Design: Randomised non-inferiority trial
Population: 106 patients with grade 1–2 immune checkpoint inhibitor pneumonitis
Primary outcome: Treatment success at 8 weeks
Effect: 66.7% (3 weeks) vs 85.2% (6 weeks); difference −18.5 points

This randomised trial, reported in the American Journal of Respiratory and Critical Care Medicine, was the first to test corticosteroid duration for immune checkpoint inhibitor pneumonitis. It enrolled 106 patients with grade 1–2 pneumonitis (median age 72; 73% grade 2), assigned to a 3-week or the guideline-recommended 6-week taper.

Treatment success at 8 weeks — room-air SpO2 of 90% or more without steroid escalation or prolongation — was 66.7% with 3 weeks and 85.2% with 6 weeks (difference −18.5 points). Non-inferiority of the shorter course was not shown, and a predefined exploratory analysis favoured the 6-week taper. Grade 3 or worse adverse events were commoner with 6 weeks (24% vs 12%) but manageable. Quality of life and overall survival did not differ.

With checkpoint inhibitors widely used, pneumonitis is increasingly managed by pulmonologists and oncologists together, and shortening steroids is tempting in older patients. This trial says not to: the shorter taper failed in about one in three. The trial was small, so it supports rather than overturns current guidance.

For grade 1–2 immune-related pneumonitis treated with steroids, use a taper of about six weeks.
Do not shorten the taper to three weeks; treatment failure was substantially more common.
Watch for steroid adverse effects — hyperglycaemia, infection, myopathy — over the longer course.
Check oxygen saturation at rest and on exertion during the taper.

Why it matters

It replaces guideline opinion with randomised evidence on a decision made every week in oncology clinics.

The statistics, in plain English

The trial used an 80% confidence interval for its non-inferiority test; the interval (−29.0 to −7.9 points) lay wholly below zero, meaning the short course was worse rather than merely not proven equal.