Article Review Report
Kornmann O, Dahl R, Centanni S, Dogra A, Owen R, Lassen C, Kramer B. Once-daily indacaterol versus twice-daily salmeterol for COPD: a placebo-controlled comparison. European Respiratory Journal 2011; 37(2):273-279.
Why are you looking at this journal?
The European Respiratory Journal is a credible source of information because it is peer-reviewed. It is an official publication of the European Respiratory Society. My interest in the journal lies in the focus of the articles it publishes. The journal publishes experimental and clinical work on all aspects of respiratory medicine including the biology of lung cells, pathophysiology, epidemiology, immunology, thoracic imaging, thoracic surgery, occupational medicine, pediatric pneumology, intensive care, and sleep medicine.
The journal has an editorial board. Its current chief editor is Marc Humbert. The editorial board/ list of reviewers reviews the accuracy of information contained in all articles to be published in the journal prior to publishing. This process is important since it ensures that only articles that meet pre-specified criteria in terms of quality and adherence to research ethics are published.
I am interested in this article because it deals with the management of COPD, a disease that affects an approximated 40% of the world population and whose prevalence is projected to increase over the next several decades. The article contains important information on the current management of COPD as well as ongoing research on long-acting beta adrenergic and anticholinergic bronchodilators. In particular, my interest in the topic of the article is spurred by the possibility of a once-daily dose of indacaterol for COPD patients from the current recommended twice daily regimen of salmeterol. This information is useful to my future practice hence the article will help me keep up to date with current happenings in my field.
The article is entitled “Once-daily indacaterol versus twice-daily salmeterol for COPD: a placebo-controlled comparison.” Both indacaterol and salmeterol are long-acting beta adrenergic agonists used to promote bronchodilatation in patients with COPD. Long-acting bronchodilators are recommended for use in the management of COPD by current treatment guidelines. Salmeterol is given twice daily while indacaterol is given once daily. Its efficacy and safety profile are being compared with those of salmeterol in the current study in order to determine whether COPD patients can be managed using once-daily indacaterol instead of twice-daily salmeterol. Therefore, the findings of this study and other similar studies on the issue are likely to influence the approach to the management of COPD adopted in future treatment guidelines and in effect, they will impact on both my current and future practice.
The study has seven authors. Kornmann is affiliated with the Pulmonary Division of Internal Medicine of the Mainz University Hospital, Germany. Dahl is affiliated with the department of respiratory diseases of the Aarhus University Hospital, Denmark. Centanni is affiliated with the department of pneumology of the Degli University, Italy. Owen and Lassen are affiliated with the department of respiratory development, Novartis Pharmaceuticals Corporation, USA. Kramer is affiliated with the Novartis Horsham Research Centre, Horsham, UK.
The abstract to the article is not divided into its individual components but is provided in continuous prose. Readers are left to discern its components for themselves. Regardless of this though, the abstract contains the following sections, the background, aims, methods, results, and conclusions.
The background section of the study highlights the magnitude of the problem of COPD noting that the disease affects an estimated 40% of the world population aged over 40 years and its prevalence is expected to increase in coming years. It also denotes the need to evaluate the efficacy and safety of indacaterol, an ultra-long-acting inhaled β2-agonist bronchodilator in relation to the efficacy and safety profiles of other long-acting bronchodilators. Indacaterol had just been approved for use as maintenance treatment for COPD patients by the European Union at the time the study was being conducted. The purpose of the study, therefore, was to investigate and compare the efficacy and safety of indacaterol with that of placebo and twice-daily salmeterol over a period of 6 months. The latter was an active control. The authors did not explicitly state any research hypothesis or research questions although the latter can be deciphered deductively from the background section. For instance, one research question can state, “In patients with COPD, is a once-daily dose of indacaterol more effective in improving 24-hour trough FEV1 than a twice-daily dose of salmeterol or placebo?”
The study was a randomized double-blind clinical trial whereby patients were randomized in a double-blind fashion in a 1:1:1 ratio to receive either once daily 150 µg indacaterol via inhaler in the morning, or 50 µg salmeterol taken twice daily (morning and evening) also via inhaler, or placebos matching both active treatments for 26 weeks. Randomization was done after a 2-week run-in period whereby concomitant medications were stabilized and baseline variables determined.
The subjects in the study were male and female patients who had been diagnosed with moderate-to-severe COPD. For the patients to be included in the study, they must have been >40 years old, have had a smoking history of more than 20 pack-years, their spirometry test results measured within 30 minutes of inhaling 400 µg salbutamol at the time of screening must have been as follows: FEV1 ≥30% but ≤80% predicted and the ratio of FEV1 to forced vital capacity (FVC) <0.7. Patients with a concomitant history of asthma were not included in the study.
In relation to the approval of the study, we know that this specific study was approved by the relevant institutional review boards from the statements of the authors who state that permission to conduct the study was sought from the institutional review boards/ ethics committees of all participating centers.
A mixed-model ANCOVA was used to analyse the study variables. In this model, treatment was treated as a fixed-effect, appropriate baseline measurement and baseline FEV1 reversibility were treated as covariates, the country and smoking status were fixed effects, and the centre as a random effect. The primary and the important secondary efficacy variables were, on account of multiplicity considerations, analyzed in a hierarchical fashion. The mixed-model ANCOVA test was appropriate for this study because the study had both fixed and random effects. The levels of the fixed effects, mostly related to treatment effects, were of primary interest to the study whilst the levels of the random effects (cities of study) were not related to the primary interest of the study. Therefore, the mixed model permitted explicit modeling of a wide variety of correlation patterns that is, for both lower and upper levels of the units of the random and fixed effects. Another reason why the mixed-model approach was suitable for this study is that the study entailed the conduct of multiple/repeated measurements on each subject over time, an aspect which would have led to the emergence of multiple correlation patterns and errors which cannot be modeled using univariate and multivariate variance-covariance structures. In addition, the mixed-model approach allows the data of subjects who drop out of a study to be incorporated in the final analysis, can handle unequal spacing of repeated measurements, and the findings of the mixed model analysis are more interpretable than those of classical repeated measures.
The differences between the active treatments and placebo for 24-hour trough FEV1 were significant for the two active treatments, indacaterol and salmeterol, at day 2 and at weeks 12 and 26 (p<0.001). The trough FEV1 was significantly greater for indacaterol than salmeterol at weeks 12 and 26 by 60 ml and 70 ml respectively (both p<0.001). The increase in trough FEV1 from baseline for indacaterol over placebo was clinically significant throughout the course of the study; at week 12, an increase of 170 ml was noted from the 130 ml noted at 24 hours following administration of the first dose (day 2), and an increase to 180 ml at the 26th week. The increase from baseline trough FEV1 was smaller and not clinically significant for both salmeterol and placebo. The treatment-induced increase in FEV1 over placebo 5 minutes after administration of the first dose on day 1 was significant for both indacaterol (110 ml) and salmaterol (60 ml) (p<0.001 for both drugs over placebo); indacaterol had an advantage over salmeterol of 50 ml(p<0.01) at this time-point and this trend was maintained throughout the study.
In relation to the health status, use of as-needed salbutamol, and symptoms, the unadjusted mean of the St. George’s Respiratory Questionnaire (SGRQ) total score decreased with indacaterol from baseline by > 4-point minimum clinically important differences during all the visits. The adjusted average SGRQ total score was also significantly lower with indacaterol than placebo and salmeterol (p<0.01) throughout the study. Importantly, at week 12 which had been specified as the time-point for SGRQ to be an important secondary variable, the difference between the two active treatments, indacaterol and salmeterol was significant (p<0.05). The adjusted mean total score was higher for the transition dyspnea index (TDI) than placebo with both indacaterol and (p<0.001) and salmeterol (p<0.05) at all visits. The mean TDI differences versus placebo were significantly bigger with indacaterol than salmeterol at week 4 (0.95 versus 0.55; p<0.05) and week 12 (1.45 versus 0.90; p< 0.05). The proportion of patients with clinically important differences in the total score for TDI (≥1 units) were 56.6-60.5% with indacaterol, 48.7-53.6% with salmeterol, and 39.5-45.7% with placebo. The OR (odds ratio) of the likelihood of attaining this improvement in TDI was significant for indacaterol over placebo at weeks 4, 8, 12, and 26 (2.26, 1.71, 2.79, and 1.87 respectively; all p<0.001). The OR for salmeterol versus placebo, on the other hand, was significant only at week 12 and week 26 (2.13 and 1.90; p≤ 0.001).
A number of parameters for the various variables were designated as pre-defined minimally clinically important differences. These points were determined from similar previous studies. For the SGRQ, the MCID was 4 points in the SGRQ total score. For the TDI, the MCID was a change of 1 point whilst the composite measure for ‘days of poor COPD control’ was defined as days during which a score of ≥2 on a scale of 0-3 for a 2 or more of the following symptoms, cough, wheeze, breathlessness, and production/color of sputum was recorded. The clinically notable differences for laboratory values were designated as serum potassium <3.0 mmol/L and increased blood glucose >9.99 mmol/L. Based on these parameters, the clinically significant differences notable from the study findings are as follows: the unadjusted mean of the SGRQ total score for indacaterol which decreased from baseline by > 4-point minimum clinically important differences during all the visits and the proportion of patients with clinically important differences in the total score for TDI who were 56.6-60.5% with indacaterol, 48.7-53.6% with salmeterol, and 39.5-45.7% with placebo. The percentage of patients with clinically notable blood glucose values was 5.8% for indacaterol, 9.0% for salmeterol, and 6.3% for the placebo group and clinically notable serum potassium values were 0.3% for indacaterol, 0.6% for salmeterol, and 0% for placebos.
The pre-designated value for statistical significance for this study was 5% with 90% power. This implies that the probability for the differences between groups for this study to be due to chance alone was 0.05 hence any values whose p was 0.05 or less was considered statistically significant. The level of significance was computed based on means and standard deviations obtained in previous studies. It is explicit from the methods section that a power study was performed. This is because the authors describe the formula used to determine the sample size that would have enabled them to obtain any statistically significant results. A power study is important because the use of too small or too large sample sizes tends to distort the study findings; a small sample does not permit detection of significant differences whilst a large sample may lead to detection of statistically significant but clinically non-significant results.
Discussion and conclusion sections:
Similar to the studies that discovered that long-acting B2-agonists are more effective than short-acting B2-agonists in the management of COPD, the study established that a once-daily dose of indacaterol is more effective than a twice daily dose of salmeterol. In addition, the effects of salmetrol of trough FEV1 noted in the current study were similar to those of past studies, the additional efficacy margin of 50-60 ml of indacaterol over salmeterol was similar to that of tiotropium over salmeterol, effects of indacaterol on morning lung function were similar to those of combined bronchodilator treatment, effects on indacaterol on other end-points such as health status and dyspnoea were similar to those observed in similar past studies, occurrence of cough post-inhalation of indacaterol and the safety and tolerability profile of indacaterol established in the current study were similar to those of past studies.
On differences, the effects of salmeterol on health status and dyspnoea noted in the current study were larger than those observed in past salmeterol studies on COPD patients. In addition, the incidence of viral and bacterial upper respiratory infections in the indacaterol and placebo groups was lesser than the incidence of the same infections noted in a previous one-year study that had used the same treatment.
In regards to the limitations of the study: the authors felt that the emphasis on FEV1 as the primary end-point of the study may have inadvarntently mitigated the power of the study to examine the effects of indacaterol on the other end-points. The study did not also have adequate power to detect the statistically significant but small reductions in “days of poor COPD control” that might have occurred with either indacaterol or salmeterol. The tool used in measuring “days of poor COPD control” was also not validated and it was prone to recall bias since it depended on accurate daily entries by subjects.
The important conclusions of the study were as follows: the findings of the current and previous studies suggest that indacaterol is more effective as a bronchodilator than salmeterol and salmeterol’s counterpart, the twice-daily dosed formoterol. The authors also concluded that indacaterol may prove in future to be as effective as the anticholinergic bronchodilator tiotropium which is given as once daily dose. The authors further concluded that indacaterol improves health status and reduces dyspnoea as compared to placebo and is better, or as effective as current bronchodilator drugs in terms of improving clinical outcomes. Lastly, the authors concluded once-daily indacaterol will become a useful additional alternative for the treatment of COPD on account of its beneficial effects such as early morning bronchodilation, sustained reductions in dyspnoea, and improvements in health status.
The unanswered question of the study is whether a once-daily dose of indacaterol is equally or more effective than a once-daily dose of tiotropium in improving 24-hour trough FEV1 in COPD patients. Therefore, future studies to investigate this phenomenon should be carried out using appropriate study designs.
The external validity of the study refers to the degree to which the conclusions of a study can be used to make accurate predictions about a population. It is determined by evaluating the sampling technique employed and the proximal similarity of the contexts of the study. In reference to the current article, the study has external validity because the sample determination criteria used permitted a fair sample to be drawn from the population hence inferences about the population can be made from the findings of the study. The authors to the study also reviewed the findings of studies conducted under similar conditions and using similar or near identical treatments. A threat to the external validity of the study, however, exists because the authors did not delineate the manner in which their sample was identified and selected from the sample frame in all the 15 study settings.
Kornmann O, Dahl R, Centanni S, Dogra A, Owen R, Lassen C, Kramer B. Once-daily indacaterol versus twice-daily salmeterol for COPD: a placebo-controlled comparison. European Respiratory Journal 2011; 37 (2):273-279. doi: 10.1183/09031936.00045810