Dropping chemotherapy in EGFR-mutant stage III lung cancer: a large signal from a tiny trial

Design

Open-label phase 3 randomised trial, stopped early at prespecified interim analysis, with a protocol-prespecified real-world validation cohort

Population

43 randomised patients (24 experimental, 19 control) with unresectable stage III EGFR-mutant NSCLC; real-world cohort n = 125

Primary outcome

Progression-free survival

Effect

Median PFS 34.0 vs 7.4 months (HR 0.15, P < 0.001); median OS not reached vs 30.5 months (HR 0.32, P = 0.09)

ADVANCE randomised patients with unresectable stage III EGFR-mutant non-small-cell lung cancer to aumolertinib 110 mg daily with definitive radiotherapy, or cisplatin-pemetrexed chemoradiotherapy. The plan was 98 patients. It stopped at 43 — 24 experimental, 19 control — after a prespecified interim analysis, for loss of equipoise in an open-label trial, slow accrual, and the size of the difference already visible.

At a median 25.5 months, median progression-free survival was 34.0 months against 7.4 (hazard ratio 0.15, P < 0.001). Median overall survival was not reached against 30.5 months (HR 0.32, P = 0.09). Neutropenia and nausea were more common in the chemotherapy arm and quality of life favoured the experimental arm. A protocol-prespecified real-world cohort of 125 patients supported the finding: TKI with radiotherapy and TKI with concurrent chemoradiotherapy performed similarly, and both beat chemoradiotherapy alone.

The difficulty is that trials stopped early for benefit systematically overestimate it, and 43 patients is very few. But the control arm's median progression-free survival of 7.4 months is itself striking — chemoradiotherapy performs poorly in EGFR-mutant disease, which is the underlying reason this question was asked. Combined with the real-world cohort and with what is already known about EGFR-mutant biology, this is enough to put a chemotherapy-free option on the table for discussion, while acknowledging the effect size will probably shrink.

• Raise a TKI-plus-radiotherapy strategy at the multidisciplinary meeting for unresectable stage III EGFR-mutant disease — it is now a defensible discussion, not a fringe one.
• Quote the hazard ratio with its caveat: a trial stopped early at 43 patients overstates the effect it stopped for.
• Note that overall survival did not reach significance (HR 0.32, P = 0.09) — progression-free survival is what this trial showed.
• Weigh the toxicity difference honestly; less neutropenia and nausea is a real patient-facing gain.
• Aumolertinib is a third-generation EGFR TKI approved in China; availability and price elsewhere, including India, differ, and osimertinib is the agent most Indian units would be discussing.

Why it matters

Chemoradiotherapy in EGFR-mutant stage III disease has been the standard while performing poorly in exactly this group.

Don't overread it

A trial stopped early at 43 of a planned 98 patients overestimates the benefit it was stopped for, and overall survival was not significant.

The statistics, in plain English

A hazard ratio of 0.15 from 43 patients is the kind of number that almost always moves towards 1 as a trial matures — early stopping for benefit selects for the moment the random variation happens to favour the treatment. The overall survival result illustrates this: HR 0.32 sounds large, but P = 0.09 means the data are still compatible with no survival difference. The real-world cohort of 125 patients is the more reassuring element, precisely because it was not stopped at a favourable moment.

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.