The entire Compass story now runs through a single meeting with the FDA. In April, the company disclosed the survival readout from its randomized phase 2/3 trial of tovecimig, a bispecific antibody that blocks both DLL4 and VEGF-A, combined with paclitaxel in second-line biliary tract cancer. The study met its primary endpoint and then its headline secondary endpoint: progression-free survival (time until the tumor grows or the patient dies, the trial's main durability measure) stretched to 4.7 months, versus 2.6 months with chemotherapy alone. That is a 56% reduction in the risk of progression. Overall survival, the more contested endpoint, did not separate statistically because more than half of the control arm crossed over to the combination after their disease progressed, which muddied the randomized comparison. The company is holding a pre-submission feedback meeting with the agency this quarter before filing a biologics license application later in 2026.
That setup makes this one of the cleanest binary positions in oncology. The stock closed at 2.38 on September 4. The share count is roughly 180 million, and the implied market value sits near 429 million. That puts a single-indication tovecimig franchise against a cash balance of about 180 million and a burn of about 44 million per half year. The price embeds skepticism: the share is down more than half from the 6.88 high set in early February, before the survival data arrived, yet the April print cleared both the primary and the PFS bar and the full dataset earned an oral presentation at ESMO in October. The interpretation is that the discount is the market's verdict on the OS miss, and the central debate is whether the crossover-distorted survival data is a flaw in the trial design or a flaw in the drug.
Three variables decide the stock over the next four quarters: the tone and scope of the FDA feedback arriving this quarter, the pre-commercialization spending rate, and the readouts from CTX-8371, the PD-1 by PD-L1 bispecific in triple-negative breast cancer, Hodgkin lymphoma, and non-small cell lung cancer, whose first expansion data are due in the fourth quarter. The falsifiable clock is short. A request for a confirmatory trial or a clinical hold ends the current case, and the next data point that tests it is the agency's feedback letter.