Less than a quarter of patients with severe to very severe COPD are likely to respond to inhaled corticosteroid therapy, a post hoc analysis of the WISDOM trial of ICS withdrawal has concluded.

Presented as a poster at ATS, and also published in Lancet Respiratory Medicine, the analysis found that withdrawal of ICS did increase the risk of subsequent moderate or severe exacerbations in patients with elevated blood eosinophil counts.

The rate was increased by 22% when the count was ≥ 2% of total white blood cells, by 63% when the count was ≥ 4% of total white blood cells, and by 82% when count was ≥ 5% of total white blood cells. Although the increase in the exacerbation rate became more pronounced as the eosinophil cut-off level rose, the interaction with treatment was significant only when the eosinophils were at least 4% or 5% of the total count.

“Similar results were seen with eosinophil cut-offs of 300 cells/μL and 400 cells/μL,” the researchers said.

However, withdrawal of ICS did not increase exacerbations in the majority of patients – about 75% – who had blood eosinophil levels <4% or <300 cells/µL.

Related story: Are blood eosinophils a biomarker of response to ICS?

Professor Peter Calverley, a lead investigator in the WISDOM trial and Professor of Pulmonary and Rehabilitation Medicine at the University of Liverpool, UK, in a previous comment on the study had said that long-acting bronchodilators were a mainstay of COPD management.

“However in clinical practice, ICS are widely used across all COPD stages,” he said. “It has been difficult to determine the subset of patients who respond to ICS. These findings will help physicians more confidently identify which patients may benefit from ICS therapy, helping minimise exposure to the risk of long-term side effects of ICS use.”

The 12-month WISDOM study enrolled 2,485 patients who had at least one exacerbation in the previous year. They were treated with 18 µg tiotropium, 100 µg salmeterol, and 1000 µg fluticasone propionate triple therapy ( total daily dose) during a 6-week run-in period, followed by ICS continuation or stepwise ICS dose reduction every 6 weeks.

The primary result of the study, published in 2014, was that the overall risk of exacerbations was similar among those who discontinued ICS and in those who continued treatment, but there was a greater decrease in lung function during the final step of ICS withdrawal.

This latest analysis found no consistent differences in exacerbation rates between the treatment groups when analysed by low versus high serum IgE, positive versus negative RAST tests, and positive versus negative skin-prick test subgroups.