Roflumilast and Exacerbations in patients receiving Appropriate Combination Therapy
REACT trial design and results
Design
Double-blind, placebo-controlled, phase 3-4 RCT at 203 centres in 21 countries
Treatment
Roflumilast 500 microg orally once daily, added to inhaled steroid and LABA
Control
Placebo, added to inhaled steroid and LABA
Population
1945 patients with severe COPD, chronic bronchitis and at least two exacerbations in the previous year
Follow-up
1 year
Primary endpoint
Rate of moderate-to-severe COPD exacerbations per patient per year
Result
0.805 with roflumilast versus 0.927 with placebo — 13.2% lower, but rate ratio 0.868 (95% CI 0.753 to 1.002, p=0.0529) on the prespecified Poisson analysis. The predefined negative binomial sensitivity analysis gave 0.823 versus 0.959 (rate ratio 0.858, 0.740 to 0.995, p=0.0424). Adverse events led to withdrawal in 11% on roflumilast versus 5% on placebo.
Our findings suggest that roflumilast reduces exacerbations and hospital admissions in patients with severe chronic obstructive pulmonary disease and chronic bronchitis who are at risk of frequent and severe exacerbations despite inhaled corticosteroid and longacting beta2 agonist therapy, even in combination with tiotropium.
The trial authors, in the published abstract
How it has aged
A trial whose interpretation depends entirely on which prespecified analysis you privilege — p=0.0529 or p=0.0424 for the same data. It supports roflumilast as an add-on in the narrow group it enrolled: severe COPD with chronic bronchitis and frequent exacerbations despite inhaled therapy. Tolerability, not efficacy, is what limits its use.
Publications
Effect of roflumilast on exacerbations in patients with severe chronic obstructive pulmonary disease uncontrolled by combination therapy (REACT): a multicentre randomised controlled trial
Martinez FJ, Calverley PM, Goehring UM, et al. Effect of roflumilast on exacerbations in patients with severe chronic obstructive pulmonary disease uncontrolled by combination therapy (REACT): a multicentre randomised controlled trial. Lancet 2015 Mar 7;385(9971):857-66.
13.2% lower exacerbation rate, not significant on the primary Poisson analysis (p=0.0529)
14.2% lower and statistically significant on the predefined negative binomial analysis (p=0.0424)
Twice the rate of withdrawal for adverse events compared with placebo