Anti-inflammatory effects of roflumilast in chronic obstructive pulmonary disease (ROBERT): a 16-week, randomised, placebo-controlled trial

Background The clinical effects of roflumilast, a selective phosphodiesterase-4 inhibitor, are well established, but little is known about the anti-inflammatory mechanisms underlying the drug's efficacy. The aim of the ROflumilast Biopsy European Research Trial (ROBERT) was to assess the anti-inflammatory effects of roflumilast on bronchial mucosal inflammation in patients with moderate-to-severe chronic obstructive pulmonary disease (COPD) and chronic bronchitis. Methods ROBERT was a randomised, double-blind, placebo-controlled trial done at 18 sites in five countries. Eligible patients were aged 40–80 years, had COPD, and had had a chronic productive cough for 3 months in each of the two previous years. Patients also had to have a post-bronchodilator predicted FEV1 30–80% and a post-bronchodilator FEV1/forced vital capacity ratio of 70% or less. Patients entered a 6-week run-in period before being randomly assigned (1:1) via a computerised central randomisation system to roflumilast 500 μg once daily or placebo for 16 weeks, in addition to bronchodilator therapy (inhaled corticosteroids were not permitted). Randomisation was stratified by concomitant use of long-acting β agonist. Both participants and investigators were masked to group assignment. Roflumilast and placebo were supplied as identical yellow, triangular tablets. Airway inflammation was assessed by quantification of inflammatory cells in bronchial biopsy samples and induced sputum samples. The primary endpoint was the change in the number of CD8 inflammatory cells in bronchial biopsy submucosa from randomisation to week 16 in the intention-to-treat population. Changes in cell counts of additional inflammatory markers, including eosinophils, were assessed as secondary endpoints. This trial is registered with ClinicalTrials.gov, number NCT01509677, and is closed to new participants, with follow-up completed. Findings Between Jan 4, 2012, and Feb 11, 2016, 158 patients were randomly assigned: 79 to the roflumilast group, and 79 to the placebo group. At week 16, the change in the number of CD8 cells in the bronchial submucosa did not differ significantly between the roflumilast and placebo groups (treatment ratio 1·03 [95% CI 0·82–1·30]; p=0·79). However, compared with placebo, roflumilast was associated with a significant reduction in eosinophils in bronchial biopsy samples at week 16 (treatment ratio 0·53 [95% CI 0·34–0·82]; p=0·0046). Significant reductions in both absolute (p=0·0042) and differential (p=0·0086) eosinophil cell counts in induced sputum were also noted with roflumilast compared with placebo, but peripheral blood eosinophil counts were not significantly affected. We noted no other significant effects of roflumilast on bronchial mucosal inflammatory cells. The most common (ie, occurring in >5% patients) moderate adverse events were worsening of COPD (three [4%] patients in the roflumilast group vs seven [9%] in the placebo group), cough (six [8%] vs four [5%]), diarrhoea (four [5%] vs three [4%]), and nasopharyngitis (three [4%] vs five [6%]). Severe adverse events included worsening of COPD, which occurred in four (5%) patients in the roflumilast group and two (3%) in the placebo group. No deaths occurred during the study. Serious adverse events occurred in eight (10%) patients in the roflumilast group and five (6%) in the placebo group. Interpretation 16 weeks of treatment with roflumilast did not affect the number of CD8 cells in bronchial submucosa compared with placebo. However, we noted significant reductions in eosinophil cell counts in bronchial biopsy samples and induced sputum, generating the hypothesis that the effect of roflumilast in COPD could be mediated by an effect on lung eosinophils. Funding Takeda and AstraZeneca.