One bronchodilator or two? Translating clinical trials into clinical practice

What inhaled maintenance treatment would you give to a newly diagnosed symptomatic chronic obstructive pulmonary disease (COPD) patient? The Global Initiative for Chronic Obstructive Lung Disease (GOLD) 2019 report recommends long-acting bronchodilator (LABD) monotherapy for initial maintenance treatment, with longacting muscarinic antagonist (LAMA)/long-acting beta agonist (LABA) combinations reserved for more highly symptomatic patients, and inhaled corticosteroid (ICS)/ LABA combinations used in patients with a history of exacerbations and higher blood eosinophil counts (a predictive biomarker of ICS effects). GOLD acknowledges the paucity of direct evidence for these recommendations, for example, from treatment-naïve or newly diagnosed COPD patients. To deal with this evidence gap, GOLD extrapolated the results of clinical trials performed in patients taking inhaled maintenance treatment(s) before the start of the study to formulate these initial treatment recommendations. The group mean data of multiple clinical trials have shown that dual bronchodilator (LAMA/LABA) combinations have greater effects on lung function, symptoms and quality of life compared to LABD monotherapies. These results underpin the GOLD 2019 recommendations to escalate treatment from an LABD monotherapy to a dual bronchodilator combination to further treat symptoms. There is wide variability between individuals in the magnitude of clinical response to LABD. Consequently, group mean clinical trial data does not predict the real-life response of every individual; not every patient in real life will experience a clinical benefit from escalation to dual bronchodilator treatment. Post hoc subgroup analysis of clinical trials has focused on individuals who were treatment naïve before the study start, in order to generate evidence regarding optimal initial pharmacological treatment. Greater benefits for dual bronchodilator combinations versus monotherapies on lung function, symptoms and quality of life have been reported. These subgroup analyses lose statistical power to demonstrate significant P-values; nevertheless, similar magnitudes of benefit were observed for treatment-naïve individuals compared to the whole population. While these results support the use of dual bronchodilator combinations as initial maintenance treatment, they do not address which individuals would benefit most (and least) from this approach. Clinical trials of dual bronchodilator combinations have not included mild COPD patients (forced expiratory volume in 1 s (FEV1) > 80% predicted), 7 so there is little evidence to guide initial maintenance treatment selection in these individuals. These patients may require a lower level of initial inhaled treatment, with many being in the GOLD A group. In a recent publication in Respirology, Muro et al. present pooled analysis results (at week 24/26) from three clinical trials that compared the dual bronchodilator indacaterol/glycopyrronium with LAMA monotherapies (tiotropium and glycopyrronium). Only individuals who were not using maintenance LABD(s) at study entry were included (sample size >300 per treatment arm). Approximately 40% of individuals were using ICS at baseline, which was continued throughout the study. This population is called ‘LABD naïve’ by the authors, and is different to the other post hoc analysis already described where patients using ICS were excluded and the population was called ‘treatment naïve’. This nuance is important, as ICS use means that the LABD-naïve population does not accurately mirror newly diagnosed patients who are not taking any inhaled treatment. However, this LABD-naïve subgroup appears to resemble the earlier stages of COPD treatment, while not being completely maintenance treatment naïve. Furthermore, these individuals are different from the rest of the trial population who had LABD treatment(s) withdrawn for the run-in period and then restarted at randomization; this scenario does not resemble clinical practice. There was a significant benefit of LAMA/LABA over LAMA monotherapy on lung function consistent with previous results. The transition dyspnoea index (TDI) and St George’s Respiratory Questionnaire (SGRQ) score are widely used patient-reported outcomes (PRO) in clinical trials. For both of these, there were greater improvements with indacaterol/glycopyrronium, but the effect sizes were small and not statistically significant when considering the confidence interval. Inconsistent results for PRO have been reported in many studies comparing dual bronchodilator combinations and LABD monotherapies, and perhaps there was insufficient statistical power in this post hoc analysis. Alternatively, it is worth considering the LABD-naïve nature of the population; a LAMA might cause a large symptom improvement in these individuals, and the numerical improvements from baseline for TDI and SGRQ with the LAMA treatments suggest that this occurred. In this situation, the PRO instruments might lack the ability to effectively measure additional effects of a second bronchodilator given simultaneously. A practical approach in clinical practice is to give one bronchodilator, assess the response and add the second later (if needed), allowing an evaluation of the benefits of each bronchodilator. The alternative approach for newly diagnosed patients is to give a dual bronchodilator combination immediately to optimize lung function and