ArriVent's Firmonertinib Misses Phase 3 FURVENT Endpoint, Stock Drops 47%

ArriVent BioPharma (AVBP) said its lead drug firmonertinib missed the primary endpoint of the Phase 3 FURVENT trial in first-line EGFR exon 20 insertion non-small cell lung cancer (NSCLC). Shares closed down 47% to $15.09 on Monday, after opening down roughly 57% and touching an intraday low near $11.36, erasing about $13 from a $28.46 prior close.

The trial randomized 398 patients with previously untreated, non-squamous locally advanced or metastatic NSCLC to one of two firmonertinib doses or platinum-based chemotherapy. The primary endpoint, progression-free survival (PFS) by blinded independent central review (BICR), missed statistical significance.

The 240 mg dose posted a median PFS of 11.0 months versus 9.5 months for chemo, a hazard ratio of 0.75 (95% CI, 0.55 to 1.02) and p=0.0654, just above the significance threshold. The 160 mg dose was numerically worse than chemo at 8.4 months. The confirmed response rate favored the higher dose at 60% versus 33% for chemo, but the PFS miss is what regulators weigh most.

Investigator assessment, a secondary measure, looked stronger at 11.1 months versus 7.1 months. Overall survival is immature, showing only “a trend toward improvement.” Safety was consistent with earlier firmonertinib trials, with grade 3 or higher treatment-related adverse events in 26% of the 240 mg arm versus 40% on chemo.

Mechanism and Why Exon 20 Is Hard

Exon 20 insertions resist first- and second-generation EGFR inhibitors, which is why chemo plus Rybrevant remains the default for these patients. The insertion loop distorts the EGFR binding pocket in a way that keeps classical TKIs like osimertinib from docking — the same receptor, a differently shaped door. Firmonertinib is a brain-penetrant inhibitor of both classical and uncommon EGFR mutations, designed to fit that distorted pocket, and it was ArriVent’s bid for a chemotherapy-free option in the roughly 9% of EGFR-mutant NSCLC driven by exon 20 insertions. Brain penetration matters for this population specifically, because exon 20 patients have high rates of CNS metastases — but a tolerable, brain-active drug still has to beat chemo on PFS to matter, and FURVENT says it did not, at least by BICR at the 240 mg dose.

The approved standard, Johnson & Johnson’s Rybrevant (amivantamab) plus chemo, reported a median PFS of 11.4 months versus 6.7 months in its own Phase 3. A same-class rival, Cullinan’s zipalertinib, already has its NDA accepted.

What It Means for the Trade

Guggenheim Securities had flagged a potential $40-per-share upside if FURVENT data at least matched the Rybrevant bar; instead the stock gave back about $13. The arithmetic of the miss is brutal: an 11.0-month median PFS versus chemo’s 9.5 is an 11.0-versus-6.7 comparator away from competitive, and the 160 mg arm was numerically worse than chemo outright. A monotherapy PFS that merely ties a chemo-combination standard gives an FDA reviewer little reason to prefer it, and it gives a payer little reason to pay for it.

The BICR miss removes ArriVent’s near-term regulatory catalyst and clouds the chemo-free thesis. Management is “evaluating the most appropriate development path” for firmonertinib, which held FDA Breakthrough Therapy Designation and was being studied jointly with partner Allist. That language typically means months of silence, not a quick pivot announcement — expect the next news to be either a rescued filing, a combination strategy, or an attempt to salvage the ex-U.S. story through Allist. A second program, ALPACCA in PACC-mutant NSCLC, is still running and is now the company’s main pipeline value.

What to watch next: whether ArriVent pivots firmonertinib to a chemo-combination setting, and whether the maturing overall survival data, which can override a PFS miss, tips in the drug’s favor.

Risks and What Could Go Wrong

The investigator-assessed PFS advantage (11.1 vs 7.1 months) and the strong response rate (60% vs 33%) are the bull case for salvage — but regulators weigh BICR precisely because local assessment is more optimistic, and a p-value of 0.0654 on the primary does not become approvable with spin. The confidence interval crossing 1.0 means the true effect could be nothing.

The competitive window is closing. Zipalertinib’s NDA is already accepted, Rybrevant owns the front line, and J&J has every incentive to consolidate the class with combination regimens. Even a successful OS salvage takes years. And with a $15 stock and no approved product, ArriVent’s financing position becomes part of the story — a pipeline company whose lead catalyst just failed will raise money into weakness if the pivot plan costs cash. A full recovery requires the OS data to be genuinely strong, and that readout is not scheduled on anyone’s near-term calendar.

Source: ArriVent press release and Fierce Biotech