Multi-Criteria Decision Analysis of Biologics in Chronic Obstructive Pulmonary Disease

Chronic obstructive pulmonary disease (COPD) is a heterogeneous condition with limited response to standard anti-inflammatory therapies. Biologics targeting specific inflammatory pathways have emerged as potential treatments, but their efficacy remains variable across distinct COPD endotypes.

To systematically evaluate the efficacy and trial quality of biologics tested in COPD patients using a multicriteria decision analysis (MCDA) approach, with attention to type 2 (T2) and non-T2 inflammatory targets.

We assessed 20 trials encompassing 12 biologics and 9294 patients with COPD. Each trial was scored (0-3 per domain, total 12 points) across four domains: exacerbation reduction, lung function improvement, biomarker stratification, and trial design quality.

Dupilumab (anti-IL-4Rα) demonstrated the most robust efficacy in eosinophilic COPD, with consistent reductions in exacerbation rates and improvements in FEV₁, supported by high trial quality. Mepolizumab and benralizumab (anti-IL-5/IL-5R) showed moderate efficacy in biomarker-enriched populations. Anti-alarmins, specifically tozorakimab (anti-IL-33), itepekimab (anti-IL-33/IL-1RL1), astegolimab (anti-ST2), and tezepelumab (anti-TSLP), showed mixed results, with modest lung function gains but largely non-significant effects on exacerbation rates. Agents targeting non-T2 pathways, including infliximab (anti-TNF-α), canakinumab (anti-IL-1β), MEDI8968 (anti-IL-1R1), CNTO6785 (anti-IL-17A), and ABX-IL8 (anti-IL-8), consistently failed to demonstrate clinical efficacy, often due to small sample sizes, early-phase design, and lack of biomarker stratification.

Biologics targeting T2 inflammation offer therapeutic promise in eosinophilic COPD when guided by biomarkers. Conversely, current biologics directed at non-T2 and alarmin pathways yield limited or inconsistent benefits, emphasizing the need for improved phenotyping and targeted intervention strategies in non-eosinophilic COPD.