This report is based on public company disclosures, filings and announcements reviewed by GSN; figures are as stated by the company and have not been independently verified.

When a new cancer pill goes head to head with chemotherapy, the question patients and doctors care about is simple: did people live longer before their cancer started growing again? This week, for one kind of lung cancer, the honest answer was “it depends who you ask.” That split verdict now sits at the centre of the story of a clinical-stage drug developer.

The developer is ArriVent BioPharma, Inc. (Nasdaq: AVBP), a clinical-stage company based in Newtown Square, Pennsylvania, which describes its business as accelerating the global development of new medicines. Its drug firmonertinib is an oral pill that blocks a growth signal called EGFR and that, according to the company, crosses readily into the brain.

What happened

ArriVent said its Phase 3 FURVENT trial did not meet its primary endpoint, the single result a trial is designed and judged on. A Phase 3 trial is the large, final stage of testing that regulators usually expect before approving a drug. FURVENT enrolled 398 patients across the United States, Europe and parts of Asia, including Japan and China, all with previously untreated advanced non-small cell lung cancer carrying a particular genetic change known as an EGFR exon 20 insertion. Patients received firmonertinib at 240 mg or 160 mg a day, or the current standard first treatment, platinum-based chemotherapy with pemetrexed. The trial is being run jointly with partner Allist, the company said. Chief executive Bing Yao called the results “not what we hoped for” and said the company is evaluating the full dataset before choosing a development path.

Two sets of judges, two different scores

The main measure was progression-free survival, meaning how long patients went before their cancer grew or they died. Crucially, it was judged by blinded independent central review: outside reviewers reading the scans without knowing which treatment each patient received.

Think of a diving competition with two panels. One panel watches from behind a screen and cannot see the divers’ names. The other sits poolside and knows exactly who is who. The trial’s rulebook said the screened panel’s score is the one that counts. On that panel’s card, patients on the higher dose went a median 11.0 months before progression against 9.5 months on chemotherapy, a gap of 1.5 months. The statistical summary, called a hazard ratio, came in at 0.75, which on its face suggests a lower risk of progression, but the plausible range around it stretched as high as 1.02, just past the point that means no difference at all. The result did not clear the bar for statistical significance.

The poolside panel, the trial’s own investigators, scored it differently: 11.1 months on the higher dose versus 7.1 months on chemotherapy. Notice what moved. The drug’s number barely changed between the two panels; chemotherapy’s number did. The independent reviewers simply judged the chemotherapy patients to be doing better than the investigators did, and that is what closed the gap.

There were other encouraging signs, according to the company. Under independent review, 60% of patients on the higher dose saw their tumours shrink meaningfully, against 33% on chemotherapy. Serious side effects judged to be caused by treatment (grade 3 or worse) affected 26% of patients on the higher dose compared with 40% on chemotherapy. ArriVent also reported an early trend toward longer overall survival, which it cautioned is not yet mature. The lower 160 mg dose, by contrast, looked much like chemotherapy on the independent read.

Where this fits in ArriVent’s story

According to the release, firmonertinib is already approved in China for several groups of lung cancer patients, including exon 20 patients whose disease has progressed after platinum chemotherapy. In the United States, it holds Breakthrough Therapy Designation for exactly the population FURVENT tested, a status the FDA uses to speed up development of drugs that show early promise. FURVENT was meant to turn that promise into first-line evidence. In other words, this was the trial that was supposed to move firmonertinib from a later option to the first thing a patient with this mutation receives, and on the score it chose in advance, it did not get there.

What this could become

One possibility is that the higher dose carries on in some form. If the full dataset and maturing survival figures hold up, ArriVent might argue that a pill with fewer serious treatment-related side effects and more tumour shrinkage still has a place, perhaps in a narrower group of patients.

Another is that overall survival becomes the deciding number. Survival is the hardest measure to argue with; if the early trend firmed up, it could change how the whole trial is read.

A third is that attention shifts elsewhere. The company describes firmonertinib as active against other EGFR mutations, including one group it calls PACC, so its development path might lean toward those uses instead.

The signpost is the one the company has named itself: what it says after it finishes reviewing the full FURVENT data, and what the survival figures show once they mature.

The desk’s view

To our eye, ArriVent deserves credit for the way it delivered bad news. It published both panels’ scores side by side, including the one that makes the miss plain, and it did not bury the lower dose’s weak showing. That is more candour than a reader can always expect.

Our doubt is about leaning on the secondary results. Independent blinded review is usually made the primary measure precisely because it is the harder test to sway, and here it is the investigators’ reading, not the drug’s, that differs. This desk’s reading is that the gap between the two chemotherapy numbers is the question worth asking about: why did the independent reviewers and the investigators disagree so much about patients on the standard treatment?

What to watch

The documents give no dates. The named next steps are ArriVent’s evaluation of the full dataset, its decision on firmonertinib’s development path, and the maturing overall survival data.

Plainly put: FURVENT missed the goal it was designed to hit. What happens to firmonertinib in first-line exon 20 lung cancer has not been decided, and the company says it is still working that out.

Sources