Compass Therapeutics Touts “Stunning” PFS Win in COMPANION-002 as Crossover Muddy OS Data

Key Points

• COMPANION‑002 met its key secondary endpoint of BICR‑assessed progression‑free survival, with median PFS 4.7 vs 2.6 months (HR 0.44, p<0.0001) and an improved ORR of 17.1% vs 5.3%, results CEO Thomas Schuetz described as “absolutely stunning.”
• Widespread patient crossover (54% of control patients; 85% of study participants received tovecimig) confounded the intent‑to‑treat overall survival analysis
• Safety was consistent with prior studies (hypertension the main differentiator; 2.8% pulmonary hypertension), and Compass plans biomarker analyses, to complete remaining endpoints in ~1 month, request an FDA meeting mid‑summer, pursue a BLA later in the year, and target potential U.S. approval in 2027.

Compass Therapeutics NASDAQ: CMPX executives highlighted what CEO and Vice Chair Dr. Thomas Schuetz called “absolutely stunning” progression-free survival results from the company’s phase II/III COMPANION-002 trial in second-line advanced biliary tract cancer, while also detailing how patient crossover complicated the study’s overall survival analysis.

Secondary endpoint: progression-free survival

Speaking on the company’s “Topline Secondary Endpoints” call, Schuetz said COMPANION-002 met its key secondary endpoint of progression-free survival (PFS) as assessed by blinded independent central review (BICR). The trial compared tovecimig plus paclitaxel versus paclitaxel alone in patients who had received one prior line of therapy.

According to Schuetz, median BICR-assessed PFS was 4.7 months for the tovecimig-plus-paclitaxel arm versus 2.6 months in the paclitaxel control arm, with a hazard ratio of 0.44 and a p-value less than 0.0001. He noted the study was powered for a hazard ratio of about 0.6, and said the observed result represented a 56% reduction in the risk of progression.

Schuetz also reiterated previously reported primary endpoint data: BICR-assessed overall response rate (ORR) was 17.1% in the combination arm versus 5.3% in the control arm, including a centrally adjudicated complete response in the combination group.

Trial design and baseline characteristics

COMPANION-002 enrolled second-line advanced biliary tract cancer patients and randomized participants 2:1 to tovecimig plus paclitaxel or paclitaxel alone. The trial allowed patients in the control arm to cross over after centrally confirmed progression to receive tovecimig plus paclitaxel.

Schuetz said 111 patients were randomized to the combination arm and 31 patients crossed over from paclitaxel alone, meaning 142 patients received tovecimig at some point during the study. He added that baseline stratification factors—including primary tumor location (intrahepatic cholangiocarcinoma versus other), ECOG performance status (0 versus 1), and extent of disease (outside the liver versus locally advanced)—were “well-balanced” between treatment arms.

Overall survival: crossover complicates intent-to-treat analysis

Schuetz emphasized that overall survival (OS) in the intent-to-treat (ITT) analysis was “confounded by crossover.” He said 54% of patients in the paclitaxel arm crossed over, and “85% of patients in the study ultimately received tovecimig in combination with paclitaxel.”

During Q&A, Schuetz acknowledged investor focus on the ITT OS hazard ratio, arguing it should not be interpreted as evidence of harm. “The reason that the hazard ratio is 1.05 is because the control arm patients who crossed over did so amazingly well,” he said, adding that crossover patients “drove the hazard ratio” in the ITT analysis.

To explore crossover effects, Schuetz discussed a pre-specified endpoint the company called PFS2, which assessed progression-free survival in crossover patients before and after crossover. He reported that crossover patients had a median PFS of 1.9 months prior to crossover on paclitaxel, improving to 3.5 months after crossover to tovecimig plus paclitaxel, with a hazard ratio of 0.36 and p=0.0016 (with time zero resetting at crossover for the post-crossover curve).

Schuetz also described an analysis splitting the control arm into crossover and non-crossover subsets. He said crossover patients (n=31) had a median OS of 12.8 months versus 6.1 months for patients who did not cross over and received paclitaxel alone. He acknowledged the potential for selection bias in such comparisons, but pointed to PFS on paclitaxel as evidence the crossover group was not inherently better prognosis, saying the crossover patients progressed significantly faster on paclitaxel than non-crossover patients.

Jefferies’ Amin asked about comparing “pure” treated populations, and Schuetz confirmed that median OS was 8.9 months in the randomized combination arm versus 6.1 months in patients who received paclitaxel without crossover.

Safety and next steps toward FDA discussions

Schuetz said top-line safety was consistent with prior studies and highlighted hypertension as the main difference between treatment arms, calling it “on mechanism for a VEGF blockade.” In response to a Piper question on discontinuations, he said he did not yet have the full dose modification dataset but noted that 4 patients (3.7%) discontinued due to hypertension.

On pulmonary hypertension, he said it was not listed among the most prominent adverse events in the slide deck, but reported three adverse events of pulmonary hypertension in the study (2.8%). He added the company monitored for it with serial BNP measurements and echocardiograms and said the observed incidence was lower than in prior phase I and phase II studies conducted in Korea.

Compass Therapeutics plans to complete remaining endpoint analyses in roughly a month and then request an FDA meeting, which Schuetz said could take place “approximately the middle of the summer.” He said the company intends to seek full approval via a planned Biologics License Application (BLA) submission, while also arguing the ORR and PFS results could support accelerated approval. “The hazard ratio here of 0.44 is without any question an approvable number,” he told H.C. Wainwright.

Schuetz said the company collected archival biopsy specimens and plans a “comprehensive biomarker analysis,” including blood samples, to better understand why crossover patients appeared to do particularly well. He also said Compass is exploring additional statistical methods to adjust OS for crossover, since the pre-specified rank preserving structural failure time (RPSFT) approach was not interpretable. “The statistical assumptions for the RPSFT analysis were not met,” he said, adding that the company was providing those figures “solely for full disclosure.”

Looking further ahead, Schuetz said Compass is already studying tovecimig added to a frontline regimen of gemcitabine, cisplatin, and durvalumab at MD Anderson, and also discussed the possibility of future studies such as tovecimig plus paclitaxel versus FOLFOX in the second-line setting.

In closing remarks, Schuetz said that if FDA discussions go well, Compass expects to be in position to submit the BLA toward the end of the year and expressed a long-term goal of a potential U.S. approval in 2027.

About Compass Therapeutics NASDAQ: CMPX

Compass Therapeutics, Inc is a clinical‐stage biotechnology company dedicated to the discovery and development of novel immuno‐oncology therapies. Headquartered in Cambridge, Massachusetts, the company focuses on engineering monoclonal antibody candidates designed to enhance T cell–mediated anti‐tumor responses. Compass leverages proprietary antibody platforms to identify and optimize biologics that modulate immune checkpoint pathways and the tumor microenvironment.

The company's lead programs include CTX-471, a bispecific antibody targeting both PD-1 and PD-L1 checkpoints, and DSP107, a CD47‐SIRPα pathway modulator aimed at disrupting “don't eat me” signals on cancer cells.

This instant news alert was generated by narrative science technology and financial data from MarketBeat in order to provide readers with the fastest reporting and unbiased coverage. Please send any questions or comments about this story to contact@marketbeat.com.