A randomised, double-blind, placebo-controlled study to evaluate the efficacy of oral azithromycin as a supplement to standard care for adult patients with acute exacerbations of asthma (the AZALEA trial)

Importance of asthma exacerbations

Asthma is the most prevalent respiratory disease; in developed countries it is diagnosed in 5–10% of adults and 10–15% of children, with around 30% of children reporting wheeze in the last year. 1 Most asthma morbidity and mortality results from acute exacerbations: 5–10% of asthmatics have been hospitalised with an exacerbation and an estimated ≈25,000 of Europeans die unnecessarily of asthma each year. Exacerbations also account for ≈50% of total expenditure on asthma care. 2 More than half of asthma patients report having an exacerbation in the last year, with more than one-third of children and more than one-quarter of adults requiring urgent medical care visits as a result. 3

Aetiology of asthma exacerbations

Respiratory viral infections are the major cause of asthma exacerbations in children (80–85%)4,5 and adults (75–80%). 6–8 However, non-viral respiratory pathogens such as Mycoplasma pneumoniae and Chlamydophila pneumoniae have also been associated with wheezing episodes and asthma exacerbations in both adults and children. 9–13 Interestingly, in two of these studies virus detection rates were ≈80%,9,11 whereas serological positivity for atypical bacterial infection/reactivation can be as high as 40–60%,9,13 indicating that viral and atypical bacterial infections probably interact in increasing the risk of asthma exacerbations.

There is little published evidence that standard bacterial infections are important in the aetiology of asthma exacerbations;14 however, patients with asthma have increased susceptibility to respiratory bacterial infections,15–17 increased carriage of pathogenic respiratory bacteria identified by culture18 and molecular techniques,19 and impaired interferon responses to bacterial polysaccharides. 20 In addition, viral infection impairs innate responses that are important in antibacterial immunity21 and increases bacterial adherence to bronchial epithelial cells. 22 There is therefore good evidence that bacterial respiratory infections are both more common and more severe in asthma and that viral infection can increase susceptibility to bacterial infection.

A recent study of 361 children with > 800 stable and exacerbation airway samples collected during the first 3 years of life and analysed for standard bacteria and respiratory viruses has confirmed that acute wheezing episodes were associated with both bacterial infection [odds ratio 2.9, 95% confidence interval (CI) 1.9 to 4.3; p < 0.001] and viral infection (odds ratio 2.8, 95% CI 1.7 to 4.4; p < 0.01). 23 We therefore hypothesise that standard bacterial infections are also likely to be important in the aetiology of asthma exacerbations in adults and aimed to investigate this in this study.

Treatment of asthma exacerbations

When asthma exacerbations occur, treatment options are limited to bronchodilators and (cortico)steroids. Beyond the addition of magnesium in severe exacerbations, treatments have developed very little in the last 50 years. Current therapeutic strategies are of limited efficacy and the development of new approaches addressing the aetiological agents causing the exacerbations is urgently needed. Current asthma guidelines recommend specifically that antibiotic therapy should not be administered routinely in asthma exacerbations. 24

New approaches to treatment of asthma exacerbations

If atypical bacteria are causal or contributory factors in asthma exacerbations, then treatment with antibiotics with activity against mycoplasma and chlamydia species would be expected to be beneficial in asthma exacerbations. Adults with acute exacerbations of asthma and treated with telithromycin (a ketolide antibiotic closely related to macrolides, with both classes being highly active against M. and C. pneumoniae) as a supplement to standard care showed a statistically significantly greater reduction in asthma symptoms (p < 0.005) and improvement in lung function (p = 0.001) and faster recovery (p = 0.03) than those treated with placebo. 13 The magnitude of the treatment effect was also highly clinically significant, with the improvement in symptoms resulting from telithromycin treatment being approximately 50% greater than with standard therapy (plus placebo). Improvement in lung function was almost 100% greater and, importantly, recovery time to a 50% improvement in clinical symptoms 3 days faster in those receiving active treatment. This treatment therefore had a clear therapeutic effect; however, this study requires confirmation in a second similar study before revision of guidelines can be considered. Ideally, confirmation would be with a further study with telithromycin; however, issues with toxicity have limited the use of telithromycin to severe life-threatening infections.

The macrolide antibiotic azithromycin is a safe and well-tolerated alternative that has been used for many years in the treatment of respiratory disease but has thus far not been studied in acute exacerbations of asthma. We therefore hypothesised that treatment with azithromycin might be of benefit in acute asthma exacerbations. The AZALEA (AZithromycin Against pLacebo for acute Exacerbations of Asthma) study investigated the effectiveness of azithromycin as a supplement to standard care for adult patients with acute exacerbations of asthma, following as closely as possible the design of the telithromycin study, with the aim of providing confirmation or otherwise of those results.

Mechanisms of activity of macrolide/ketolide antibiotics in the treatment of asthma exacerbations

Macrolide/ketolide antibiotics might have therapeutic effects/benefits in treating asthma exacerbations through treatment of either standard or atypical bacterial infections or both. In addition, both macrolide and ketolide antibiotics have anti-inflammatory properties that are independent of their antibacterial activity, which may be beneficial in reducing airway inflammation, and is known to be important in the pathogenesis of asthma exacerbations. 7,25 In addition to these three possible mechanisms of action, we also believe that antiviral activity is a fourth possible mechanism.

We have previously reported that impaired type I and type III interferon production by virus-infected bronchial epithelial cells and macrophages is important in the pathogenesis of asthma exacerbations. 20,26 We have also recently shown that azithromycin, but not erythromycin or telithromycin, significantly increased rhinovirus-induced type I and type III interferon and interferon-stimulated antiviral protein production in primary bronchial epithelial cells, as well as significantly reducing rhinovirus replication and release in bronchial epithelial cells. 27 Azithromycin has also been shown to reduce illness severity in a mouse model of viral bronchiolitis. 28 Thus, azithromycin has the potential to have direct antiviral activity by augmenting the production of those interferons that we have already shown to be deficient in asthma exacerbations,20,26 and this activity may make it a better treatment option than telithromycin, which does not appear to have this property. 27 A further mechanistic aim of our study, therefore, was to investigate the frequencies of standard bacterial, atypical bacterial and viral infections in these exacerbations to determine the relative importance of each of these infections, and of possible coinfections with one or more agents, in the aetiology of acute exacerbations of asthma in adult subjects. We have also performed subgroup analyses to determine whether or not any treatment benefit observed is greater in those with evidence of one or more of these infections, with the aim of shedding some light on the possible mechanism(s) of action of azithromycin in this context.

Concerns regarding antimicrobial resistance

This clinical trial is important as there are significant concerns regarding the development of resistance against macrolide antibiotics. Although these concerns are somewhat mitigated by the short course of therapy being studied (relative, for example, to ongoing clinical trials investigating long-term treatment in severe asthma), determining whether or not azithromycin has efficacy in this context will, if the study is negative, help limit the inappropriate use of antibiotics (in a recent study of adult asthma exacerbations, 57% of subjects received antibiotics29).

If the study is positive, then determining the frequencies of detection of standard bacterial, atypical bacterial and viral infections in these exacerbations, combined with the subgroup analyses assessing the efficacy of the intervention in those with evidence of one or more of these infections, would help guide the use of such therapies in subgroups of asthma exacerbations that may respond better to such therapies, as well as guiding future investigation of the efficacy of alternative antibiotics with shorter durations of action or different spectra of antimicrobial/viral/inflammatory activity.

Choice of and duration of therapy

Although the course of azithromycin therapy is only 3 days, azithromycin has a multiple-dose tissue half-life of 68 hours and will therefore persist in the lung at significant concentrations for around 10 days after a 3-day course of therapy. 30 The main aim of this study was to determine whether or not the telithromycin results could be validated in a study with a similar antibiotic, with a similar mechanism and duration of action. Telithromycin was given for 10 days (the standard licensed duration of therapy for other respiratory indications) and the primary outcome variable was assessed at 10 days. 13 As our aim was to determine whether or not the telithromycin results can be validated, we felt that it was important to use the same primary outcome variable and as similar a duration of action as is possible (given that we cannot use telithromycin because of liver toxicity). This was one reason that we chose to study azithromycin rather than other macrolide antibiotics. Other reasons for choosing azithromycin include its antiviral activity, not shared with other macrolides,27 a more favourable drug interaction profile30 and excellent concentration at sites of infection. 30

Are patterns of airway inflammation associated with aetiology and treatment outcomes?

Different patterns of airway inflammation have been identified in both stable asthma and during exacerbations; these have been classified as neutrophilic, eosinophilic, mixed granulocytic or pauci-granulocytic. However, it is not known whether or not these different patterns of inflammation are associated with different aetiologies for the exacerbation, nor whether or not they are related to treatment outcome. Our final aim was, therefore, to characterise the inflammatory cell profiles in sputum at presentation, to determine if exacerbation aetiology as well as any possible treatment benefit was related to the types of airway inflammation present (neutrophilic, eosinophilic, mixed granulocytic or pauci-granulocytic).

Need for the AZALEA study

There are no systematic reviews of and no published reports of clinical trials investigating the efficacy of azithromycin in the treatment of (acute) asthma exacerbations. At the time of protocol development for this study there were no similar studies registered on ClinicalTrials.gov. The only somewhat similar study is the AZMATICS (AZithroMycin/Asthma Trial In Community Settings) study (NCT00266851),31 which planned to enrol 200 adult patients with asthma, either stable persistent or in exacerbation, and treat for 3 months to answer the question, ‘Will a 12-week treatment with the antibiotic, azithromycin, result in a statistically significant and clinically meaningful improvement in overall asthma symptoms and other patient-oriented asthma outcomes 1 year after initiation of treatment of adult primary care patients with asthma?’. Thus, the aims, design, timing of outcome analysis and treatment length are clearly very different from those of the AZALEA study.

Inclusion criteria

• Adults of either sex aged 18–55 years or aged 56–65 years with a < 20 pack-year smoking history or aged > 65 years with a < 5 pack-year smoking history.

• Patients with a documented history of asthma for > 6 months and presenting within 48 hours (of initial presentation to medical care) with an acute deterioration in asthma control (increased wheeze, dyspnoea and/or cough and/or reduced PEF) and requiring a course of oral steroids.

• Patients with a PEF or FEV₁ of < 80% of predicted normal or patient’s best at presentation, at recruitment or in the time elapsed between presentation and recruitment.

• Patients must be able to complete diaries and quality-of-life questionnaires.

• Patients must sign and date an informed consent form prior to any study procedures.

Exclusion criteria

• Patients with known prolongation of the QT interval, with a history of torsades de pointes, congenital long QT syndrome, bradyarrhythmias or uncompensated heart failure, on drugs known to prolong the QT interval, with ongoing proarrhythmic conditions such as uncorrected hypokalaemia or hypomagnesaemia or clinically significant bradycardia and receiving class IA (quinidine, procainamide) or class III (dofetilide, amiodarone, sotalol) antiarrhythmic agents.

• Smokers aged 56–65 years with a ≥ 20 pack-year history or aged > 65 years with a ≥ 5 pack-year history.

• Patients requiring immediate transfer/referral to an intensive care unit.

• Patients who have used oral or systemic antibiotics within 28 days prior to enrolment.

• Patients with known impaired hepatic function (alanine aminotransferase/aspartate aminotransferase) more than two times the upper limit of normal.

• Patients with significant lung disease [including chronic obstructive pulmonary disease (COPD)] other than asthma.

• Patients taking > 20 mg of oral corticosteroids daily as maintenance therapy.

• Patients requiring other antibiotic therapy.

• Patients who are receiving other medications or who have other disease conditions or infections that could interfere with the evaluation of drug efficacy or safety.

• Women who are breastfeeding or who are pregnant, as demonstrated by a urine pregnancy test carried out before exposure to study medication or the start of any study procedure that could pose a risk to the fetus.

• Patients with suspected or known hypersensitivity to, or suspected serious adverse reaction to, azithromycin or any of the macrolide or ketolide class of antibiotics, erythromycin or any excipients thereof.

• Patients who have received treatment with any other investigational drug within 1 month prior to study entry or who have such treatment planned for the study period during the treatment or follow-up phase.

• Patients with a concomitant condition (including clinically relevant cardiovascular, hepatic, neurological, endocrine or other major systemic disease) making implementation of the protocol or interpretation of the study results difficult.

• Patients with a mental condition rendering them unable to understand the nature, scope and possible consequences of the study.

• Patients unlikely to comply with the protocol, for example unco-operative attitude or an inability to return for follow-up visits.

• No patient was allowed to enrol in this study more than once.

Primary outcome

The primary outcome was the diary card summary symptom score, with symptoms including wheezing, breathlessness and coughing, assessed at 10 days after randomisation.

Secondary outcomes

• Additional efficacy end points assessed at baseline and 5 and 10 months post randomisation were:

  • health status assessed by Acute AQLQ

  • health status assessed by Mini AQLQ

  • pulmonary function tests (FEV₁, FVC, FEV₁ : FVC ratio, PEF, FEF_25–75%, FEF_50%).

• Primary and secondary outcomes assessed 5 and 10 days post randomisation to permit better estimation of the optimum timing of the assessment of primary/secondary outcome variables in future similar studies (the efficacy of telithromycin was assessed at 10 days only).

• Time to a 50% reduction in symptom score.

Exploratory analyses

• Trends in primary and secondary outcomes over the time course of the exacerbation up to 10 days.

• Assessment of efficacy outcomes in relation to initial C. pneumoniae and/or M. pneumoniae status.

• Assessment of efficacy outcomes in relation to initial standard bacteriological status.

• Assessment of efficacy outcomes in relation to initial virological status.

• Assessment of efficacy outcomes in relation to initial sputum inflammatory cell status.

Electronic case record form

Data management was carried out using the InForm™ ITM (Integrated Trial Management) System version 4.6, a web-based data entry system that builds an Oracle database for each individual clinical trial (Oracle Corporation, Redwood City, CA, USA). Trial data were captured on a bespoke web-based electronic case report form (eCRF), with built-in validation rules to identify data entry errors in real time and a full audit trail of data entry and changes. All those entering data were trained prior to start-up and given personal login details, with access to forms restricted according to site and role. The eCRF was designed in accordance with the requirements of the trial protocol.

Timescale of trial evaluations

Schedule of investigations

A summary of the tests and investigations undertaken is provided in Table 1.

Pulmonary function tests

A spirometer meeting all American Thoracic Society recommendations was used for the pulmonary function tests, performed at visits 1–3. Pulmonary function tests were measured three times in a consistent position (standing or sitting) throughout the study. The best FEV₁, FVC, FEV₁ : FVC ratio, FEF_25–75%, FEF_50% and PEF were recorded on the eCRF as follows:

1. FEV₁ in litres

2. FVC in litres

3. FEV₁ : FVC ratio

4. FEF_25–75% in litres per second

5. FEF_50% in litres per second

6. PEF in litres per minute.

Patients’ daily recordings

All patients were supplied with a diary in which to record salbutamol (reliever) use, asthma symptom ratings and number of night-time awakenings because of asthma symptoms. The daily diary included four questions measuring daytime asthma symptoms, with patients rating the frequency and severity of their symptoms on a 7-point scale (with 0 denoting no symptoms and 6 denoting severe symptoms). The summary diary symptom score was calculated as the mean of these four daytime activity scores (the frequency of asthma symptoms, the severity of asthma symptoms, the level of activity performed today and the effect of asthma on activity) recorded at the end of the day. At visit 1, patients were instructed on how to record information in the diary (Box 1) and were asked to complete the diary at the end of each day for 10 days (with the nocturnal questions referring to the previous night). They were reminded of the recording instructions at visits 2 and 3 and to return all of the completed diary cards to the site at visit 4. All diary cards were retained in the participant files for data entry and monitoring.

Clinical sample collection

Health outcomes data

Health outcomes were measured to determine overall assessment of symptom resolution during the first 10 days based on global subject diary assessment.

Health status was assessed at visits 1–3 using the Acute AQLQ and Mini AQLQ.

Each site was provided with:

• an Acute AQLQ

• a Mini AQLQ

• background information on the AQLQ, administration information and details on the analysis of the AQLQ

• AQLQ coloured cards.

All staff at the different sites who were responsible for administering the AQLQs were asked to familiarise themselves with the contents of the different items before administering the questionnaires.

The questionnaires were interviewer administered and not self-administered. The Acute AQLQ contained a response sheet with columns used to record the responses at visits 1, 2 and 3. A new Mini AQLQ was printed for each visit when the Mini AQLQ was administered and patient responses were recorded directly onto the AQLQ. All patient responses/completed AQLQs were kept in the participant files for source data verification (SDV) and were also entered into the InForm eCRF database.

If possible, staff at the different sites were asked to complete the AQLQ first during clinic visits, before any discussion with a health professional, as this may have influenced how patients completed the questionnaires.

Definitions

Causality

The assignment of the causality of AEs and ARs was made by the investigator responsible for the care of the participant using the definitions in Table 2. If any doubt about causality existed the local investigator would inform the chief investigator.

Period of observation

For the purposes of this study, the period of observation extended from the time that a subject gave informed consent until 7 days after the last dose of study medication.

Reporting procedures

All AEs that occurred after a subject had signed the informed consent were documented on the pages provided on the case report form. The trial eCRF included dedicated forms for reporting SAEs.

Serious adverse events and suspected unexpected serious adverse reactions

Fatal or life-threatening SAEs and SUSARs were reported to the chief investigator (who reported to the sponsor) on the day that the local site became aware of an event. The SAE form included the nature of the event, the date of onset, severity, corrective therapies given, outcome and causality (i.e. unrelated, unlikely, possible, probably, definitely). Additional information was sent to the chief investigator and sponsor within 5 days if the event had not resolved at the time of reporting.

Serious adverse events

Investigators were advised to report SAEs via the eCRF within 24 hours of becoming aware of an event and to include an assessment of expectedness and causality in the SAE report. Each SAE report was reviewed by the Imperial Clinical Trials Unit (ICTU) and chief investigator. The flow chart provided in Figure 2 shows the reporting procedures used.

FIGURE 2.

Reporting procedure for AEs. CRGO, Clinical Research and Governance Office.

Suspected unexpected serious adverse reactions

If an AR was considered to be serious, unexpected and related to the IMP (possibly, probably or definitely related), this would have met the definition of SUSAR, requiring expedited reporting to the MHRA, REC and sponsor. However, there were no SUSARs in the AZALEA study.

Annual safety reports

Annual safety reports were provided to the REC and MHRA, in accordance with clinical trial regulations, on the anniversary of the clinical trial authorisation each year. A total of three annual safety reports were submitted over the course of the trial.

Sample size

The sample size calculation was based on the primary outcome: change from baseline in diary card summary asthma symptom scores at 10 days after randomisation. Our previous study [the Telithromycin, Chlamydophila, and Asthma trial (TELICAST)]13 found a mean decrease in symptom score of 1.3 in the treatment group and 1.0 in the control group, resulting in a difference of –0.3 [standard deviation (SD) 0.783] between the groups at 10 days. In the previous study this was accompanied by a difference in lung function (FEV₁) of 290 ml and by a 3-day faster time to a 50% improvement in symptoms, both of which were clinically impressive. We therefore felt that a difference of 0.3 in symptom score was clinically relevant.

Using a two-sided t-test at a 1% significance level with 80% power, 161 patients in each group were needed to detect the same difference in asthma scores between the groups. A significance level of 1% was chosen to provide greater certainty in the assessment of the primary outcome variable, as well as greater power for the subgroup exploratory analyses, as those subgroup analyses that were performed in the 280-patient TELICAST study13 were uninformative.

Taking into account the dropout rate of 15% in our previous study,13 we proposed to recruit 190 patients in each arm of the study. To be able to run the trial within the project timelines, we initially intended to involve 10 centres.

Randomisation

Randomisation was web based via access to a secure Imperial College London server and was performed using the InForm ITM System. Patient allocation was stratified by centre performed in random length blocks. Either the research nurse or the principal investigator (PI) at each site enrolled and then randomised each patient into the study using the InForm database.

The randomisation lists were generated by an ICTU statistician. Details such as the block size were kept confidential and held separately by the ICTU.

Blinding

This was a double-blind trial and so all participants and care providers and therefore those assessing outcomes were blinded to study treatment. Members of the trial team managing and analysing the data were also blind to the treatment received.

The identity of the study medications was blinded and the medications were packaged and supplied by Sharp Clinical Services (Crickhowell, UK) with code-break envelopes. Overencapsulated azithromycin capsules and placebo capsules were placed into child-resistant tamper-evident containers and a randomised label applied to each container.

Definition of day 1 for patient diary scores

Day 1 was defined in the statistical analysis plan (SAP) as follows: ‘the day 1 diary card of each patient will be defined to be the first diary that is completed within 24 hours of randomisation’. This definition was not applicable because the actual time of completing the diary cards was not recorded in InForm (nor the time of randomisation); therefore, if the date of the diary was the day after the randomisation it was not possible to determine whether or not that was within 24 hours.

After careful consideration, the study team agreed to amend the definition of day 1 to the following: ‘the day of the administration of the first dose of study drug, or if it is not available, then the day of the randomisation. If no diary card has been completed on that day, their day 1 diary card will be treated as missing. Diary cards for days 2–10 will be determined in a similar way’.

The date of study drug administration was available for all randomised patients.

This had no effect on the secondary outcomes as both the Acute AQLQ, Mini AQLQ and pulmonary function tests were conducted only three times and the analysis was based on the order of visits not on the days.

Missing data

Before starting data analysis, the level and pattern of missing data in the baseline variables and outcomes were analysed by forming appropriate tables. Additionally, the likely causes of any missingness were investigated. This information was used to determine whether the level and type of missing data had the potential to introduce bias into the analysis or to substantially reduce the precision of estimates related to treatment effects. Missing data in the patient diaries took one of several forms: no patient diary returned for any day (patient missingness), all data missing for ≥ 1 day (day missingness) and data missing for some but not all of the individual questions for a particular day (item missingness). Of these, the level of item missingness was expected to be minimal. According to the SAP, if any item missingness occurred the scores for the missing questions were interpolated from the previous and subsequent day scores. This process was conducted for two missing entries.

If any item missingness occurred in AQLQ scores, the summary score for that day was treated as missing.

Missing data for the pulmonary function tests were expected to occur because of the spirometer not recording some measures. As this was unrelated to patient outcome, it was reasonable to assume that this missingness was uninformative and that multilevel models fitted to all observed data would provide unbiased parameter estimates.

Modelling patient diary scores

The modelling process for patient diary scores was carried out based on the methods outlined in the SAP, which was finalised before the unblinding. All patients who returned at least one diary card (and received the study drug) were included in the analysis but only those diary cards that were collected in the investigated 10-day time frame were included. Clinical efficacy analyses were carried out on an intention-to-treat basis. A data check for outliers was also planned in the SAP, which included a series of longitudinal plots (one for each centre) of diary score for each patient, differentiating between treatment arm; box plots of diary score by treatment arm for each day were also produced as well as a table of summary statistics of diary score by day and treatment arm (including mean, SD, median and lower and upper quartiles).

Multilevel modelling was used to calculate the unbiased estimates of differences in diary scores between the treatment arms for each day. The terms used in the model were set out in the SAP and are explained in the following section. Different relationships between time and diary scores were compared including linear, quadratic and square root relationships. These models differed in their ‘time’ covariate. Fixed and random effects and the use of splines were also investigated. The goodness of fit of these models was assessed by residual plots, which are provided in Appendix 1. The final model includes only time and time × treatment interaction as covariates and diary score as the outcome variable. Treatment by itself was not included in the model, which means that there was not any difference in diary scores assumed at baseline between the treatment arms. Any observed difference could happen only by chance (caused by the randomisation).

The three main components of the model

Let DS_id represent the diary score for patient i on day d, d = 1, . . .,10 and t(i) represent the treatment given to individual i (azithromycin or placebo). Then, DS_id was modelled as the sum of three components: an intercept term, a change over time term and a residual error term, that is:

Possible choices for each of these components are outlined in the following sections. The options explored for the primary analysis were determined by the results of the exploratory analysis and the final choice was the simplest model that satisfies standard checks of model fit (e.g. residual plots).

Modelling Acute Asthma Quality of Life Questionnaire score, Mini Asthma Quality of Life Questionnaire score and pulmonary function

Similar models were used to assess the day 10 differences in change in Acute AQLQ and Mini AQLQ scores and pulmonary function between the two treatment arms.

Statistical analysis plan

A SAP was prepared by the trial investigators and trial statistician and reviewed and agreed by the Trial Steering Committee (TSC) and Data Monitoring and Ethics Committee (DMEC) prior to the end of the recruitment period.

Statistical software

All analyses were performed using Stata 13 (StataCorp LP, College Station, TX, USA).

Trial management

The UK Clinical Research Collaboration-registered ICTU was responsible for trial management, quality assurance, trial statistics and development and maintenance of the trial database. A dedicated trial manager and clinical trials monitor were appointed through the ICTU to oversee the day-to-day management and monitoring of the project from set-up to close.

Trial sponsor

The sponsor of the trial was Imperial College London. The sponsor’s role is clearly set out in the European Clinical Trials Directive36 and NHS Research Governance Framework37 documents. Imperial College London signed a clinical trial agreement with each of the participating centres prior to the start of recruitment at each centre.

Ethical considerations

The trial was conducted in accordance with the Declaration of Helsinki [see www.wma.net/ (accessed 4 April 2016)] on research involving human subjects. The study protocol, patient information sheet and consent form were submitted to the REC prior to the start of the study and a favourable opinion was obtained on the 15 June 2011.

Consent

Patients were provided with the patient information sheet (see Appendix 2) and were given sufficient time to consider participation and discuss the trial with the research staff prior to consent and enrolment. Full written informed consent was taken using the ethically approved consent form.

Research governance

The trial was carried out in accordance with the NHS Research Governance Framework37 and local NHS permission was granted by the research and development departments at each participating site prior to recruitment commencing.

Regulatory requirements

As a randomised trial of an IMP, the AZALEA trial was conducted in accordance with the European Clinical Trials Directive36 and the Medicines for Human Use (Clinical Trials) Regulations 200438 as well as ICH Good Clinical Practice guidelines. 39,40 The trial received Clinical Trial Authorisation from the MHRA on 21 July 2011 and was registered in the European Community with a EudraCT number of 2011–001093–26.

Trial registration

The trial was registered on the ClinicalTrials.gov clinical trial database with a reference number of NCT01444469.

National Institute for Health Research Clinical Research Network portfolio

The AZALEA trial was adopted on the National Institute for Health Research (NIHR) Clinical Research Network (CRN) portfolio with a UKCRN ID number of 11358. Accrual data were uploaded onto the NIHR CRN database on a monthly basis.

Summary of protocol amendments

The following amendments were made to the trial protocol following approval of the first version of the document by the REC and MHRA.

• Version 2: in addition to minor typographical clarifications to the wording of the protocol, the following changes were made: addition of a throat swab (in case a sufficient sample was not obtained from the nasal mucus and nasal swab); refinement of inclusion criteria to include FEV₁ as well as PEF as a measurement of lung function; refinement of exclusion criteria to clarify the type of antibiotic use that would be excluded; and inclusion of a statement that hospitalisation as a direct result of the asthma exacerbation was not a SAE.

• Version 3: refinement of inclusion criteria to include patients aged > 65 years with a smoking history of < 5 pack-years.

• Version 4: refinement of the eligibility criteria to include patients presenting within 48 hours (of initial presentation to medical care) with an acute deterioration of asthma control (instead of 24 hours as in the previous protocol version); recruitment extension to April 2014; and minor administrative changes.

• Version 5: protocol amendment to introduce participant reimbursements for completing study visits and returning all symptom diaries.

• Version 6: addition of an extra exclusion criterion to reflect guidelines released from the US Food and Drug Administration on the use of azithromycin.

Trial committees

Data management

Predefined data ranges were included in the eCRF, which raised automated queries if data outside the expected range were entered. In addition to the automated queries, the trial data were reviewed on a regular basis by the trial monitor to look for discrepancies and errors. In addition to the regular checks performed by the trial monitor, the trial statistician also performed a series of checks on snapshots of data to look for inconsistencies.

Risk assessment and monitoring plan

A risk assessment was performed by the ICTU quality assurance manager prior to the start of the trial. The result of the risk assessment indicated that the study was of medium risk and that 50% of trial data, 100% of consent forms and 100% of SAEs should be source verified. A monitoring plan was prepared in accordance with the risk assessment to specify the frequency of monitoring visits and amount of SDV required.

The requirements of a medium-risk trial for monitoring are:

• at least two monitoring visits to be performed in total or one to three visits per annum

• SDV should be carried out for 50% of subjects to check for eligibility, existence, drug delivery (to patients), end points and AEs

• SDV should be carried out for 100% of consent forms and SAEs

• at each monitoring visit the monitor will verify research approvals, drug accountability, regulatory documents and archiving.

Monitoring visits

A site initiation visit was performed at all participating centres. Interim monitoring visits were carried out depending on recruitment rates and close-out visits were carried out at all centres following the final follow-up visit for the last patient recruited. The monitoring visits were conducted mainly by the trial monitor.

Investigational medicinal product manufacturer

The overencapsulation of azithromycin capsules and production of matching placebo was undertaken by Sharp Clinical Services, a MHRA-licensed manufacturing unit with expertise in manufacturing and overencapsulating IMPs.

Patient and public involvement

Patient representatives were consulted during preparation of the patient information sheet. The TSC membership included a patient representative from Asthma UK who was invited to attend all TSC meetings and who was included in all relevant correspondence.

In addition, the trial manager attended the National Heart and Lung Institute at the Royal Brompton Hospital, London, on several occasions to meet the respiratory consumer group. At the group meetings an update on study progress was given and any relevant issues were discussed, including the following:

1. Patient leaflet and poster – members of the group were asked for their comments and feedback on the language and appropriateness of the patient leaflet and poster, used as tools to help introduce the study to patients.

2. Ongoing issues affecting recruitment such as whether or not the group felt that patients would be more likely to participate in the study if they were approached by a study doctor rather than a nurse (in looking at why consenting to the study was low) and whether or not introducing patient payments for visits would increase the number of visits attended (in looking at why the level of missing visits was high).

Useful feedback was received from the group and incorporated into study documents and procedures when relevant as the study progressed.

Screening

In total, 4582 patients were screened at the participating hospitals/centres over the duration of the trial. Of the 390 patients meeting the eligibility criteria, 199 were randomised to the trial as 191 declined to participate. Table 3 summarises the numbers of screened and eligible patients recruited to the trial and Table 4 summarises the reasons for non-recruitment.

Recruitment and retention

Recruitment lasted for 2.5 years, from September 2011 to April 2014. The actual recruitment period was longer than the original target of 1 year. The delays in starting the trial and continuing issues with slower recruitment were associated with the following:

• delays in opening sites to recruitment

• in the initial year of recruitment we encountered an unusually mild winter with lower than anticipated numbers of asthma exacerbations

• one of the greatest factors restricting recruitment was patients being prescribed antibiotics by doctors [in accident and emergency (A&E) departments and by general practitioners (GPs)], despite the current British Thoracic Society guidelines41 stating that antibiotics should not be routinely given to treat asthma exacerbations

• research teams at some recruiting sites were based at a different hospital to the A&E department; consequently, the research nurse and investigator were not always available to travel to the other site to recruit patients.

Because of these reasons, recruitment proved extremely challenging. To counteract this, further recruiting centres were opened, inclusion criteria were relaxed and the recruitment time was extended. However, despite all of these efforts only 199 subjects were recruited (of the original target of 380) by the medication expiry date and, in the absence of any further funding, the study had to be terminated, despite not reaching its recruitment target.

Recruitment rate

The target recruitment rate for the study was three to four patients per month per centre, based on the original 10 centres recruiting and a target recruitment figure of 380.

Accrual of patients during the whole study period is presented in Figure 4. Monthly and cumulative monthly accrual of patients is shown in Figure 5.

FIGURE 4.

Cumulative monthly accrual of patients into the AZALEA trial.

FIGURE 5.

Monthly and cumulative monthly accrual of patients into the AZALEA trial.

Clinical characteristics

Baseline characteristics of all randomised patients were summarised by treatment group using the median and interquartile range for continuous variables and number and percentage for categorical variables (Table 5). To check for any differences in baseline characteristics between centres, the same variables for centres that recruited at least 10 patients were compared (Table 6). Baseline characteristics were well balanced across treatment arms and centres.

Biological samples

Sputum bacteriology results by treatment arm are shown in Tables 7 and 8. Sputum virology results are shown in Table 9 and nasal virology results are shown in Table 10. The joint sputum and nasal swab virology results are shown in Table 11.

Pulmonary function tests

Pulmonary function test results at baseline (visit 1) are shown in Table 12 by treatment arm and in Table 13 for the centres that recruited ≥ 10 patients.

Missing diary cards

Figure 7 shows the extent of the missingness for the diary cards. Day 1 was defined as the day of administration of the study drug (see Chapter 3). The highest level of missingness was observed on day 10 (30%), although the second highest was observed on day 1 (26%).

FIGURE 7.

Number and percentage of missing diary cards by day (n = 199).

A breakdown of diary card missingness for centres with at least 10 recruited patients is shown in Table 16. A relatively high rate of missingness was observed in the largest centre (Queen Alexandra Hospital, Portsmouth), where 41% of the diary cards were missing on the last 2 days.

Table 17 shows the most common patterns of diary card missingness. Regardless of whether day 1 was defined as the day of randomisation or the day of administration of the study drug, 10% of the patients did not complete their diary card on the first day. Apart from that, the missingness of the diary score records can be considered as standard dropout.

We identified two frequent types of missingness in our study data, related to the time of the study drug admission. Some patients (potentially 17) could have received their study drug late in the day and so they started to complete the diary cards on the following day, which means that their first completed diary card was for day 2 and their last diary card was for day 11 by definition. There were some patients (potentially 13) who could have completed their first diary card by assessing their status on the previous day and so they completed diary cards for days 0–9 by definition.

There were also 23 patients who were randomised but who did not return any diary cards.

Missing asthma questionnaires

Asthma questionnaires were missing if a patient did not attend visits 2 or 3. In addition, there were 18 missing overall Acute AQLQ scores (three at visit 1, seven at visit 2 and eight at visit 3) and 18 missing overall Mini AQLQ scores (three at visit 1, seven at visit 2 and eight at visit 3) because one or more items were missing. The missingness was balanced between the treatment arms.

Missing pulmonary function test results

Results from the pulmonary function tests were missing for unattended visits; in addition, some test results were missing for other patients. There were complete results for 160, 150 and 142 patients at visits 1, 2 and 3, respectively (out of a possible 199, 171 and 163 patients who attended visits 1, 2 and 3, respectively). Most of the missing results were for FEF_25−75% and FEF_50%, which were missing in five cases out of the six recruited patients at Rowden Medical Partnership (GP surgery), Chippenham, and in five cases out of the eight recruited patients at New Cross Hospital, Wolverhampton.

Exploratory analysis of the primary outcome

As a check for outliers and imbalances, a series of longitudinal plots (one for each centre) of diary score for each patient, differentiating between treatment arms, were produced (see Appendix 1). Figure 8 shows the mean diary score and standard error (SE) for each treatment arm on each day. Box plots of diary scores by treatment arm for each day were also produced to show the distribution of the observed scores graphically (Figure 9). Table 18 shows the observed mean diary scores and SDs for each treatment arm by day and the numbers of observations. Additionally, a table of summary statistics of the diary scores by day and treatment arm was produced, including the numbers of observations, means, SDs, medians and lower and upper quartiles (Table 19).

FIGURE 8.

Observed symptom diary scores by day (with SEs). Reproduced with permission from JAMA Internal Medicine. 2016. http://dx.doi.org/10.1001/jamainternmed.2016.5664. Copyright © 2016 American Medical Association. All rights reserved.

FIGURE 9.

Box plots of observed symptom diary scores. Vertical bars represent the highest/lowest value within 1.5× the interquartile range from the upper/lower quartile. Closed circle represents value(s) outside 1.5× the interquartile range from the upper/lower quartile. Reproduced with permission from JAMA Internal Medicine. 2016. http://dx.doi.org/10.1001/jamainternmed.2016.5664. Copyright © 2016 American Medical Association. All rights reserved.

Results

A linear change was assumed in the model for the diary score over time, with different slopes for the two treatment arms. Additionally, equal mean scores were assumed at baseline for the two groups as any inequality could have occurred only by chance, because of randomisation. To reduce bias caused by the observed difference at baseline, the main effect of the interaction term was not included in the model as an independent covariate. Sensitivity analysis with the inclusion of this covariate was conducted. The estimated mean diary score at baseline (day 1) in the whole study population was 3.66 (95% CI 3.41 to 3.90). The decrease in diary score observed in the azithromycin group was slightly greater than the decrease in diary score observed in the placebo group. On average the difference in change in diary scores from baseline for the azithromycin group compared with the placebo group was –0.018 per day (95% CI –0.074 to 0.037). The estimated differences with their 95% CIs for each day are provided in Table 20. The mean ‘natural’ background daily decrease in diary score (decrease in the placebo group) was –0.18 (95% CI for the first day alone –0.22 to 0.14). On day 10, the difference between the two groups was not statistically significant, with the estimated mean diary score being lower in the azithromycin group by –0.166 (95% CI –0.670 to 0.337). On day 5 the difference between the groups was –0.074 (95% CI –0.298 to 0.150).

Acute Asthma Quality of Life Questionnaire and Mini Asthma Quality of Life Questionnaire analysis

Figure 10 shows the mean Acute AQLQ scores and SEs for each treatment arm at each visit. Box plots of observed Acute AQLQ scores by treatment arm for each visit are shown in Figure 11.

FIGURE 10.

Observed mean Acute AQLQ scores and SEs by treatment arm and visit. Reproduced with permission from JAMA Internal Medicine. 2016. http://dx.doi.org/10.1001/jamainternmed.2016.5664. Copyright © 2016 American Medical Association. All rights reserved.

FIGURE 11.

Box plots of observed Acute AQLQ scores by treatment arm and visit. Vertical bars represent the highest/lowest value within 1.5× the interquartile range from the upper/lower quartile. Closed circle represents value(s) outside 1.5× the interquartile range from the upper/lower quartile (circles for active group, diamonds for placebo group). Reproduced with permission from JAMA Internal Medicine. 2016. http://dx.doi.org/10.1001/jamainternmed.2016.5664. Copyright © 2016 American Medical Association. All rights reserved.

Table 21 shows the statistics of the Acute AQLQ scores for each treatment arm by visit and the numbers of observations.

As for the primary outcome, multilevel modelling was carried out assuming equal mean scores at baseline and a linear change for the Acute AQLQ and Mini AQLQ scores over time, with different slopes for the two treatment arms. Differences in the change in Acute AQLQ scores from baseline for each visit for the azithromycin group compared with the placebo group are provided in Table 22. At visit 3 (day 10) there was no statistically significant difference between the two groups. According to the model, at visit 3 the AQLQ score was higher by 0.130 (95% CI –0.276 to 0.539) in the azithromycin group than in the placebo group. Details of the model can be found in Appendix 1.

The same analyses that were conducted for the Acute AQLQ scores were conducted for the Mini AQLQ scores. Figure 12 shows the mean Mini AQLQ scores and SEs for each treatment arm and visit. Box plots of Mini AQLQ scores by treatment arm and visit are shown in Figure 13. Table 23 shows the statistics of the Mini AQLQ scores for each treatment arm and visit.

FIGURE 12.

Observed mean Mini AQLQ scores and SEs by treatment arm and visit. Reproduced with permission from JAMA Internal Medicine. 2016. http://dx.doi.org/10.1001/jamainternmed.2016.5664. Copyright © 2016 American Medical Association. All rights reserved.

FIGURE 13.

Box plots of observed Mini AQLQ scores by treatment arm and visit. Vertical bars represent the highest/lowest value within 1.5× the interquartile range from the upper/lower quartile. Closed circle represents value(s) outside 1.5× the interquartile range from the upper/lower quartile (circles for active group, diamonds for placebo group). Reproduced with permission from JAMA Internal Medicine. 2016. http://dx.doi.org/10.1001/jamainternmed.2016.5664. Copyright © 2016 American Medical Association. All rights reserved.

Differences in the change in Mini AQLQ scores from baseline for each visit with 95% CIs for azithromycin compared with placebo are shown in Table 24. At visit 3 (day 10) there was no statistically significant difference between the two groups. According to the model, at visit 3 the Mini AQLQ score was lower by –0.042 (95% CI –0.409 to 0.325) in the azithromycin group than in the placebo group. Details of the model can be found in Appendix 1.

Pulmonary function test analysis

For the pulmonary function tests, similar exploratory analyses and multilevel modelling were conducted as for the AQLQ scores. Details of the models used can be found in Appendix 1. Table 25 shows the observed pulmonary function test values (mean and SE) for each visit by treatment arm.

Table 26 shows the estimated differences in change in pulmonary function from baseline with 95% CIs for azithromycin compared with placebo.

Time to a 50% reduction in symptom score analysis

Time from day 1 to a 50% reduction in the initial symptom score was calculated using the mean daytime scores from the diary cards. This variable was calculated as the number of days until the first occasion on which the score was at least 50% lower than the day 1 score (it is possible for this score to come down but then to increase above 50% again). Patients who reached the 50% reduction were considered to have had this event; all other patients were treated as censored at the day of their last recorded diary score (usually day 10 but earlier for patients with missing diary card scores). The observed median time to a 50% reduction in symptom score was 7 days in the placebo group and 6 days in the active group.

Kaplan–Meier curves of time to a 50% reduction in symptom score for each treatment arm (truncated at 10 days) are shown in Figure 14.

FIGURE 14.

Kaplan–Meier curves of time to a 50% reduction in symptom score for each treatment arm (truncated at 10 days). Reproduced with permission from JAMA Internal Medicine. 2016. http://dx.doi.org/10.1001/jamainternmed.2016.5664. Copyright © 2016 American Medical Association. All rights reserved.

Linear model with random intercept and slope

FIGURE 20.

Observed and fitted values for randomly selected examples.

FIGURE 21.

Observed and fitted values for randomly selected examples.

FIGURE 22.

Residuals by day for linear model with random intercept and slope.

FIGURE 23.

Level 2 residuals (slope) for linear model with random intercept and slope (by definition it is constant over time).

FIGURE 24.

Level 2 residuals (intercept) for linear model with random intercept and slope (by definition it is constant over time).

Quadratic model with random intercept and slope

FIGURE 25.

Observed and fitted values for randomly selected examples.

FIGURE 26.

Observed and fitted values for randomly selected examples.

FIGURE 27.

Residuals by day for quadratic model with random intercept and slope.

FIGURE 28.

Level 2 residuals (slope) for quadratic model with random intercept and slope (by definition it is constant over time).

FIGURE 29.

Level 2 residuals (intercept) for quadratic model with random intercept and slope (by definition it is constant over time).

Square root model

FIGURE 30.

Observed and fitted values for randomly selected examples.

FIGURE 31.

Observed and fitted values for randomly selected examples.

FIGURE 32.

Residuals by day for quadratic model with random intercept and slope.

FIGURE 33.

Level 2 residuals (slope) for square root model with random intercept and slope (by definition it is constant over time).

FIGURE 34.

Level 2 residuals (intercept) for square root model with random intercept and slope (by definition it is constant over time).

Splines (at day 3 and 7)

FIGURE 35.

Observed and fitted values for randomly selected examples.

FIGURE 36.

Observed and fitted values for randomly selected examples.

FIGURE 37.

Residuals by day for splines model with random intercept and slope.

What is the study about?

Acute attacks (exacerbations) of asthma are common and cause a great deal of suffering in asthmatic patients. Current treatments for asthma attacks are not completely effective and new and better treatments are needed. Viruses often cause asthma attacks and bacterial lung infections have also been associated with asthma attacks. However, the role for bacteria is uncertain. Current asthma guidelines for doctors treating asthma exacerbations do not recommend the routine use of antibiotics. We would like to investigate whether or not azithromycin, which is a safe and well-tolerated antibiotic (an antibacterial) that has been used for many years in the treatment of respiratory disease, might be of benefit in asthma attacks. As there is some evidence that azithromycin has antiviral properties this may add to its benefits (antibiotics do not usually affect viruses). By looking at the effect of azithromycin on asthma attacks this will help us to show whether or not azithromycin should be recommended during an acute asthma attack in addition to the usual care that is provided to these patients as it may help them recover quicker from the exacerbation. We will also be able to look at why azithromycin may be effective – if it is having an antibacterial and/or antiviral effect.

Why have I been invited?

You have been invited to take part in this research study because you have presented with an exacerbation of your asthma, which is the condition we are looking at. We are planning to enrol 380 participants into this study who present to medical care with an asthma exacerbation across different AZALEA research sites across the UK.

Do I have to take part?

It is up to you to decide to join the study. We will describe the study and go through this information sheet. If you agree to take part, we will then ask you to sign a consent form. You are free to withdraw at any time, without giving a reason, and this would not affect the standard of care you receive.

What will happen to me if I agree to take part?

In addition to your current visit we will ask you to attend an additional three visits during which we will take some samples from you (see below for details) and ask you to complete some questionnaires. These visits will be at 5 and 10 days from your current visit and then a final follow-up visit in 6 weeks time; the total time you will be in the research will be 6 weeks.

The study will involve a brief interview and medical examination to find out if you are suitable for the study. A member of the research team will review your medical records and ask some questions about you and any medicines you might be taking. These data will be entered into a central study database and be given a code so that you cannot be identified from the information in the database. Only the study team will have access to identifying information about you and this will be kept confidential as described below.

A summary of the research procedures at each of the study visits is shown in the table below. The research procedures that will occur as part of your current visit should take no longer than 2 hours and all subsequent visits will take around 1 hour.

Study procedures

What are the alternatives for diagnosis or treatment?

The alternative to participating in this research is to continue through the routine process for treatment of your exacerbation and not participate in the study or take the study drug/placebo.

Are there any disadvantages to taking part in this study?

The disadvantages of taking part in this study are those associated with collecting the samples as well as the time inconvenience to you of attending the AZALEA research site for further visits. We will try to arrange the time of your study visits to be convenient for you.

What are the side effects of the study drug?

• Azithromycin is a licensed antibiotic with a good safety record that has been widely used for a long time. Side effects and allergic reactions are rare. The most common side effects that you may experience are gastrointestinal, that is, nausea, diarrhoea, vomiting and abdominal pain/cramps.

• There is a low risk of cardiovascular side effects. In rare incidences azithromycin can cause abnormal changes in the electrical activity of the heart, particularly in patients who already have abnormal heart rhythms (arrhythmias) or who are taking medication for these conditions. These patients will therefore be excluded from taking part in the study.

• If you suspect that you have these or any other symptoms you must inform the study doctor. If you experience severe side effects or an allergic reaction please contact the study doctor immediately on XXXX or attend A&E or phone for an ambulance if you are concerned.

• If you have reacted badly in the past to azithromycin or related antibiotics and their ingredients then you will not be eligible to participate in this study. Similarly, if you are taking any other medications or have other disease conditions or infections that could interfere with the way that the drug works or with our monitoring of how it works then you will not be included in the study.

Are there any benefits to taking part in this study?

You may benefit from this study if you are randomised to receive azithromycin and it is found that this is effective in improving recovery from an asthma exacerbation. However, we do not know whether or not it is effective so we cannot promise that the study will help you. Although this study will not necessarily benefit you directly, it may help us to understand more about the best ways of treating asthma exacerbations in future patients with this condition.

What happens when the research study stops?

You will return to the care of your local doctor (GP) and the hospital doctors you would see routinely.

What if there is a problem?

Any complaint about the way you have been dealt with during the study or any possible harm you might suffer will be addressed. The detailed information on this is given in Part 2.

Will my taking part in the study be kept confidential?

Yes. The details are included in Part 2.

This completes Part 1.

If the information in Part 1 has interested you and you are considering participation, please read the additional information in Part 2 before making any decision.

What if relevant new information becomes available?

Sometimes new information about a treatment being used in a study becomes available during the course of the study. If this happens, the study doctor will let you know. You can then make a decision about whether you want to continue in the study. If you decide not to continue in the study, the study doctor will ensure that your ‘normal’ care continues. If you decide to continue in the study, you will be asked to sign an updated consent form.

After receiving new information, the study doctor might want you to withdraw from the study. He may feel it is in your best interest to do this. If this happens, the doctor will explain the reason for this and arrange for your ‘normal’ care to continue. Your participation in the study may be stopped for any of the following reasons:

• failure to comply with the study instructions

• a serious reaction, which may require treatment or observation

• the doctor decides it is in the best interest of your health and welfare to discontinue.

What will happen to the samples I give?

• The above research samples that we take from you will be used to find out the following information:

  • whether azithromycin treatment is effective during an acute exacerbation of asthma

  • the type of viral and/or bacterial infection present in participants during an exacerbation

  • why azithromycin might be an effective treatment, that is, how it works

  • whether there are ‘markers’ in participants’ blood or spit that would be able to tell us in advance about the severity of and recovery from the acute exacerbation.

• The results from testing your research samples will not affect the care that you receive as part of your routine treatment as we will not know the best treatment until the outcome of the study.

• All of the samples will be labelled with a code with no identifying information about you. The samples will then be tested in designated laboratories but the laboratory scientists will not be able to identify them as your samples. Your identifying information will be accessible only to authorised members of the research team.

• If you agree, we would also like to keep any of your samples that are left over after our research. These would be stored and used for future ethically approved research and would be accessible only to authorised members of the research team or regulatory bodies. At any time you want you could ask for these samples to be destroyed.

Can I withdraw from participation in this study?

Your participation in this study is voluntary and you have the right to refuse to participate. You are free to withdraw at any time and do not have to give a reason for this, even after you have agreed to take part. Being part of this study will not affect your normal medical care, either now or in the future.

What will happen if I don’t want to carry on with the study?

You are free to withdraw from the study at any time, without needing to give a reason why. If you withdraw from the study, we would like to keep the samples and data collected up to your withdrawal for our analysis. Any stored blood or tissue samples that can still be identified as yours can be destroyed though if you wish.

What if something goes wrong?

Imperial College London holds insurance policies that apply to this study. If you experience serious and enduring harm or injury as a result of taking part in this study, you may be eligible to claim compensation without having to prove that Imperial College is at fault. This does not affect your legal rights to seek compensation.

If you are harmed because of someone’s negligence, then you may have grounds for a legal action. Regardless of this, if you wish to complain, or have any concerns about any aspect of the way you have been treated during the course of this study, then you should immediately inform the investigator on the contact details below. The normal National Health Service complaints mechanisms are also available to you. If you are still not satisfied with the response, you may contact the Imperial AHSC Joint Research Office.

Who is organising and funding the research?

The research is co-ordinated by Imperial College London and has been funded by the National Institute for Health Research (NIHR) and Medical Research Council. It has been reviewed and given a favourable opinion by the London – Bloomsbury Research Ethics Committee and has been given a favourable opinion by the Medicines and Healthcare products Regulatory Agency (MHRA), the agency that reviews and approves drug studies.

The study is being carried out by experts in asthma from around the country as a collaboration between NHS trusts and universities (a list of these collaborators is available to you should you wish). None of the investigators performing the research or participants taking part will benefit financially from the study.

Will the information on me be kept confidential?

Yes, all personal information will be kept confidential and secure. Only people involved in the study will have access to your personal information. When we send your samples for analysis in designated laboratories they will be labelled with a code and have no identifying information on them. This study will be published in medical journals but it will not be possible to identify you from what is written. With your permission we would also like to retain any of your samples not used in this study for future research projects.

Involvement of the general practitioner/family doctor

With your permission we will inform your GP of your participation in this study.

Expenses and payments

A maximum payment of £50 will be available to you at visit 4 if you have completed all study visits and completed and returned all 10 symptom diaries (this payment is made up of £10 for attending visit 1, £10 for visit 2, £10 for visit 3, £10 for visit 4 and £10 for returning the symptom diaries).

We will also reimburse the cost of your travel expenses for the additional three visits you will be asked to attend for research purposes.

What if I have any problems or would like further information about the study?

If you have a concern about any aspect of this study, you should ask to speak to the researchers on XXXX, who will do their best to answer your questions. If you remain unhappy and wish to complain formally, you can do this through the normal NHS complaints procedure.

Thank you for taking the time to consider participating in our research.