Tovecimig (CTX-009)

Tovecimig is an investigational bispecific antibody that binds delta-like ligand 4 and vascular endothelial growth factor A, given intravenously at 10 mg/kg on days 1 and 15 of each 28-day cycle with paclitaxel 80 mg/m2 on days 1, 8, and 15. A population pharmacokinetic model describes a two-compartment disposition with parallel linear and Michaelis-Menten elimination and a terminal half-life of approximately 5.7 days. The product is not approved in any jurisdiction. The FDA granted Fast Track designation in April 2024 for previously treated metastatic or locally advanced biliary tract cancer and orphan drug designation for biliary tract cancer in April 2026.

The registrational evidence is COMPANION-002, an open-label randomized phase 2/3 trial that enrolled 168 participants at 34 United States sites and allocated them 2:1 to tovecimig plus paclitaxel (111 participants) or paclitaxel alone (57 participants), with crossover permitted after centrally confirmed progression. Eligibility required radiologically documented progression after one prior gemcitabine and platinum containing regimen, an Eastern Cooperative Oncology Group performance status of 0 or 1, and ineligibility for an approved molecularly targeted therapy. The trial met its primary endpoint of objective response rate by blinded independent central review: 17.1% (19 of 111) including one complete response versus 5.3% (3 of 57), p=0.031, revised in the final analysis to 18.0% (20 of 111), p=0.0228. Median progression-free survival was 4.7 months versus 2.6 months (hazard ratio 0.44, p<0.0001).

The overall survival endpoint was not met. Median overall survival was 8.9 months with tovecimig plus paclitaxel versus 9.4 months in the control arm (hazard ratio 1.05, p=0.78), and a prespecified rank-preserving structural failure time adjustment gave 8.9 versus 9.4 months (hazard ratio 1.13, p=0.65). The sponsor attributes this to crossover: 31 of 57 control participants (54%) crossed over, so 142 of 168 participants (85%) received tovecimig and the control arm is not a tovecimig-naive comparator for survival. Three analyses favour tovecimig but are not randomized comparisons. In the 31 crossover participants, median progression-free survival was 3.5 months after crossover versus 1.9 months before it (hazard ratio 0.36, p=0.0016). Within the control arm, median overall survival was 12.8 months in crossover participants versus 6.1 months in non-crossover participants (hazard ratio 0.54, p=0.04), but the same participants had progressed faster on paclitaxel alone, with median progression-free survival of 1.9 versus 3.6 months (hazard ratio 2.31, p=0.007), indicating the two subgroups differed. Pooled median overall survival across all 142 participants who received tovecimig was 9.8 months.

Safety was reported for 108 and 53 participants. The most frequent treatment-emergent adverse events with tovecimig plus paclitaxel were hypertension (69%) and fatigue (67%), and the most common related grade 3 or higher events were hypertension (44%) and neutropenia (36%). Grade 3 or higher hypertension occurred in 52% of the tovecimig arm against 6% of the control arm. The sponsor reported no new safety signals.

Supporting evidence is early phase and smaller. In a Korean phase 1b/2 study, the objective response rate with tovecimig plus paclitaxel was 37.5% (9 of 24 participants, 95% CI 18.8 to 59.4), reaching 63.6% in the second-line subgroup and 15.4% in the third-line subgroup, with median duration of response and median progression-free survival both 9.4 months and 12-month overall survival of 53.0%; grade 3 or higher treatment-related adverse events occurred in 75% and neutropenia in 50.0%. The first-in-human phase 1 study dosed 45 participants with no dose-limiting toxicities, established recommended phase 2 doses of 10 and 12.5 mg/kg, and reported an objective response rate of 18.8% (3 of 16) at the recommended dose and hypertension in 37.8%. In COMPANION-003, a study of tovecimig monotherapy in third- and fourth-line metastatic colorectal cancer, the posted registry results record 2 responses among 49 participants, disease control in 28 of 49, median progression-free survival of 3.9 months, and median overall survival of 10.2 months.

Several limitations bear on interpretation. As of September 21, 2026 no COMPANION-002 publication or congress presentation existed, so all pivotal results come from sponsor documents filed with the U.S. Securities and Exchange Commission and are not peer reviewed; the complete dataset was selected for oral presentation at the October 2026 European Society for Medical Oncology Congress. The trial was powered to detect a 23% absolute difference in response rate with 90% power and a two-sided alpha of 0.05, so the observed difference of about 12 percentage points is smaller than the effect the sample size was planned around. Duration of response, participant disposition, deaths, serious adverse events, discontinuations due to adverse events, disease control rate, and the EORTC QLQ-C30 quality-of-life endpoint specified in the design publication have not been reported. All sites were in the United States, and eligibility excluded participants who could receive an approved molecularly targeted therapy, so the results apply to the population without an actionable alteration.