This individual participant data meta-analysis evaluated whether ACQ-5 symptom burden predicts future severe asthma attacks and response to anti-inflammatory therapy, using data from 22 RCTs (ORACLE2, n=6,513 control-group participants) plus four additional translational/observational datasets (total 7,161 distinct participants).
Each 0.5-point rise in baseline ACQ-5 added only a modest increase in attack risk (aRR 1.09, 95% CI 1.06–1.12; ΔR²=0.02). ACQ-5 did not predict benefit from IV mepolizumab, whereas patients with high blood eosinophils AND high FeNO had the greatest absolute risk reduction (2.81 vs 1.17 attacks; aRR 0.38, 95% CI 0.25–0.57). ACQ-5 was also unassociated with post-corticosteroid lung function change.
- Control-group data from RCTs may not fully represent real-world asthma populations. - Additional translational datasets were small (n=53–74) and selected based on data availability, limiting generalisability. - ACQ-5 is a self-reported measure subject to recall and response bias, and unmeasured confounders (e.g., adherence, inhaler technique) were not accounted for.
Symptom scores alone are poor guides for escalating anti-inflammatory therapy in asthma — type 2 biomarkers (blood eosinophils and FeNO) more reliably identify who is at high risk and who will benefit from biologic treatment. Clinicians should interpret ACQ-5 results in the context of objective biomarkers rather than using symptoms as the primary trigger for anti-inflammatory escalation.
