Tezepelumab worked in the patients severe asthma trials usually leave out

Design

Phase 4, multicentre, single-arm, open-label, 12-month study (PASSAGE, NCT05329194)

Population

286 US patients aged 12 and over with severe uncontrolled asthma, enrolled across phenotypes and underrepresented groups including smokers and those with coexisting COPD

Primary outcome

Annualised asthma exacerbation rate in the 12 months before versus after starting tezepelumab

Effect

2.88 to 0.87, a 70% reduction (95% CI 63-75); 54-77% across subgroups; FEV1 +0.122 L overall, +0.212 L if below 80% predicted

The standing objection to severe asthma biologic data is that the trials enrol a narrow population. PASSAGE was built to answer it: a phase 4, single-arm, open-label, 12-month US study that deliberately recruited across phenotypes and underrepresented groups - blood eosinophils above and below 300 cells per microlitre, with and without allergy, Black patients, adolescents, people with 10 or more pack-years of smoking, and people whose severe asthma coexists with mild-to-moderate COPD.

Across 286 participants the annualised exacerbation rate fell 70% (95% CI 63-75), from 2.88 in the year before tezepelumab to 0.87 in the year after, and the reduction held between 54% and 77% in every phenotype and every underrepresented group examined. Pre-bronchodilator FEV1 at week 52 rose by a least-squares mean of 0.122 L (0.07-0.17), and by 0.212 L (0.15-0.28) in those who started below 80% predicted. Between half and nine in ten participants recorded meaningful improvements on asthma control and quality-of-life questionnaires. No new safety signals.

The design is the limitation and it is a serious one. A single-arm before-and-after comparison in patients enrolled because their asthma was uncontrolled will capture regression to the mean and any improvement in adherence that follows joining a study, and neither of those is the drug. The consistency across subgroups is the more persuasive finding here than the size of the drop.

Where this lands is in the referral decision. A patient with severe uncontrolled asthma, a low eosinophil count and a smoking history is often not offered a biologic at all; tezepelumab acts upstream of the type 2 pathway, and this is the most directly relevant evidence that such patients benefit. In Indian practice the constraint is cost rather than eligibility, but the eligibility argument should stop being used as the reason not to discuss it.

• Do not exclude a patient from biologic discussion because the eosinophil count is under 300
• A significant smoking history did not abolish the benefit here - review those patients again
• Confirm the diagnosis and inhaler technique before attributing failure to disease severity
• Record a baseline exacerbation count and FEV1 so response can actually be judged at 12 months
• Treat the effect size cautiously: single-arm designs flatter any treatment given to uncontrolled patients

The statistics, in plain English

A 70% reduction here compares each patient with their own previous year, not with a control group - the single most important thing to know about this number. Patients enter such a study when their asthma is at its worst, and severity tends to drift back towards average on its own, so part of the improvement belongs to that rather than to the drug. What a randomised trial provides and this design cannot is the size of that part. The consistency of the effect across very different subgroups is the finding that survives the objection.

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.