: considerable heterogeneity.
Assessment of reporting biases
If we are able to pool more than 10 trials for future versions, we will create and examine a funnel plot to explore possible small study biases and publication bias.
Data synthesis
In view of the considerable clinical heterogeneity between the included studies, we used a random‐effects model.
Data on outcomes were combined at 6 months and 12 months. Where data for other time points were reported, these were also described.
Subgroup analysis and investigation of heterogeneity
Provided sufficient studies were included, we planned to carry out subgroup analyses according to:
-
age (0 to 5 years, 6 to 16 years, 17 years and older);
-
eosinophilic individuals versus non‐eosinophilic individuals; and
-
dose of intervention (posthoc subgroup identified);
using the outcomes:
-
HRQoL; and
-
asthma symptoms.
If more studies are included in the future, we will use the formal test for subgroup interactions in RevMan.
Sensitivity analysis
We planned to carry out the following sensitivity analyses if sufficient studies were included.
-
Excluding studies with an overall high risk of bias.
-
Excluding cross‐over trials and cluster randomised trials.
Summary of findings table
We created 'Summary of findings' tables using the following outcomes.
-
HRQoL.
-
Asthma exacerbation as defined by a hospital admission or treatment with a course of oral corticosteroids.
-
Serious adverse events.
We used the five GRADE considerations (study limitations, consistency of effect, imprecision, indirectness and publication bias) to assess the quality of the body of evidence as it relates to the studies that contribute data to the meta‐analyses for the prespecified outcomes. We used methods and recommendations described in Section 8.5 and Chapter 12 of the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011) using GRADEpro software. We have justified all decisions to downgrade or upgrade the quality of studies using footnotes, and we have made comments to aid the reader's understanding of the review where necessary.
Results
Description of studies
Results of the search
We identified 154 records in our literature searches: 129 in database searches in November 2013 and a further 25 in November 2014 (Figure 1). Eight studies met our inclusion criteria ('Characteristics of included studies' table), and two others were included in the ongoing studies category ('Characteristics of ongoing studies' table). The eight included studies had 25 records: one for Buttner 2003; seven for Flood‐Page 2003, one for Flood‐Page 2007, four for Haldar 2009, one for Leckie 2000; five for Nair 2009; two for Ortega 2014 and four for Pavord 2012. The remaining 127 records were excluded for various reasons ('Characteristics of excluded studies' table).
Study flow diagram
Included studies
We included eight studies ('Characteristics of included studies' table), involving 1707 total participants distributed as follows: Buttner 2003, 19; Flood‐Page 2003, 24; Flood‐Page 2007, 362; Haldar 2009, 61; Leckie 2000, 24; Nair 2009, 20; Ortega 2014, 576 Pavord 2012, 621. Table 1 compares the design, numbers, interventions and patient groups in the included trials. The severity of asthma among participants varied from mild atopic asthma to persistent eosinophilic asthma with recurrent exacerbations. The mepolizumab was administered exclusively through intravenous route in seven of the studies, with dosage varying from 2.5 mg/kg or 10 mg/kg, or 75 mg, 250 mg and 750 mg with different dosing regimens over a range of treatment periods. Only one study had a subcutaneous (SC) arm, with a dose of 100 mg (Ortega 2014).
| Study | Design | Baseline Asthma severity | Baseline treatment | SC or IV | Intervention (mepolizumab) | Follow‐up | Primary and secondary outcomes | No. participants |
| RCT double‐blind, placebo | Mild allergic asthma | SABA as required | IV | 10 mg/kg versus 2.5 mg/kg versus placebo | 16 weeks | Blood eosinophils, sputum eosinophils, histamine PC20 (mg/mL), late asthmatic reaction (maximum % fall in FEV1) | 24 | |
| RCT parallel group, multicentre double blind | Mild or moderate asthma (FEV1 50‐80% predicted) | 1000 mcg BDP or equivalent and stable | IV | Three 750 or 250 mg or placebo every 4 weeks for 6 months | 6 months | Blood eosinophil, ECP, interferon‐c producing CD4 T‐cells | 19 | |
| 2‐centre double‐blind, placebo‐controlled, parallel‐group study | Mild atopic (skin prick positive) asthma (FEV1 >70% predicted) | SABA as required and no corticosteroids in previous 8 weeks | IV | Three doses of either 750 mg or placebo over 20 weeks (at weeks 0, 4 and 8) | 20 weeks | Airway eosinophils, bone marrow eosinophils, blood eosinophils, airway hyperresponsiveness, FEV1, PEFR | 24 | |
| Multicentre, randomised, double‐blind, placebo‐controlled trial | Moderate asthma (FEV1 between 50% and 80% predicted) | maximum dose (BDP) or equivalent, 1000 mcg/d | IV | Three doses of either 750 mg or 250 mg or placebo over 20 weeks (at weeks 0, 4 and 8) | 20 weeks | Change from baseline morning PEFR recorded at weeks 12 and 20; asthma summary symptom scores; use of rescue medication such as albuterol (salbutamol); quality of life scores; asthma exacerbation rates; eosinophil counts in blood and sputum | 362 | |
| Randomised, double‐blind, placebo‐controlled trial.
| Eosinophilic asthma | Prednisolone treatment with high‐dose ICS | IV | Five doses of either 750 mg or placebo (administered in 5 monthly infusions)
| 26 weeks | Juniper ACQ in patients with asthma or by Likert symptom scores + FEV1 in patients with eosinophilic bronchitis without asthma); the prednisone‐sparing effect of mepolimuzab or placebo as indicated by the absolute and percentage dose reduction possible without a clinical exacerbation (defined as % sputum eosinophilia, FEV1 % predicted and methacholine PC20); blood eosinophils; frequency of rescue salbutamol use; time to exacerbation | 20 | |
| RCT double‐blind, placebo, parallel‐group | Eosinophilic asthma and exacerbations | Sputum eosinophilia of more than 3% despite high‐dose ICS treatment, and at least two exacerbations in previous 12 months | IV | 750 mg versus matched placebo (150 mL of 0.9% saline) at monthly intervals for 1 year | 50 weeks | Severe exacerbations per person; secondary outcomes included a change in asthma symptoms (AQLQ); FEV1 after use of a bronchodilator; airway hyperresponsiveness; eosinophil counts in the blood, sputum | 61 | |
| Multicentre, double‐blind, placebo‐controlled trial
| Eosinophilic asthma and exacerbations | High dose ICS (i.e. ≥ 880 mcg/day FP or equivalent daily) for at least 12 months | IV | 13 infusions in total given every 4 weeks of 750 mg, 250 mg, 75 mg or placebo | 52 weeks | Exacerbations; time to first clinically significant exacerbation; frequency of exacerbations requiring hospitalisation; time to first exacerbation requiring hospitalisation or ED visit; mean change from baseline in clinic pre‐bronchodilator FEV1; mean change from baseline in clinic post‐bronchodilator FEV1; mean change from baseline in ACQ score | 621 | |
| Randomised, double‐blind, double‐dummy, phase 3 study | Persistent eosinophilic asthma | High dose ICS in the 12 months prior to visit 1 with or without maintenance OCS | IV and SC | 75 mg IV dose versus 100 mg SC dose versus placebo every 4 weeks for 32 weeks | 32 | Exacerbations per year; mean change from baseline in clinic pre‐bronchodilator FEV1 at week 32; mean change from baseline in the SGRQ total score at week 32 | 576 |
ACQ: Asthma Control Questionnaire; AQLQ: Asthma Quality of Life Questionnaire; BDP: beclomethasone dipropionate; ECP: eosinophil cationic protein; ED: emergency department; FEV1 : Forced expiratory volume in 1 second; FP; fluticasone propionate; ICS; inhaled corticosteroid; IV: intravenous; PC20 : histamine provocative concentration causing a 20% drop in FEV1; PEFR: peak expiratory flow rate; RCT: randomised controlled trial; SABA: short‐acting beta‐agonists; SC: subcutaneous; SGRQ: St George's Respiratory Questionnaire.
Excluded studies
We excluded 127 records from the review. Of these, 119 (94%) were excluded because mepolizumab had not been included in the study, 4 (3%) were excluded because they did not include a placebo arm, another 2 (2%) were excluded because the focus was on steroid reduction, 1 (1%) was non‐randomised, and the remaining study (1%) was conducted on healthy participants without a diagnosis of asthma ('Characteristics of excluded studies' table).
Risk of bias in included studies
Details of our risk of bias assessments are available in the 'Characteristics of included studies' table, and a summary of our assessment can be seen in Figure 2 and Figure 3.
Risk of bias graph: review authors' judgements about each risk of bias item presented as percentages across all included studies.
Risk of bias summary: review authors' judgements about each risk of bias item for each included study.
Allocation
We deemed only three studies (Nair 2009; Pavord 2012; Ortega 2014) to be at low risk of bias for both random sequence generation and allocation concealment. We judged Haldar 2009 to be at low risk of bias for random sequence generation, but its bias with regard to allocation concealment was unclear. The risk of bias for the remaining four studies (Buttner 2003; Flood‐Page 2003; Flood‐Page 2007; Leckie 2000) was unclear for both random sequence generation and allocation concealment (Figure 3).
Blinding
With regard to performance bias and detection bias, we determined that all eight studies were at low risk of bias (Figure 3).
Incomplete outcome data
In terms of attrition bias, we considered seven of the studies to be at low risk of bias, while the risk of bias in Buttner 2003 was unclear (Figure 3).
Selective reporting
One study noted that there was no significant difference in HRQoL but did not provide any data (Flood‐Page 2007), so we considered it to be at high risk of bias. We deemed all other studies to be at low risk of bias as there was no apparent evidence of selective reporting.
Effects of interventions
See: Summary of findings for the main comparison Intravenous mepolizumab compared to placebo for asthma; Summary of findings 2 Subcutaneous mepolizumab compared to placebo for asthma
Primary outcomes
HRQoL (as measured by a validated questionnaire)
Three studies (participants = 1044) measured quality of life using the Asthma Quality of Life Questionnaire (AQLQ) (Flood‐Page 2007; Haldar 2009; Pavord 2012). One study noted that there was no significant difference but did not provide any data (Flood‐Page 2007).
Intravenous mepolizumab versus placebo
Pavord 2012 reported data at 52 weeks for three different dose groups of Intravenous (IV) mepolizumab (75 mg, 250 mg, 750 mg), which we combined and presented as one group. Haldar 2009 reported data at 50 weeks. Combining the two studies, Analysis 1.1 showed a non‐significant difference between IV mepolizumab and placebo (MD 0.21, 95% CI − 0.01 to 0.44; participants = 682), favouring IV mepolizumab. Our confidence in this result is low, as the mean difference is less the clinical minimally important difference of 0.5 units, and no responder analysis is reported (summary of findings Table for the main comparison).
Ortega 2014 measured quality of life using the St. George's Respiratory Questionnaire (SGRQ) and found a significant difference favouring IV mepolizumab over the placebo (MD 6.40, 95% CI 3.15 to 9.65; participants = 382; Analysis 1.2). We only have moderate confidence in this result, as IV delivery is not currently a licenced route of administration for mepolizumab (summary of findings Table for the main comparison).
Subcutaneous mepolizumab versus placebo
Ortega 2014 measured quality of life using the St. George's Respiratory Questionnaire (SGRQ) and found a significant difference between subcutaneous (SC) mepolizumab and placebo, in favour of mepolizumab (MD − 7.00, 95% CI − 10.19 to − 3.81; participants = 385; Analysis 2.1). We have moderate confidence in this result from a single study (summary of findings Table 2).
Asthma exacerbation as defined by a hospital admission or treatment with a course of oral corticosteroids
Six studies (participants = 1664) reported on asthma exacerbations (Flood‐Page 2003; Flood‐Page 2007; Haldar 2009; Nair 2009; Pavord 2012; Ortega 2014). Increase in oral corticosteroids is included in the definition of exacerbation for three studies (Haldar 2009; Ortega 2014; Pavord 2012). Two studies did not include an increase in oral corticosteroids in the definition of exacerbation (Flood‐Page 2007; Nair 2009), while one study did not provide a definition of exacerbation (Flood‐Page 2003).
IV Mepolizumab versus placebo
Four studies ( Flood‐Page 2003; Flood‐Page 2007; Haldar 2009; Nair 2009) reported the number of patients experiencing an exacerbation. Analysis 1.6, which used a random‐effects model, did not show a significant difference between IV mepolizumab and placebo (Risk Ratio 0.67, 95% CI 0.34 to 1.31; participants = 468 I2 = 59%). Our confidence in this result is low due to the wide confidence intervals (summary of findings Table for the main comparison).
Pavord 2012 reported the rate ratio of exacerbations for each of the three different dose groups of IV mepolizumab compared to placebo. Ortega 2014 reported the percentage reduction in the rate ratio for clinically significant exacerbations for 75 mg IV mepolizumab compared to placebo. We combined the results for groups taking the 75 mg dose from these studies, both of which included participants with severe eosinophilic asthma.
Analysis 1.3 shows similar results for the rate of clinically significant exacerbations, which include a course of oral steroids, emergency department (ED) visit or admission. For the 75 mg dose, the rate of ED visits or hospital admissions for people on mepolizumab was half that of the placebo group (rate ratio 0.52, 95% CI 0.43 to 0.64; participants = 690; studies = 2). For the 250 mg dose, the result was similar (rate ratio 0.61, 95% CI 0.46 to 0.81; participants = 307; studies = 1) and also for the 750 mg dose (rate ratio 0.48, 95% CI 0.36 to 0.64; participants = 311; studies = 1). Our confidence in this result is moderate, as IV delivery is not currently a licenced delivery route for mepolizumab (summary of findings Table for the main comparison).
Analysis 1.4 shows the rate ratio for the combined results of these two studies in terms of exacerbations requiring hospital admission, and there is not a significant difference for the 75 mg mepolizumab dose (rate ratio 0.61, 95% CI 0.33 to 1.13; participants = 690; studies = 2). The 750 mg IV mepolizumab group compared to placebo showed a reduction in the risk of being admitted to hospital (rate ratio 0.37, 95% CI 0.16 to 0.86; participants = 311; studies = 1). The 250 mg dose did not show a statistically significant reduction (rate ratio 0.65, 95% CI 0.31 to 1.37; participants = 307; studies = 1), but the difference between doses was not significant (test for subgroup differences: Chi2 = 1.14, degree of freedom (df) = 2 (P = 0.57), I2 = 0%).
Analysis 1.5 shows the combined results on exacerbations requiring a visit to the ED or hospital admission. For the 75 mg dose, there was a significant reduction in the exacerbation rate for this outcome (rate ratio 0.52, 95% CI 0.31 to 0.87; participants = 690; studies = 2), and although the reduction in rate was similar for the other doses, it did not reach statistical significance (250 mg dose: rate ratio 0.58, 95% CI 0.30 to 1.12; participants = 307; studies = 1; and 750 mg dose: rate ratio 0.52, 95% CI 0.27 to 1.02; participants = 311; studies = 1). Again there was no significant difference between the results according to dose (test for subgroup differences: Chi2 = 0.08, df = 2 (P = 0.96), I2 = 0%).
SC Mepolizumab versus placebo
Ortega 2014 also found a reduction in the rate of all of the above types of exacerbations favouring SC mepolizumab in comparison to placebo. Analysis 2.2 shows the results for hospital admission (rate ratio 0.31, 95% CI 0.11 to 0.91; participants = 385; studies = 1). Analysis 2.3 shows the reduction in either ED visits or hospital admission (rate ratio 0.39, 95% CI 0.18 to 0.83; participants = 385; studies = 1). Analysis 2.4 shows the reduction in clinically significant exacerbations (rate ratio 0.47, 95% CI 0.35 to 0.63; participants = 385; studies = 1). We have moderate confidence in these results from a single study (summary of findings Table 2).
Serious adverse events
Five studies (participants = 1640) reported information on serious adverse events.
Nair 2009 stated that there were no serious adverse events, while Pavord 2012 reported that the overall frequency of serious adverse events was similar across treatment groups and that no serious life‐threatening anaphylactic reactions were observed; however, three patients in the IV mepolizumab groups died during the study for reasons that the physician investigator judged to be unrelated to the treatment.
Flood‐Page 2007 reported nine serious adverse events: four in patients receiving placebo (vertigo, bladder carcinoma, unintended pregnancy and asthma exacerbation), three in patients receiving IV mepolizumab 250 mg (hydrocephalus/cerebrovascular disorder, constipation and gastrointestinal disturbance), and two in patients receiving IV mepolizumab 750 mg (asthma exacerbation). None of these serious adverse events was considered to be related to the study medication, and there were no significant differences between the treatment groups.
Haldar 2009 reported that hospitalisation for asthma was a serious adverse effect for 10% (3/29) of participants in the IV mepolizumab arm and 34% (11/32) in the placebo arm.
Ortega 2014 reported that the incidence of serious adverse events (including asthma‐related events) was 7% in the intravenous mepolizumab group, 8% in the subcutaneous mepolizumab group, and 14% in the placebo group.
Analysis 1.7 indicated that there was a significant difference between IV mepolizumab versus placebo (Risk Ratio 0.49, 95% CI 0.30 to 0.80; participants = 1441; studies = 5; I2 = 0%), favouring IV mepolizumab. Our confidence in this result is moderate, as IV delivery is not currently a licenced route of administration for mepolizumab (summary of findings Table for the main comparison).
Secondary outcomes
Measures of lung function: forced expiratory flow in one second (FEV1), peak expiratory flow rate (PEFR)
Seven studies (participants = 1688) report on lung function (Flood‐Page 2003; Flood‐Page 2007; Haldar 2009; Leckie 2000; Nair 2009; Pavord 2012; Ortega 2014).
IV Mepolizumab versus placebo
Flood‐Page 2003 reported no difference between IV mepolizumab and placebo for median FEV1 and median PEFR at 12 weeks (Table 2).
| Intervention | Control | ||||||||
| Study | Outcome | N | Pre‐dose median (IQR) | Post‐dose median (IQR) | N | Pre‐dose median (IQR) | Post‐dose median (IQR) | Median difference | P value (between groups) |
| FEV1 L/s | 11 | 3.05 (2.69 to 3.28) | 3.1 (2.82 to3.85) | 13 | 3.1 (2.65 to 3.51) | 3.06 (2.65 to 3.45) | 0.15 | 0.22 | |
| Morning PEFR L/min | 11 | 433 (402 to 497) | 436 (417 to 503) | 12 | 459.5 (408 to 481) | 448 (370 to 490) | 21 | 0.09 | |
FEV1 : Forced expiratory volume in 1 second; IQR: interquartile range; PEFR: peak expiratory flow rate
Flood‐Page 2007 reported mean change from placebo for FEV1 (L) and PEFR L/min at weeks 12 and 20. Analysis 1.8 indicates there was no significant difference in FEV1 between IV mepolizumab and placebo at week 20. Analysis 1.9 shows a significant difference for IV mepolizumab 250 mg compared to placebo (MD 13.49; 95% CI 0.71 to 26.27), but not for the 750 mg compared to placebo group (MD 3.42, 95% CI − 9.40 to 16.24). However, the test for subgroup difference was not significant (Chi2 = 1.19, df = 1 (P = 0.280), I2 = 15.9%).
Haldar 2009 and Nair 2009 reported no significant difference in post‐bronchodilator FEV1 (L) between IV mepolizumab and placebo at one year and six weeks, respectively (Analysis 1.10). Nair 2009 also reported no difference between IV mepolizumab and placebo for percentage predicted FEV1 after bronchodilation (Analysis 1.11).
Pavord 2012 found no significant difference between any dose of IV mepolizumab and placebo in pre‐bronchodilator FEV1 (mL) at one year (Analysis 1.13).
Leckie 2000 reports no significant difference between IV mepolizumab and placebo in late asthmatic reaction (maximum percentage fall in FEV1) (Analysis 1.14).
Ortega 2014 reported a statistically significant difference favouring IV mepolizumab for both pre‐ and post‐bronchodilator FEV1 (MD 0.10 L ; 95% CI 0.01 to 0.19); (MD 0.15 L, 95% CI 0.05 to 0.24), (Analysis 1.10; Analysis 1.12).
SC Mepolizumab versus placebo
Ortega 2014 reported a statistically significant difference favouring SC mepolizumab for both pre‐ and post‐bronchodilator FEV1 (MD 0.10, 95% CI 0.02 to 0.18; participants = 385; studies = 1 and MD 0.14, 95% CI 0.04 to 0.23; respectively) (Analysis 2.5; Analysis 2.6;).
Asthma symptoms
Five studies (participants = 1640) measured asthma symptoms (Flood‐Page 2007; Haldar 2009; Nair 2009; Pavord 2012; Ortega 2014).
IV Mepolizumab versus placebo
Flood‐Page 2007 reported results at 20 weeks using the asthma summary symptom score. Nair 2009 reported data at 4 weeks using a symptom score, a cough score and the Juniper Asthma Cough Questionnaire (JACQ) score. Haldar 2009 reported data at one year using the visual analogue scale symptom score and a modified Juniper Asthma Control Score. Pavord 2012 reported data using the asthma control questionnaire at one year. Ortega 2014 reported data at 32 weeks using the five‐item Asthma Control Questionnaire (ACQ‐5).
There were no significant differences between IV mepolizumab at 250 mg or 750 mg and placebo using an asthma symptom score or the JACQ, but there was a significant difference between 75 mg and placebo (MD − 0.30, 95% CI − 0.55 to − 0.04; participants = 690; studies = 2; Analysis 1.15), although test for subgroup difference was again non‐significant (Chi2 = 0.81, df = 2 (P = 0.67), I2 = 0%).
SC Mepolizumab versus placebo
There was also a statistically significant improvement in symptoms on SC mepolizumab compared to placebo (MD − 0.44, 95% CI − 0.64 to − 0.24; participants = 385; studies = 1); Analysis 2.7). However, there was no responder analysis, and this mean difference is less than the minimal clinically important difference of − 0.5 units.
Adverse events/side effects
Six studies (participants = 1664) reported adverse events (Flood‐Page 2003; Flood‐Page 2007; Haldar 2009; Nair 2009; Pavord 2012; Ortega 2014).
Flood‐Page 2003 reported that all of the 24 volunteers completed the study without reporting adverse events.
Flood‐Page 2007 reported that there were no significant differences between the treatment groups for any adverse events reported. The most common adverse events (at least 5% of participants in any treatment group) were upper respiratory tract infection, asthma, headache, rhinitis, bronchitis, sinusitis, viral infection, injury, back pain, nausea and pharyngitis.
Haldar 2009 reported that one patient withdrew due to rash after mepolizumab infusion.
Nair 2009 reported that one patient in the IV mepolizumab group withdrew because of increased shortness of breath, thought to be due to heart failure. One patient in the placebo group died six months after the study because of sudden cardiac arrest; one patient in the IV mepolizumab group reported aches and tiredness when prednisolone was reduced, and one patient in the placebo group had hypoadrenalism when prednisolone was reduced.
Pavord 2012 found that the most frequently reported adverse events were headache (27 (17%) individuals given placebo, 32 (21%) given 75 mg IV mepolizumab, 32 (21%) given 250 mg IV mepolizumab, and 32 (21%) given 750 mg IV mepolizumab) and nasopharyngitis (24 (15%), 34 (22%), 33 (22%), and 29 (19%) for the four groups, respectively). The most frequently reported drug‐related adverse event was infusion‐related reaction (e.g. non‐allergic reactions), which was reported by 10 (6%) patients given placebo, 8 (5%) given 75 mg mepolizumab, 12 (8%) given 250 mg IV mepolizumab, and 19 (12%) given 750 mg IV mepolizumab. Hypersensitivity deemed to be possibly related to investigational product was reported by three patients (2%) given placebo, none given 75 mg IV mepolizumab, one (< 1%) given 250 mg IV mepolizumab, and two (1%) given 750 mg IV mepolizumab.
In the Ortega 2014 study, the overall incidence of adverse events during treatment was similar in the three groups (84% in the IV mepolizumab group, 78% in the SC mepolizumab group, and 83% in the placebo group). The most frequently reported adverse events were nasopharyngitis and headache. The incidence of adverse events that were considered by the study investigators to be related to a study drug was 17% in the IV mepolizumab group, 20% in the SC mepolizumab group, and 16% in the placebo group. The incidence of injection‐site reactions was more frequent in the SC mepolizumab group (9%) than in the IV mepolizumab group or the placebo group (3% in each).
Eosinophil counts in peripheral blood, sputum or bronchioalveolar lavage fluid
All eight studies (participants = 1707) report on eosinophil counts (Buttner 2003; Flood‐Page 2003; Flood‐Page 2007; Haldar 2009; Leckie 2000; Nair 2009; Pavord 2012; Ortega 2014).
Buttner 2003 found that "[a]sthmatic patients received three consecutive intravenous infusions of either IV mepolizumab (250 mg or 750 mg per dose) or placebo at 4‐week intervals. Remarkably, almost a complete disappearance of peripheral blood eosinophils was observed after the first infusion. Eosinophil counts remained low or absent until week 24, 12 weeks after the last infusion. In contrast, there were no significant changes in eosinophil counts in the placebo group. The marked fall in peripheral blood eosinophils was accompanied by a significant decrease in ECP concentrations. The kinetics of ECP (serum eosinophil cationic protein) levels resembled the eosinophil counts. These qualitative and quantitative changes were observed in both treatment groups without a significant difference between the 250 and 750 mg dosage."
Flood‐Page 2003 found that at four weeks after the first dose of IV mepolizumab, there was a significant decrease in peripheral blood eosinophil counts in the actively treated group when compared with placebo (P = 0.002). This decrease was maintained throughout the dosing period and was still evident at the time of the repeat bronchoscopy and bone marrow aspirate, [at] Week 10 (P = 0.02). There was a median reduction of 100% from baseline of eosinophils in the actively treated group at Weeks 4 and 10 (interquartile range, 67–100%). A return of blood eosinophil counts toward baseline was observed at a mean of 9 weeks after the last dose (range 4–20 weeks, data not shown). IV mepolizumab produces a 79% median reduction in bronchoalveolar lavage fluid (BALF) eosinophils (interquartile range, 42‐99%) (P = 0.4 when compared with placebo) (Table 3).
| Outcome | Intervention | Control | P value | ||||
| N | Pre‐dose median (IQR) | Post‐dose median (IQR) | N | Pre‐dose median (IQR) | Post‐dose median (IQR) | ||
| BALF (% cell type on cytospin) Eosinophils | 11 | 1.4 (0.9 to 10.2) | 0.3 (0.01 to 0.8) | 13 | 1.2 (0.2 to 6) | 1.2 (0.3 to 1.6) | 0.4 |
BALF: bronchoalveolar lavage fluid; IQR: interquartile range
Flood‐Page 2007 found a significant reduction in the eosinophil counts in the 250 mg and 750 mg groups at week 1 (P < 0.001). Also, 32 patients gave samples at baseline and week 12; 17 had sputum eosinophils > 3%. There was a significant decrease in both the 250 mg (P = 0.006) and the 750 mg group (P = 0.004), which was also significant when compared to placebo.
Haldar 2009 reports a significant difference between IV mepolizumab and placebo for geometric mean sputum eosinophil percentage during exacerbation, and a sputum eosinophil count > 3% during exacerbation (% of episodes), Table 4. The study also reports, "[T]he geometric mean of eosinophil counts in the blood during the treatment phase, as compared with the baseline value, was reduced by a factor of 6.6 in the mepolizumab group and by a factor of 1.1 in the placebo group, with the changes from baseline differing between the groups by a factor of 6.1 (95% CI, 4.1 to 8.9; P < 0.001)."
| Outcome | Intervention | Control | |||
| N | Percentage | N | Percentage | P value | |
| Geometric mean sputum eosinophil % during exacerbation | 29 | 1.5% | 32 | 4.4% | 0.005 |
| Sputum eosinophil count >3% during exacerbation (% of episodes) | 29 | 36% | 32 | 59% | 0.04 |
Results from Leckie 2000 are presented in Table 5. There was a significant reduction in blood eosinophils pre‐allergen challenge in the group given mepolizumab 10 mg/kg. Postinhaled allergen, there was a significant reduction in blood eosinophils in both groups given mepolizumab. There was a dose dependent reduction in sputum eosinophils in both mepolizumab groups. This result reached statistical significance in the 10 mg/kg group.
| Intervention Mepolizumab (10 mg/kg) N=8 | Intervention Mepolizumab (2.5 mg/kg) N=7 | ||
| Outcome | Day | Difference (95%CI) | Difference |
| Difference in blood eosinophils vs placebo pre‐allergen | Day − 14 | 0.08 (− 0.09 to 0.26), P = 0.4960 | 0.18 (0.01 to 0.36), P = 0.0292 |
| Day 8 | 0.17 (0.04 to 0.30), P = 0.0054 | 0.01 (− 0.16 to 0.19), P = 1.00 | |
| Day 29 | 0.21 (0.10 to 0.33), P < 0.0001 | 0.02 (− 0.14 to 0.18), P = 1.00 | |
| Difference in blood eosinophils vs placebo post‐allergen | Day − 13 | 0.38 (0.07 to 0.69), P = 0.0144 | 0.23 (− 0.11 to 0.58), P = 0.2136 |
| Day 9 | 0.34 (0.13 to 0.55), P = 0.0006 | 0.32 (0.11 to 0.53), P = 0.0012 | |
| Day 30 | 0.49 (0.28 to 0.7), P < 0.0001 | 0.43 (0.22 to 0.65), P = 0.0002 | |
| Difference in sputum eosinophils vs placebo | Day ‐13 | − 2.0 (− 16.2 to 12.3), P = 1.00 | − 2.1 (− 16.3 to 12.2), P = 1.00 |
| Day 9 | 11.3 (2.6 to 20.1), P = 0.0076 | 5.0 (− 5.9 to 16.0), P = 0.6108 | |
| Day 30 | 12.1 (3.1 to − 21.0), P = 0.0050 | 5.9 (− 5.0 to 16.8), P = 0.4454 | |
Ortega 2014 found a significant decrease in both treatment groups in blood eosinophil count.
Results from Nair 2009 can be found in Table 6. In phase 1 of the trial, a single infusion of mepolizumab 750 mg resulted in a reduction in the number of sputum and blood eosinophils.
| Outcome | Visit | Intervention | Control | P value | ||
| N | median (range) | N | median (range) | |||
| Sputum eosinophils (%) median | Visit 1 | 9 | 16.6 (1.6 to 54.3) | 11 | 4.0 (0 to 35.3) | P < 0.05 compared to baseline |
| 4 weeks after first dose | 9 | 0 (range 0 to 4.0) | 10 | 3.0 (0 to 16.3) | P < 0.05 compared to baseline | |
| mean (SD) | mean (SD) | |||||
| Blood eosinophils (x 109/L) | Visit 1 | 9 | 664.4 (492.5) | 11 | 352.1 (± 253.7) | P < 0.05 compared to baseline |
| 4 weeks after first dose | 9 | 49.5 (37.5) | 10 | 295.8 (± 207.4) | P < 0.05 compared to baseline | |
Pavord 2012 found that compared with placebo, the ratios of geometric means at 52 weeks showed that the drug reduced blood eosinophil counts (ratios of geometric means 0.22, 95% CI 0.18 to 0.27) in individuals given 75 mg mepolizumab (P < 0.0001; ratios of geometric means 0.14; 95% CI 0.12 to 0.18), in those given 250 mg mepolizumab, (P < 0.0001; ratios of geometric means 0.12; 95% CI 0.09 to 0.14) and in those given 750 mg mepolizumab, (P < 0.0001). In the subgroup of 94 participants who had sputum induction, the drug also caused decreases in sputum eosinophil counts compared with placebo (ratio 0.68. 95% CI 0.13 to 3.52), in individuals given 75 mg mepolizumab (P = 0.6429; ratios of geometric means 0.35, 95% CI 0.08 to 1.52), in those given 250 mg mepolizumab (P = 0.1577; 0.12 95% CI 0.02 to 0.56) and in those given 750 mg (P = 0.0082).
Discussion
Summary of main results
Eight studies met our inclusion criteria for this systematic review (Buttner 2003; Flood‐Page 2003; Flood‐Page 2007; Haldar 2009; Leckie 2000; Nair 2009; Pavord 2012; Ortega 2014). Six studies included adults participants only, while Pavord 2012 and Ortega 2014 included participants aged 12 years and over, with a mean age of around 50 years and no separate reporting of results in adolescents. In total, 1707 people participated.
The results suggest that mepolizumab leads to an improvement in HRQoL and a reduction in asthma exacerbation rates for people with severe eosinophilic asthma randomised to received mepolizumab compared to placebo, with no significant increase in serious adverse events on treatment.
With regard to the secondary outcome measures, mepolizumab did not lead to a significant increase in measures of lung function (FEV1 or PEFR). There was no significant difference in asthma symptoms using an asthma symptom score or the JACQ between IV mepolizumab at 250 mg or 750 mg and placebo. However, there was a significant difference between 75 mg IV mepolizumab and placebo (although a non‐significant test for subgroup difference) and between SC mepolizumab and placebo, in participants with severe eosinophilic asthma. There were minimal significant adverse events related to mepolizumab, but headache and nasopharyngitis were commonly reported side effects. Due to the variety of dosing regimens and protocols, direct comparison of eosinophil counts in peripheral blood, sputum and bronchoalveolar fluid was not possible.
Peripheral blood eosinophil counts, sputum eosinophil counts and eosinophil counts in bronchoalveolar fluid all showed a significant reduction after treatment with mepolizumab.
There were only two studies that included paediatric patients, down to the age of 12 years old (Ortega 2014; Pavord 2012), but there was no separate reporting of results in adolescents, so we have insufficient evidence to undertake a subgroup analysis based on age.
Overall completeness and applicability of evidence
Although the precise definition of asthma exacerbation is subject to debate, with the consequent variability in reporting, it is nevertheless considered to be one of the core outcomes to be measured in asthma studies (Fuhlbrigge 2012). We found evidence of a reduction in the rate of clinically significant exacerbations in adults with severe eosinophilic asthma given IV or SC mepolizumab. Health‐related quality of life (HRQoL) improved with intervention compared to placebo by a mean of seven units in the single study using SGRQ (Ortega 2014), but the mean change in AQLQ was less than the minimal clinically important difference and was not accompanied by responder analyses. These two primary outcomes are clinically important outcomes for the individual. Secondary outcomes of asthma symptoms scores, cough scores, lung function and airway hyperreactivity were not influenced by mepolizumab. Most studies examined eosinophils, inflammatory markers and mediators using a combination of peripheral blood, sputum and bronchoalveolar lavage and showed reductions in those who received mepolizumab. The clinical relevance of this finding to patients may not be clinically important. There were no studies in children under 12 and only two studies included children aged 12 years or older (but without disaggregating results for the participating adolescents). The asthma population examined in this review was too heterogeneous to draw any conclusions about the general asthma population.
Quality of the evidence
Using the GRADE system, we considered the quality of evidence for IV mepolizumab to be limited, as this is not a licenced delivery route (so we would regard this as indirect evidence). We felt that the HRQoL results were of moderate quality, and further research may have an important effect on the results presented. There was a risk of reporting bias in the assessment of HRQoL for one paper: Flood‐Page 2007 noted no significant changes in HRQoL but did not provide any data, thus no data could be included in the meta‐analysis. We are aware of the limitations in some of the studies and have detailed them in the results section, Figure 2 and Figure 3. We determined that the risk of performance bias and detection bias based on the blinding processes was low in all eight studies. We also found that selection bias was low in only three studies for both random sequence generation and allocation concealment (Nair 2009; Ortega 2014; Pavord 2012) but unclear in four others (Buttner 2003; Flood‐Page 2003; Flood‐Page 2007; Leckie 2000). Haldar 2009 had a low risk of bias for random sequence generation, but the risk of bias was unclear with respect to allocation concealment. Publication bias was not formally assessed through the construction of a funnel plot due to the small number of included studies. However, we performed a thorough search strategy, including searching conference abstracts and ongoing studies, in order to identify unpublished studies.
Potential biases in the review process
We acknowledge the potential for publication bias in this review, as it is possible that we failed to identify unpublished trials that may have provided positive or negative outcomes, which in turn could have altered the treatment benefits. However, to the best of our knowledge, we identified a significant number of trials meeting our inclusion criteria through comprehensive and systematic database searches. We tried to address any study selection bias by having two review authors who independently evaluated all the identified studies. We also ensured that the assessment of each trial was consistently in line with the inclusion criteria.
Agreements and disagreements with other studies or reviews
Our review follows on from Liu 2013, which also considered the efficacy of mepolizumab in patients with asthma. The present review includes one extra study (Ortega 2014), and its findings are consistent with Liu 2013. Both reviews highlight the need for further research in this area.
Risk of bias graph: review authors' judgements about each risk of bias item presented as percentages across all included studies.
Risk of bias summary: review authors' judgements about each risk of bias item for each included study.
Comparison 1 IV Mepolizumab versus placebo, Outcome 1 Health‐related quality of life (AQLQ).
Comparison 1 IV Mepolizumab versus placebo, Outcome 2 Health‐related quality of life (SGRQ).
Comparison 1 IV Mepolizumab versus placebo, Outcome 3 Rate of clinically significant exacerbations.
Comparison 1 IV Mepolizumab versus placebo, Outcome 4 Rate of exacerbations requiring admission.
Comparison 1 IV Mepolizumab versus placebo, Outcome 5 Rate of exacerbations requiring ED or admission.
Comparison 1 IV Mepolizumab versus placebo, Outcome 6 People with one or more exacerbations.
Comparison 1 IV Mepolizumab versus placebo, Outcome 7 Serious adverse events.
Comparison 1 IV Mepolizumab versus placebo, Outcome 8 FEV1 (litres).
Comparison 1 IV Mepolizumab versus placebo, Outcome 9 PEFR (L/min).
Comparison 1 IV Mepolizumab versus placebo, Outcome 10 Post bronchodilator FEV1 (L).
Comparison 1 IV Mepolizumab versus placebo, Outcome 11 Percentage predicted FEV1 after bronchodilation.
Comparison 1 IV Mepolizumab versus placebo, Outcome 12 Pre‐bronchodilator FEV1 (L) at week 32.
Comparison 1 IV Mepolizumab versus placebo, Outcome 13 Pre‐bronchodilator FEV1 (mL) at week 52.
Comparison 1 IV Mepolizumab versus placebo, Outcome 14 Late asthmatic reaction (maximum % fall in FEV1).
Comparison 1 IV Mepolizumab versus placebo, Outcome 15 Asthma symptoms.
Comparison 1 IV Mepolizumab versus placebo, Outcome 16 Asthma symptoms (JACQ).
Comparison 2 SC Mepolizumab versus placebo, Outcome 1 Health‐related quality of life.
Comparison 2 SC Mepolizumab versus placebo, Outcome 2 Rate of exacerbations requiring admission.
Comparison 2 SC Mepolizumab versus placebo, Outcome 3 Rate of exacerbations requiring ED or admission.
Comparison 2 SC Mepolizumab versus placebo, Outcome 4 Rate of clinically significant exacerbations.
Comparison 2 SC Mepolizumab versus placebo, Outcome 5 Pre bronchodilator FEV1 (litres).
Comparison 2 SC Mepolizumab versus placebo, Outcome 6 Post bronchodilator FEV1 (litres).
Comparison 2 SC Mepolizumab versus placebo, Outcome 7 Asthma symptoms.
| IV mepolizumab compared to placebo for asthma | ||||||
| Patient or population: adults with asthma of varying degrees of severity | ||||||
| Outcomes | Illustrative comparative risks* (95% CI) | Relative effect | No. of participants | Quality of the evidence | Comments | |
| Assumed risk | Corresponding risk | |||||
| Placebo | IV mepolizumab | |||||
| Change in HRQoL assessed with AQLQ. Scale from 1 to 7 (higher is better) Follow‐up: 52 weeks | The mean change in HRQoL ranged from 0.18 to 0.71 units | The mean change in HRQoL in the intervention group was 0.21 units more (0.01 fewer to 0.44 more) | ‐ | 682 | ⊕⊕⊝⊝ | Trial participants had severe eosinophilic asthma |
| Change in HRQoL assessed with SGRQ. Scale from: 0 to 100 (lower is better) Follow‐up: 32 weeks | The mean change in HRQoL was − 9.0 units | The mean change in HRQoL in the intervention group was 6.4 units lower (3.15 lower to 9.65 lower) | ‐ | 382 | ⊕⊕⊕⊝ | Trial participants had severe eosinophilic asthma |
| Rate of clinically significant exacerbations ‐ 75 mg mepolizumab versus placebo Follow‐up: range 32 to 52 weeks | The mean rate of clinically significant exacerbations on placebo was 1 per patient per yearc | The mean rate of clinically significant exacerbations in the intervention group was 0.48 less per patient per year (0.57 less to 0.46 less) | Rate ratio 0.52 (0.43 to 0.64) | 690 | ⊕⊕⊕⊝ | Trial participants had severe eosinophilic asthma |
| Rate of clinically significant exacerbations ‐ 250 mg mepolizumab versus placebo Follow‐up: 52 weeks | The mean rate of clinically significant exacerbations on placebo was 0.43 per patient per year | The mean rate of clinically significant exacerbations in the intervention group was 0.17 less per patient per year (0.08 less to 0.23 less) | Rate ratio 0.61 (0.46 to 0.81) | 307 | ⊕⊕⊕⊝ | Trial participants had severe eosinophilic asthma |
| Rate of clinically significant exacerbations ‐ 750 mg mepolizumab versus placebo Follow‐up: 52 weeks | The mean rate of clinically significant exacerbations on placebo was 0.43 per patient per year | The mean rate of clinically significant exacerbations in the intervention group was 0.22 less per patient per year (0.17 less to 0.25 less) | Rate ratio 0.48 (0.36 to 0.64) | 311 | ⊕⊕⊕⊝ | Trial participants had severe eosinophilic asthma |
| People with one or more exacerbations Follow‐up: 20 to 50 weeks | 264 per 1000 | 177 per 1000 | Risk ratio 0.67 | 467 | ⊕⊕⊝⊝ | Variety of asthma severity in the trials |
| Serious adverse events Follow‐up: 20 to 52 weeks | 82 per 1000 | 40 per 1000 | Risk ratio 0.49 | 1441 | ⊕⊕⊕⊝ | Variety of asthma severity in the trials |
| *The basis for the assumed risk is the mean control group risk across studies. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). | ||||||
| GRADE Working Group grades of evidence | ||||||
| aThe intravenous route is not currently licenced for mepolizumab; one point deducted for indirectness. | ||||||
| Subcutaneous mepolizumab compared to placebo for asthma | ||||||
| Patient or population: adults with severe eosinophilic asthma | ||||||
| Outcomes | Illustrative comparative risks* (95% CI) | Relative effect | No. of participants | Quality of the evidence | Comments | |
| Assumed risk | Corresponding risk | |||||
| Placebo | SC mepolizumab | |||||
| Change in HRQoL assessed with SGRQ. Scale from: 0 to 100 (lower is better) Follow‐up: 32 weeks | The mean HRQoL was − 9.0 units | The mean HRQoL ‐ SGRQ in the intervention group was 7 units fewer (10.19 fewer to 3.81 fewer) | ‐ | 385 | ⊕⊕⊕⊝ | |
| Rate of exacerbations requiring admission Follow‐up: 32 weeks | The mean rate of exacerbations requiring admission on placebo was 0.10 per patient per year | The mean rate of exacerbations requiring ED visit or admission in the intervention group was 0.07 less per patient per year (0.01 less to 0.09 less) | Rate ratio 0.31 | 385 | ⊕⊕⊕⊝ | |
| Rate of exacerbations requiring ED or admission Follow‐up: 32 weeks | The mean rate of exacerbations requiring ED or admission on placebo was 0.20 per patient per year | The mean rate of exacerbations requiring ED or admission in the intervention group was 0.12 less per patient per year (0.03 less to 0.16 less) | Rate ratio 0.39 | 385 | ⊕⊕⊕⊝ | |
| Rate of clinically significant exacerbations Follow‐up: 32 weeks | The mean rate of clinically significant exacerbations on placebo was 1.75 per patient per year | The mean rate of clinically significant exacerbations in the intervention group was 0.93 less per patient per year (0.65 less to 1.14 less) | Rate ratio 0.47 | 385 | ⊕⊕⊕⊝ | |
| Asthma symptoms measured on Asthma Control Questionnaire Scale from: 0 to 6 (lower is better)b Follow‐up: 32 weeks | The mean change in asthma symptoms was − 0.5 units | The mean asthma symptoms in the intervention group was 0.44 units fewer (0.64 fewer to 0.24 fewer) | ‐ | 385 | ⊕⊕⊕⊝ | |
| *The basis for the assumed risk was the event rate in the placebo arm of the single included study. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). | ||||||
| GRADE Working Group grades of evidence | ||||||
| aThis finding is from a single study so we do not know how well this will match further research; one point deducted. cThe mean difference is less than the clinical minimally important difference (0.5 units), and no responder analysis is available; one point deducted. | ||||||
| Study | Design | Baseline Asthma severity | Baseline treatment | SC or IV | Intervention (mepolizumab) | Follow‐up | Primary and secondary outcomes | No. participants |
| RCT double‐blind, placebo | Mild allergic asthma | SABA as required | IV | 10 mg/kg versus 2.5 mg/kg versus placebo | 16 weeks | Blood eosinophils, sputum eosinophils, histamine PC20 (mg/mL), late asthmatic reaction (maximum % fall in FEV1) | 24 | |
| RCT parallel group, multicentre double blind | Mild or moderate asthma (FEV1 50‐80% predicted) | 1000 mcg BDP or equivalent and stable | IV | Three 750 or 250 mg or placebo every 4 weeks for 6 months | 6 months | Blood eosinophil, ECP, interferon‐c producing CD4 T‐cells | 19 | |
| 2‐centre double‐blind, placebo‐controlled, parallel‐group study | Mild atopic (skin prick positive) asthma (FEV1 >70% predicted) | SABA as required and no corticosteroids in previous 8 weeks | IV | Three doses of either 750 mg or placebo over 20 weeks (at weeks 0, 4 and 8) | 20 weeks | Airway eosinophils, bone marrow eosinophils, blood eosinophils, airway hyperresponsiveness, FEV1, PEFR | 24 | |
| Multicentre, randomised, double‐blind, placebo‐controlled trial | Moderate asthma (FEV1 between 50% and 80% predicted) | maximum dose (BDP) or equivalent, 1000 mcg/d | IV | Three doses of either 750 mg or 250 mg or placebo over 20 weeks (at weeks 0, 4 and 8) | 20 weeks | Change from baseline morning PEFR recorded at weeks 12 and 20; asthma summary symptom scores; use of rescue medication such as albuterol (salbutamol); quality of life scores; asthma exacerbation rates; eosinophil counts in blood and sputum | 362 | |
| Randomised, double‐blind, placebo‐controlled trial.
| Eosinophilic asthma | Prednisolone treatment with high‐dose ICS | IV | Five doses of either 750 mg or placebo (administered in 5 monthly infusions)
| 26 weeks | Juniper ACQ in patients with asthma or by Likert symptom scores + FEV1 in patients with eosinophilic bronchitis without asthma); the prednisone‐sparing effect of mepolimuzab or placebo as indicated by the absolute and percentage dose reduction possible without a clinical exacerbation (defined as % sputum eosinophilia, FEV1 % predicted and methacholine PC20); blood eosinophils; frequency of rescue salbutamol use; time to exacerbation | 20 | |
| RCT double‐blind, placebo, parallel‐group | Eosinophilic asthma and exacerbations | Sputum eosinophilia of more than 3% despite high‐dose ICS treatment, and at least two exacerbations in previous 12 months | IV | 750 mg versus matched placebo (150 mL of 0.9% saline) at monthly intervals for 1 year | 50 weeks | Severe exacerbations per person; secondary outcomes included a change in asthma symptoms (AQLQ); FEV1 after use of a bronchodilator; airway hyperresponsiveness; eosinophil counts in the blood, sputum | 61 | |
| Multicentre, double‐blind, placebo‐controlled trial
| Eosinophilic asthma and exacerbations | High dose ICS (i.e. ≥ 880 mcg/day FP or equivalent daily) for at least 12 months | IV | 13 infusions in total given every 4 weeks of 750 mg, 250 mg, 75 mg or placebo | 52 weeks | Exacerbations; time to first clinically significant exacerbation; frequency of exacerbations requiring hospitalisation; time to first exacerbation requiring hospitalisation or ED visit; mean change from baseline in clinic pre‐bronchodilator FEV1; mean change from baseline in clinic post‐bronchodilator FEV1; mean change from baseline in ACQ score | 621 | |
| Randomised, double‐blind, double‐dummy, phase 3 study | Persistent eosinophilic asthma | High dose ICS in the 12 months prior to visit 1 with or without maintenance OCS | IV and SC | 75 mg IV dose versus 100 mg SC dose versus placebo every 4 weeks for 32 weeks | 32 | Exacerbations per year; mean change from baseline in clinic pre‐bronchodilator FEV1 at week 32; mean change from baseline in the SGRQ total score at week 32 | 576 | |
| ACQ: Asthma Control Questionnaire; AQLQ: Asthma Quality of Life Questionnaire; BDP: beclomethasone dipropionate; ECP: eosinophil cationic protein; ED: emergency department; FEV1 : Forced expiratory volume in 1 second; FP; fluticasone propionate; ICS; inhaled corticosteroid; IV: intravenous; PC20 : histamine provocative concentration causing a 20% drop in FEV1; PEFR: peak expiratory flow rate; RCT: randomised controlled trial; SABA: short‐acting beta‐agonists; SC: subcutaneous; SGRQ: St George's Respiratory Questionnaire. | ||||||||
| Intervention | Control | ||||||||
| Study | Outcome | N | Pre‐dose median (IQR) | Post‐dose median (IQR) | N | Pre‐dose median (IQR) | Post‐dose median (IQR) | Median difference | P value (between groups) |
| FEV1 L/s | 11 | 3.05 (2.69 to 3.28) | 3.1 (2.82 to3.85) | 13 | 3.1 (2.65 to 3.51) | 3.06 (2.65 to 3.45) | 0.15 | 0.22 | |
| Morning PEFR L/min | 11 | 433 (402 to 497) | 436 (417 to 503) | 12 | 459.5 (408 to 481) | 448 (370 to 490) | 21 | 0.09 | |
| FEV1 : Forced expiratory volume in 1 second; IQR: interquartile range; PEFR: peak expiratory flow rate | |||||||||
| Outcome | Intervention | Control | P value | ||||
| N | Pre‐dose median (IQR) | Post‐dose median (IQR) | N | Pre‐dose median (IQR) | Post‐dose median (IQR) | ||
| BALF (% cell type on cytospin) Eosinophils | 11 | 1.4 (0.9 to 10.2) | 0.3 (0.01 to 0.8) | 13 | 1.2 (0.2 to 6) | 1.2 (0.3 to 1.6) | 0.4 |
| BALF: bronchoalveolar lavage fluid; IQR: interquartile range | |||||||
| Outcome | Intervention | Control | |||
| N | Percentage | N | Percentage | P value | |
| Geometric mean sputum eosinophil % during exacerbation | 29 | 1.5% | 32 | 4.4% | 0.005 |
| Sputum eosinophil count >3% during exacerbation (% of episodes) | 29 | 36% | 32 | 59% | 0.04 |
| Intervention Mepolizumab (10 mg/kg) N=8 | Intervention Mepolizumab (2.5 mg/kg) N=7 | ||
| Outcome | Day | Difference (95%CI) | Difference |
| Difference in blood eosinophils vs placebo pre‐allergen | Day − 14 | 0.08 (− 0.09 to 0.26), P = 0.4960 | 0.18 (0.01 to 0.36), P = 0.0292 |
| Day 8 | 0.17 (0.04 to 0.30), P = 0.0054 | 0.01 (− 0.16 to 0.19), P = 1.00 | |
| Day 29 | 0.21 (0.10 to 0.33), P < 0.0001 | 0.02 (− 0.14 to 0.18), P = 1.00 | |
| Difference in blood eosinophils vs placebo post‐allergen | Day − 13 | 0.38 (0.07 to 0.69), P = 0.0144 | 0.23 (− 0.11 to 0.58), P = 0.2136 |
| Day 9 | 0.34 (0.13 to 0.55), P = 0.0006 | 0.32 (0.11 to 0.53), P = 0.0012 | |
| Day 30 | 0.49 (0.28 to 0.7), P < 0.0001 | 0.43 (0.22 to 0.65), P = 0.0002 | |
| Difference in sputum eosinophils vs placebo | Day ‐13 | − 2.0 (− 16.2 to 12.3), P = 1.00 | − 2.1 (− 16.3 to 12.2), P = 1.00 |
| Day 9 | 11.3 (2.6 to 20.1), P = 0.0076 | 5.0 (− 5.9 to 16.0), P = 0.6108 | |
| Day 30 | 12.1 (3.1 to − 21.0), P = 0.0050 | 5.9 (− 5.0 to 16.8), P = 0.4454 | |
| Outcome | Visit | Intervention | Control | P value | ||
| N | median (range) | N | median (range) | |||
| Sputum eosinophils (%) median | Visit 1 | 9 | 16.6 (1.6 to 54.3) | 11 | 4.0 (0 to 35.3) | P < 0.05 compared to baseline |
| 4 weeks after first dose | 9 | 0 (range 0 to 4.0) | 10 | 3.0 (0 to 16.3) | P < 0.05 compared to baseline | |
| mean (SD) | mean (SD) | |||||
| Blood eosinophils (x 109/L) | Visit 1 | 9 | 664.4 (492.5) | 11 | 352.1 (± 253.7) | P < 0.05 compared to baseline |
| 4 weeks after first dose | 9 | 49.5 (37.5) | 10 | 295.8 (± 207.4) | P < 0.05 compared to baseline | |
| Outcome or subgroup title | No. of studies | No. of participants | Statistical method | Effect size |
| 1 Health‐related quality of life (AQLQ) Show forest plot | 2 | Mean Difference (Random, 95% CI) | Subtotals only | |
| 1.1 AQLQ | 2 | 677 | Mean Difference (Random, 95% CI) | 0.21 [‐0.01, 0.44] |
| 2 Health‐related quality of life (SGRQ) Show forest plot | 1 | Mean Difference (Random, 95% CI) | Subtotals only | |
| 2.1 SGRQ | 1 | Mean Difference (Random, 95% CI) | ‐6.4 [‐9.65, ‐3.15] | |
| 3 Rate of clinically significant exacerbations Show forest plot | 2 | Rate Ratio (Random, 95% CI) | Subtotals only | |
| 3.1 75 mg mepolizumab versus placebo | 2 | 690 | Rate Ratio (Random, 95% CI) | 0.52 [0.43, 0.64] |
| 3.2 250 mg mepolizumab versus placebo | 1 | 307 | Rate Ratio (Random, 95% CI) | 0.61 [0.46, 0.81] |
| 3.3 750 mg mepolizumab versus placebo | 1 | 311 | Rate Ratio (Random, 95% CI) | 0.48 [0.36, 0.64] |
| 4 Rate of exacerbations requiring admission Show forest plot | 2 | Rate Ratio (Random, 95% CI) | Subtotals only | |
| 4.1 75 mg mepolizumab versus placebo | 2 | 690 | Rate Ratio (Random, 95% CI) | 0.61 [0.33, 1.13] |
| 4.2 250 mg mepolizumab versus placebo | 1 | 307 | Rate Ratio (Random, 95% CI) | 0.65 [0.31, 1.37] |
| 4.3 750 mg mepolizumab versus placebo | 1 | 311 | Rate Ratio (Random, 95% CI) | 0.37 [0.16, 0.86] |
| 5 Rate of exacerbations requiring ED or admission Show forest plot | 2 | Rate Ratio (Random, 95% CI) | Subtotals only | |
| 5.1 75 mg mepolizumab versus placebo | 2 | 690 | Rate Ratio (Random, 95% CI) | 0.52 [0.31, 0.87] |
| 5.2 250 mg mepolizumab versus placebo | 1 | 307 | Rate Ratio (Random, 95% CI) | 0.58 [0.30, 1.12] |
| 5.3 750 mg mepolizumab versus placebo | 1 | 311 | Rate Ratio (Random, 95% CI) | 0.52 [0.27, 1.02] |
| 6 People with one or more exacerbations Show forest plot | 4 | 467 | Risk Ratio (M‐H, Random, 95% CI) | 0.67 [0.34, 1.31] |
| 7 Serious adverse events Show forest plot | 5 | 1441 | Risk Ratio (M‐H, Random, 95% CI) | 0.49 [0.30, 0.80] |
| 8 FEV1 (litres) Show forest plot | 1 | Mean Difference (Random, 95% CI) | Subtotals only | |
| 8.1 250 mg mepolizumab versus placebo | 1 | 246 | Mean Difference (Random, 95% CI) | ‐0.03 [‐0.13, 0.07] |
| 8.2 750 mg mepolizumab versus placebo | 1 | 242 | Mean Difference (Random, 95% CI) | 0.02 [‐0.10, 0.14] |
| 9 PEFR (L/min) Show forest plot | 1 | Mean Difference (Random, 95% CI) | Subtotals only | |
| 9.1 250 mg mepolizumab versus placebo | 1 | 246 | Mean Difference (Random, 95% CI) | 13.49 [0.71, 26.27] |
| 9.2 750 mg mepolizumab versus placebo | 1 | 242 | Mean Difference (Random, 95% CI) | 3.42 [‐9.40, 16.24] |
| 10 Post bronchodilator FEV1 (L) Show forest plot | 3 | Mean Difference (Random, 95% CI) | Totals not selected | |
| 10.1 6 weeks | 1 | Mean Difference (Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 10.2 32 weeks | 1 | Mean Difference (Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 10.3 1 year | 1 | Mean Difference (Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 11 Percentage predicted FEV1 after bronchodilation Show forest plot | 1 | Mean Difference (Random, 95% CI) | Totals not selected | |
| 11.1 6 weeks | 1 | Mean Difference (Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 12 Pre‐bronchodilator FEV1 (L) at week 32 Show forest plot | 1 | Mean Difference (Random, 95% CI) | Totals not selected | |
| 12.1 75 mg mepolizumab versus placebo | 1 | Mean Difference (Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 13 Pre‐bronchodilator FEV1 (mL) at week 52 Show forest plot | 1 | Mean Difference (Random, 95% CI) | Subtotals only | |
| 13.1 75 mg mepolizumab versus placebo | 1 | 308 | Mean Difference (Random, 95% CI) | 61.00 [‐37.00, 161.00] |
| 13.2 250 mg mepolizumab versus placebo | 1 | 307 | Mean Difference (Random, 95% CI) | 81.00 [‐18.51, 180.51] |
| 13.3 750 mg mepolizumab versus placebo | 1 | 311 | Mean Difference (Random, 95% CI) | 56.0 [‐41.00, 155.00] |
| 14 Late asthmatic reaction (maximum % fall in FEV1) Show forest plot | 1 | Mean Difference (Random, 95% CI) | Subtotals only | |
| 14.1 2.5 mg/kg mepolizumab versus placebo | 1 | 16 | Mean Difference (Random, 95% CI) | 3.50 [‐3.46, 10.46] |
| 14.2 7.5 mg/kg mepolizumab versus placebo | 1 | 16 | Mean Difference (Random, 95% CI) | 0.3 [‐6.50, 7.10] |
| 15 Asthma symptoms Show forest plot | 5 | Mean Difference (Random, 95% CI) | Subtotals only | |
| 15.1 75 mg mepolizumab versus placebo | 2 | 690 | Mean Difference (Random, 95% CI) | ‐0.30 [‐0.55, ‐0.04] |
| 15.2 250 mg mepolizumab versus placebo | 2 | 553 | Mean Difference (Random, 95% CI) | ‐0.24 [‐0.48, 0.01] |
| 15.3 750 mg mepolizumab versus placebo | 4 | 631 | Mean Difference (Random, 95% CI) | ‐0.02 [‐0.57, 0.54] |
| 16 Asthma symptoms (JACQ) Show forest plot | 2 | 80 | Mean Difference (IV, Random, 95% CI) | ‐0.04 [‐0.42, 0.35] |
| Outcome or subgroup title | No. of studies | No. of participants | Statistical method | Effect size |
| 1 Health‐related quality of life Show forest plot | 1 | Mean Difference (Random, 95% CI) | Totals not selected | |
| 1.1 SGRQ | 1 | Mean Difference (Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 2 Rate of exacerbations requiring admission Show forest plot | 1 | Rate Ratio (Random, 95% CI) | Totals not selected | |
| 3 Rate of exacerbations requiring ED or admission Show forest plot | 1 | Rate Ratio (Random, 95% CI) | Totals not selected | |
| 4 Rate of clinically significant exacerbations Show forest plot | 1 | Rate Ratio (Random, 95% CI) | Totals not selected | |
| 5 Pre bronchodilator FEV1 (litres) Show forest plot | 1 | Mean Difference (Random, 95% CI) | Totals not selected | |
| 5.1 32 weeks | 1 | Mean Difference (Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 6 Post bronchodilator FEV1 (litres) Show forest plot | 1 | Mean Difference (Random, 95% CI) | Totals not selected | |
| 6.1 32 weeks | 1 | Mean Difference (Random, 95% CI) | 0.0 [0.0, 0.0] | |
| 7 Asthma symptoms Show forest plot | 1 | Mean Difference (Random, 95% CI) | Totals not selected | |
