Abstract

Introduction Inhaled corticosteroids (ICS) can improve clinical outcomes in patients with chronic obstructive pulmonary disease (COPD) and eosinophilic airway inflammation, but they also increase the risk of side effects like pneumonia. Blood eosinophils guide ICS use, though evidence is limited. The predictors of treatment REsponse to ICS in COPD: a randomised controlled trials (RCTs) individual participant Data re-Evaluation (ICS-RECODE) research programme will leverage data from large RCTs to identify patients who benefit most from ICS with minimal risk. This protocol details an individual participant data (IPD) meta-analysis, assessing ICS safety, efficacy and treatment×covariate interactions to identify predictors of treatment response. Methods and analysis This meta-analysis will adhere to Cochrane, IPD handbook and Grading of Recommendations Assessment, Development and Evaluation (GRADE) guidance. We will conduct a two-stage IPD meta-analysis of RCTs evaluating the addition of ICS to maintenance COPD treatments. Only RCTs with at least 500 participants across all eligible arms will be included, to allow for treatment×covariate interaction evaluation. Primary outcomes are severe and moderate or severe exacerbation rates; secondary outcomes assess both safety and efficacy. Data from each RCT will be reanalysed using rigorous, consistent statistical methods. Treatment×covariate interactions will be assessed at the RCT level. Trial treatment effects and the coefficients of treatment×covariate interaction analyses will be pooled using random effects model meta-analysis. Risk of bias will be appraised using RoB-2 informed by IPD, and certainty of evidence will be assessed with GRADE and the Instrument to assess the Credibility of Effect Modification Analyses. The ICS-RECODE IPD meta-analysis will make use of the best available data to define evidence-based, precision medicine approaches for ICS use in COPD. Ethics and dissemination The Health Research Authority approved the ICS-RECODE study, exempting it from ethics review (HRA UK, Reference: 24/HRA/0460). Our findings will be published in peer-reviewed journals and shared with the scientific and broader stakeholder communities. PROSPERO registration number CRD42024508286.