A chemo-free TKI-plus-radiotherapy strategy in EGFR-mutant stage III lung cancer

Design

Phase 3 open-label randomised trial, stopped early, with real-world validation

Population

43 randomised patients with EGFR-mutant unresectable stage III NSCLC

Primary outcome

Progression-free survival

Effect

Median 34.0 vs 7.4 months; hazard ratio 0.15 (p<0.001)

Standard treatment for unresectable stage III NSCLC is chemoradiotherapy, but EGFR-mutant disease responds poorly to chemotherapy and well to targeted agents. The phase 3 ADVANCE trial tested a chemotherapy-free strategy of the EGFR inhibitor aumolertinib with radiotherapy.

The signal was dramatic but the trial was small. It was stopped early after only 43 of a planned 98 patients were randomised, because of a large observed difference and loss of equipoise in an open-label setting. Median progression-free survival was 34.0 months with aumolertinib plus radiotherapy versus 7.4 months with chemoradiotherapy (hazard ratio 0.15), and overall survival favoured the experimental arm without reaching significance. A prespecified real-world cohort of 125 patients supported the finding.

The practical read is that a chemo-free TKI-and-radiotherapy approach is highly promising for EGFR-mutant unresectable stage III NSCLC, but the randomised evidence rests on very few patients. Treat it as a strong signal, supported by real-world data, rather than settled practice.

• Phase 3 trial of aumolertinib plus radiotherapy versus chemoradiotherapy in EGFR-mutant stage III NSCLC.
• Stopped early after only 43 of 98 planned patients were randomised.
• Median progression-free survival was 34.0 versus 7.4 months (hazard ratio 0.15).
• Overall survival favoured the experimental arm but was not significant (p=0.09).
• A prespecified real-world cohort of 125 patients supported the result.

Why it matters

It offers EGFR-mutant patients, who respond poorly to chemotherapy, a targeted alternative to chemoradiotherapy.

Don't overread it

Only 43 patients were randomised before early termination in an open-label trial, which can inflate the apparent effect; the real-world cohort strengthens but does not replace a full trial.

The statistics, in plain English

A hazard ratio of 0.15 is extreme and, from 43 patients stopped early, almost certainly overstates the true benefit. The consistent direction in a 125-patient real-world cohort is what makes the signal credible.

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.