Use eosinophils and FeNO, not symptom scores alone, to decide asthma escalation

Design

Individual participant data meta-analysis of trials and prospective cohorts

Population

7161 people with asthma of varying severity, including ORACLE2 (n=6513)

Primary outcome

Association of ACQ-5 with severe attacks and treatment response

Effect

Attack risk RR 1.09 per 0.5 ACQ-5 points (95% CI 1.06–1.12); high eosinophils and FeNO with mepolizumab RR 0.38 (0.25–0.57)

This individual patient data meta-analysis used control-group data from 22 randomised trials (ORACLE2, 6513 people) plus four further datasets, 7161 participants in all. It asked whether ACQ-5 symptom scores reflect inflammation, predict severe attacks or predict response to anti-inflammatory treatment. It was published in August 2026.

Symptom scores bore little consistent relation to lung function or inflammatory markers. Each 0.5-point rise in ACQ-5 raised attack risk only slightly, and adding it to a prediction model barely improved accuracy. Symptom score did not predict response to mepolizumab. Patients with both raised blood eosinophils and raised FeNO had the most attacks prevented, with a rate ratio of 0.38.

Much asthma escalation is driven by symptoms: breathlessness leads to more inhaler, more steroid, or a biologic. Breathlessness can come from obesity, deconditioning, dysfunctional breathing or anxiety, none of which respond to anti-inflammatory therapy. Measure eosinophils, and FeNO where available, before stepping up, and treat non-inflammatory causes of symptoms directly.

• Check blood eosinophil count before escalating anti-inflammatory treatment
• Measure FeNO where available
• Look for obesity, dysfunctional breathing, reflux and anxiety when symptoms are high but biomarkers are low
• Use attack history and biomarkers together to judge future risk
• Consider biologic referral for patients with high eosinophils and FeNO

Why it matters

It challenges the assumption that worse symptoms mean more airway inflammation needing more anti-inflammatory treatment.

Don't overread it

The analysis is secondary use of trial and cohort data; it does not test a biomarker-guided strategy against symptom-guided care directly.

The statistics, in plain English

A rate ratio of 1.09 (95% CI 1.06–1.12) per 0.5 ACQ-5 points is a small increase in risk, and ΔR² of 0.02 means the symptom score explained almost nothing extra beyond other predictors. The rate ratio of 0.38 (0.25–0.57) in patients with high eosinophils and FeNO is a large treatment effect in a clearly defined group.

Read the rest in the app

You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.

Get Daily Doctor — free

Scan to keep reading on your phone. No account needed to start.