31 May 2011
A pragmatic randomised study found that a leukotriene receptor antagonist (LTRA) was equivalent to an inhaled corticosteroid (ICS) as first-line preventer therapy, and to a long-acting beta2 agonist (LABA) as add-on therapy to ICS, in adults with asthma. However, this study has important limitations and does not change current recommended practice.
Level of evidence:
Level 2 (limited quality patient-oriented evidence) according to the SORT criteria.
Health professionals should continue to follow the recently revised SIGN/BTS guideline on the management of asthma. For patients not adequately controlled on a short-acting beta2-agonist (SABA) when required (step 1), ICS are the first choice regular preventer therapy (step 2). An LTRA (montelukast or zafirlukast) may be considered in children under five years if an ICS cannot be used.
A proportion of patients with asthma may not be adequately controlled on an ICS alone at step 2. For adults, adolescents and children aged 5 to 12 years, the addition of a LABA (salmeterol or formoterol) to an ICS should be considered next (step 3). For children under five years, the first choice add-on therapy to an ICS is an LTRA. However, before adding or changing treatment, practitioners should check concordance with existing therapy, check the patient’s inhaler technique and eliminate trigger factors.
What is the background to this?
ICS are the first-line preventer therapy for achieving overall treatment goals. Current SIGN/BTS guidelines state that LTRAs are an alternative, but less effective preventer therapy in patients taking SABA alone. Therefore, LTRAs are recommended as first-line preventer therapy (step 2) only in children under five years who cannot use an ICS. LTRAs are also recommended as the first choice add-on therapy to ICS (step 3) in children under five years only.
In adults, adolescents and children over five years, LTRAs have a much more limited role. At step 3, if there has been no response to the addition of a LABA to regular dose ICS, the SIGN/BTS guideline recommends that the LABA is stopped and the dose of ICS is increased (to 800 micrograms/day in adults and adolescents, 400 micrograms/day in children aged 5 to 12 years). If asthma control still remains inadequate, then sequential trials of other add-on therapies (e.g. an LTRA) may be considered.
This open-label pragmatic study consisted of two parallel equivalence trials in adults and adolescents with impaired asthma-related quality of life (QoL) — the ‘first-line therapy’ trial compared LTRA with ICS as first-line preventer therapy (i.e. step 2), and the ‘add-on therapy’ trial compared LTRA with LABA as add-on therapy to ICS (i.e. step 3). The aim of this study design was to evaluate the role of LTRAs in primary care patients under ‘real-world’ conditions, rather than under ‘idealised’ conditions in randomised controlled trials (RCTs).
What does this study claim?
This study claimed that an LTRA was equivalent to ICS as first-line preventer therapy (n=306) and also to LABA as add-on therapy (n=352) for a diverse group of primary care patients (aged 12 to 80 years). The pre-defined equivalence criteria were met at two months (based on Mini Asthma QoL Questionnaire [miniAQLQ] scores), but not at two years. There were no significant differences in asthma exacerbation rates and Asthma Control Questionnaire (ACQ) scores. See ‘study details’ below for more information.
How does this relate to other studies?
ICS vs. LTRA as first-line preventer therapy:
A Cochrane review of 27 RCTs in mainly adults with mild to moderate persistent asthma concluded that ICS (400 micrograms/day beclometasone or equivalent) was more effective than LTRAs given in the usual licensed doses. Another meta-analysis of 18 RCTs in children and adolescents (<18 years) with mild to moderate asthma also found that ICS was more effective than montelukast for preventing severe asthma exacerbations (requiring oral steroids).
LABA vs. LTRA as add-on therapy to ICS:
A 2011 Cochrane review of 17 RCTs concluded that in adults with asthma that is inadequately controlled on low dose ICS (and showing significant reversibility with beta2-agonists), addition of a LABA was superior to LTRA in reducing asthma exacerbations (requiring oral steroids). There was some suggestion of a possible increased risk of serious adverse events with a LABA, compared with LTRA.
An accompanying editorial suggests that the results of this study challenge current guideline recommendations. There appeared to be little difference in QoL scores, although the absence of a placebo group means that it is difficult to judge whether the changes observed in QoL in either group from baseline were clinically significant. When this pragmatic study is set in the context of the existing evidence for LTRAs, these new findings should not change current practice, especially in view of its limitations.
Meta-analyses of double-blind RCTs that measure patient-oriented outcomes provide the best available evidence. This evidence supports recommendations in the SIGN/BTS asthma guideline that ICS are the first-choice preventer therapy for all people with asthma, and LABAs are the first-choice add-on therapy to ICS for people five years and over. LTRAs have an earlier role only in children under five years — at step 2, only if an ICS cannot be used. At step 3, they are first-choice add-on therapy to an ICS, instead of a LABA. However, data are limited in children and the findings of this new study apply only to adults, as few patients were younger than 25 years.
The aim of this pragmatic study to allow evaluation in a ‘real world’ setting was worthwhile, and may be more representative of the patient populations in clinical practice than RCTs. However, as the authors acknowledged, as a pragmatic study allows usual practice to occur, the interpretation of the results may be limited by crossover between treatment groups and lack of placebo group. For example, treatment changes at two years were made for more patients in the LTRA groups in both trials (first-line therapy: 31.0% vs. 20.6%; add-on therapy: 25.4% vs. 0%). Furthermore, this small (n=658), open-label study was powered for equivalence in the MiniAQLQ score — an important patient-oriented outcome. However, the study was not powered to demonstrate equivalence in exacerbation frequency, another important patient-oriented outcome frequently reported in RCTs. Due to a lack of exacerbations overall, the number of patients was unlikely to be sufficient to demonstrate true equivalence in exacerbation frequency. An NEJM article discusses further the pros (e.g. consistency with usual clinical care) and cons (e.g. unblinded treatment, patient self-assessment) of this trial, and pragmatic trials in general.
Adherence rates in this study (measured by prescriptions issued) were low, but appeared to be better in the LTRA groups (first-line therapy: 65% in LTRA group vs. 41% in ICS group; add-on therapy: 74% in LTRA group vs. 46% in LABA group). Adherence may have been improved if the ICS and LABA had been given in a combination inhaler. A pragmatic equivalence trial with a substantial rate of non-adherence may not demonstrate equivalence robustly and may not reflect the maximal effect of treatments. However it does highlight the SIGN/BTS recommendation to check concordance with existing therapy, check the patient’s inhaler technique and eliminate trigger factors, before considering any step up treatment. Furthermore, stepping down therapy once asthma is controlled is recommended, but often not implemented, leaving some patients over-treated.
Open label, randomised, pragmatic equivalence study in 53 UK primary care practices. Patients recruited from May 2002 to February 2005.
658 patients aged 12 to 80 years (few patients <25 years) with asthma diagnosis. In the first-line therapy trial, patients had asthma symptoms that required initiation of preventer therapy as assessed by physician. In the add-on therapy trial, patients had received ICS for >12 weeks and had symptoms that required an increase in therapy.
Intervention and comparison
First-line therapy trial — LTRA (montelukast [89%] or zafirlukast [11%]) vs. ICS (beclometasone [93%], budesonide [5%] or fluticasone [2%]). Add-on therapy trial — ICS plus LTRA (montelukast [95%] or zafirlukast [5%]) vs. ICS plus LABA (salmeterol [92%] or formoterol [8%]). Participants obtained their own medication.
|First-line therapy trial||LTRA mean score (no. of patients)||ICS mean score(no. of patients)||Adjusted mean difference (95%CI)LTRA vs. ICS|
|ITT MiniAQLQ||At 2 months||5.25 + 1.03 (122)||5.28 + 1.10 (132)||–0.02 (–0.24 to 0.20)|
|At 2 years||5.52 + 1.07 (145)||5.63 + 1.16 (155)||–0.11 (–0.35 to 0.13)|
|PP MiniAQLQ||At 2 months||5.26 + 1.02 (115)||5.35 + 1.04 (127)||–0.08 (–0.31 to 0.15)|
|At 2 years||5.61 + 1.03 (98)||5.65 + 1.16 (120)||–0.12 (–0.38 to 0.14)|
|Add-on therapy trial||LTRA mean score (no. of patients)||LABA mean score(no. of patients)||Adjusted mean difference (95%CI)LTRA vs. LABA|
|ITT MiniAQLQ||At 2 months||5.09 + 1.15 (153)||5.04 + 1.11 (156)||–0.10 (–0.29 to 0.10)|
|At 2 years||5.43 + 1.14 (169)||5.42 + 1.08 (181)||–0.11 (–0.32 to 0.11)|
|PP MiniAQLQ||At 2 months||5.11 + 1.15 (147)||5.04 + 1.11 (156)||–0.09 (–0.29 to 0.11)|
|At 2 years||5.59 + 1.04 (121)||5.44 + 1.08 (176)||–0.01 (–0.21 to 0.24)|
MiniAQLQ scores range from 1 to 7, with higher scores representing better quality of life. PP – per protocol analysis i.e. patients for whom the randomly assigned treatment was not changed. Plus-minus values are means + standard deviation. Red text – equivalence not proved.
Secondary outcomes included ACQ score, Royal College of Physicians 3-item asthma questionnaire, frequency of asthma exacerbations (requiring oral steroids or hospitalisation for asthma).
Sponsorship Study was funded by the National Coordinating Centre for Health Technology Assessment UK and others.
Make sure you are signed up to NPC Email updates — the free email alerting system that keeps you up to date with the NPC news and outputs relevant to you