readthroughSign in
Sep 22, 2026Clinical readout

FDA tells Compass Therapeutics to run survival trial before BLA for tovecimig

Compass disclosed on September 22, 2026 that the FDA wants a trial showing survival benefit before it accepts a BLA filing for tovecimig in biliary tract cancer, but the company disputes that a new study is needed.

Compass Therapeutics said on September 22, 2026 that the FDA recommended it conduct a trial demonstrating a survival benefit before proceeding with a Biologics License Application submission for tovecimig in patients with previously treated, advanced biliary tract cancer.1

The company pushed back on that recommendation. Compass said it does not believe a new trial is warranted ahead of a BLA submission, pointing to the data from the Phase 2/3 COMPANION-002 study and what it called the urgent unmet medical need for BTC patients.1 Compass said it plans to keep engaging with the FDA while continuing to prepare its BLA submission.1

On the trial data, tovecimig plus paclitaxel showed a statistically significant improvement in the primary endpoint of objective response rate, 18.0% versus 5.3% for paclitaxel alone (p=0.0228), along with a highly significant progression-free survival improvement in previously treated BTC patients.1 Median PFS was 4.7 months versus 2.6 months, with a hazard ratio of 0.44 (p<0.0001).1 Overall survival results were confounded by high crossover and by notably prolonged survival among control-arm patients who crossed over to tovecimig, so OS did not reach statistical significance.1 Compass said the safety profile was generally consistent with earlier tovecimig studies.1

Tovecimig is described as an investigational DLL4 x VEGF-A bispecific antibody built to block two angiogenic pathways at once1, and it has received Fast Track and Orphan Drug designations from the FDA.1 The 8-K also noted Compass's intention to present the complete COMPANION-002 dataset at a medical conference during 2026, as stated in the filing's forward-looking statements regarding the expectation to present the complete dataset at a medical conference this year.1

Written by readthrough’s AI from the linked primary sources and fact-checked against them automatically before publishing. Not investment advice.