Taiho and Cullinan’s Lung Cancer Drug Aces a Phase 3 Trial, Heating Up a Crowded Race

An experimental drug from Taiho Oncology and Cullinan Therapeutics worked well enough in a hard-to-treat lung cancer that its late-stage trial was stopped early.

Zipalertinib targets non-small cell lung cancer driven by EGFR exon 20 insertion mutations. In a Phase 3 trial of zipalertinib plus chemotherapy versus chemotherapy alone as first-line treatment, an interim analysis found a “statistically significant and clinically meaningful improvement” in progression-free survival — enough to halt the study.

Why exon 20 insertions are different

EGFR-mutant lung cancer is one of oncology’s success stories, and exon 20 insertions were long the exception that did not benefit.

Common EGFR mutations — deletions in exon 19 and a point mutation in exon 21 — alter the receptor in ways that make it both more active and more susceptible to inhibitor drugs. Patients with those mutations respond dramatically to EGFR inhibitors.

Exon 20 insertions activate the receptor without creating that susceptibility. The insertion changes the shape of the drug-binding region such that standard EGFR inhibitors bind poorly — so patients had an EGFR-driven cancer that EGFR drugs could not treat, and received chemotherapy instead.

The subtype is uncommon, accounting for a small percentage of EGFR-mutant cases, which is part of why it was neglected for years.

What stopping early means

Trials are halted at interim analysis for efficacy when an independent monitoring committee judges the benefit sufficiently clear that continuing to randomise patients to the control arm is no longer justified.

That is a meaningful signal, and it comes with a statistical caveat worth noting. Effect sizes from trials stopped early for benefit tend to be overestimates, because the decision to stop is triggered by a favourable interim result — and interim results fluctuate. A trial run to completion frequently shows a smaller effect than the one that prompted the halt.

The direction is real; the magnitude should be treated as provisional.

The first-line move

Testing in the first-line setting rather than after chemotherapy failure is the strategically important element.

Targeted drugs in lung cancer generally enter after chemotherapy, then move earlier as evidence accumulates. Establishing first-line use means patients receive the targeted agent before the disease has evolved under selective pressure, and avoids exposing them to chemotherapy first.

Earlier Phase 2 data had shown tumour shrinkage in more than a third of patients on the drug alone, which supported moving forward.

The competitive picture

Zipalertinib enters a busy market. Johnson & Johnson’s Rybrevant — over $700 million in 2025 sales — and AstraZeneca and Dizal’s Zegfrovy are already approved for this mutation, and ArriVent’s firmonertinib has Phase 3 data expected later in 2026.

That is a striking transformation. A subtype with essentially no targeted option a few years ago now has multiple approved drugs and more coming — and the drivers are the same as elsewhere in precision oncology: molecular testing became routine, so patients could be identified; regulatory pathways accommodated small biomarker-defined populations; and demonstrated commercial viability attracted competitors.

How the competitors differ

The approved options are not equivalent, which leaves room for differentiation.

One is a bispecific antibody given by infusion, targeting EGFR and MET simultaneously, with a side effect profile including infusion reactions and skin toxicity. The others are oral small molecules with different tolerability.

For a chronic therapy in a population that may take it for a year or more, tolerability and route can determine which drug is used as much as efficacy does — and full zipalertinib data will need to be read on those dimensions rather than response rate alone.

What happens next

An FDA decision on zipalertinib is expected by February 27 for a second-line use, with the new first-line data strengthening the case for a broader role.

The sequencing is conventional — approval in later-line use first, then expansion — and the first-line result arriving before that decision puts the company in a strong position to file promptly for the broader label.

Why testing determines who benefits

A drug for a rare mutation only reaches patients who are tested for it, and the testing picture in lung cancer is better than it was and still incomplete.

Comprehensive molecular profiling at diagnosis is now guideline-recommended in advanced non-small cell lung cancer, precisely because so many actionable alterations exist. Adherence in practice varies considerably, with community settings testing less consistently than academic centres and some patients starting treatment before results return.

Exon 20 insertions carry an additional complication: they are structurally diverse, comprising many distinct insertions rather than one recurring change. Older testing methods designed to detect specific known mutations can miss them entirely, and detection generally requires sequencing approaches that read the whole region.

So a patient can be tested, reported as EGFR-negative, and have an exon 20 insertion that the assay was not designed to find. Expanding treatment options for this subtype therefore depends as much on which test is used as on whether testing happens at all.

For patients with EGFR exon 20 lung cancer, who long had few good options, more effective competing therapies is straightforwardly welcome news. Clinical and business news, not medical advice.