BioNTech and its new partner Bristol Myers Squibb have trotted out the first global data for their PD-L1xVEGF bispecific in small cell lung cancer, results BioNTech’s chief medical officer, Özlem Türeci, M.D., describes as “very consistent” with the drug’s previous encouraging findings from China.
In a global phase 2 trial, the two companies linked pumitamig (BNT327), used in combination with chemotherapy, to a confirmed overall response rate (ORR) of 76.3% among 38 evaluable patients with treatment-naïve extensive-stage small cell lung cancer (ES-SCLC), according to data presented at the 2025 World Conference on Lung Cancer.
The median duration of response (DoR) was five months, Türeci told Fierce Biotech in an interview. All patients were considered to have their disease under control.
The result was collected at a Aug. 7 data cutoff across patients who received pumitamig at two dose levels, 20 mg/kg and 30 mg/kg. The ORR was 85% for the lower dose and 66.7% for the higher dose, although Türeci cautioned against reading too much into the seemingly inversed dose-ORR relationship given the small number of patients in each subgroup.
The median progression-free survival (PFS) reached 6.8 months, with this measure landing at 6.3 months for the low dose and seven months for the high dose. Median overall survival (OS) was not mature at the time of the analysis.
No new safety signals were observed beyond those known for chemo, checkpoint inhibitors and anti-VEGF drugs. Out of 43 patients evaluated for safety, grade 3 or above pumitamig-related adverse events happened in one patient (grade 3 hemoptysis) at the lower dose and five patients at the higher dose.
No treatment-related deaths occurred, Türeci confirmed.
The results are “exciting and compelling” as they compare favorably to the historical performance of the existing immunotherapy-based standard-of-care first-line treatment in ES-SCLC, Türeci noted, acknowledging that the follow-up time is still short.
In 2019, the FDA approved Roche’s Tecentriq in combination with chemo for first-line ES-SCLC based on results from the phase 3 IMpower133 trial. The study linked the regimen to a 60% ORR, 4.2 months of median DoR, and 5.2 months of median PFS.
The ultimate goal for BioNTech and BMS is essentially to replace Tecentriq and other PD-(L)1 inhibitors with pumitamig as a next-gen immuno-oncology treatment backbone.
In March, BioNTech unveiled early phase 2 data from a separate China-only study, which showed an 85% confirmed ORR, a median DoR of 5.5 months, 6.9 months of median PFS and 16.8 months of median OS for pumitamig at 30 mg/kg plus chemo in first-line ES-SCLC.
BioNTech obtained full control over pumitamig through a $800 million-plus acquisition of its Chinese partner Biotheus in 2024 and attracted BMS as a codevelopment and co-commercialization partner in June 2025 under a deal worth up to $11.1 billion.
Despite the global numbers coming slightly below the Chinese results, Bryan Campbell, BMS’ head of program leadership of hematology, oncology and cell therapy, agreed that the latest data set is consistent with the Chinese readout.
“I think it supports our approach to move into phase 3, which is ongoing,” Campbell said. “And I also think it supports the potential of this drug to set a new standard of care in small cell extensive-stage small cell lung cancer.”
Based on the results, BioNTech and BMS have decided that the ongoing Rosetta Lung-01 phase 3 trial in first-line ES-SCLC will continue with pumitamig at a dose of 1500 mg given every three weeks, or 1200 mg in patients with a body weight below 50 kg, as the equivalent to the 20-mg/kg dose tested in the phase 2 trial, a BioNTech spokesperson told Fierce Biotech. The companies will talk to the FDA about their choice. The trial is pitting pumitamig-chemo against Tecentriq-chemo.
Previously, BioNTech opened two arms with two different pumitamig dose levels for the phase 3 trial while running dose optimization in the phase 2 in parallel to determine which dose to take to the end.
As to why the partners are switching from a weight-based approach to a fixed-dose regimen, Campbell explained that the latter option is more convenient.
Campbell declined to lay out a timeline for a potential readout from the phase 3 trial, noting that it’s event-driven. The study currently bears an estimated primary completion date in April 2028, according to ClinicalTrials.gov.
In SCLC, pumitamig could eventually compete with DLL3-targeted agents, a group led by Amgen’s T-cell engager Imdelltra. A global phase 3 trial recently showed that Imdelltra reduced the risk of death by 40% compared with standard-of-care chemo in SCLC patients who have tried one line of platinum-based chemo. in first-line ES-SCLC, Amgen is running two other phase 3 trials, DeLLphi-305 and DeLLphi-312.
“The more compounds with potential are developed and approved, the better for the patients,” Türeci said. “At the end of the day, it will be about combinations, about combining these compounds in order to get even better outcomes for patients. And given that we see our compound as a next-generation I-O backbone—also across lines, across indications—that means that we actually even are more open to any new compound.”
Combinations with ADCs
While BioNTech and BMS work to make pumitamig a new immunotherapy backbone, the pair has simultaneously started looking for ways to replace the chemo component of the regimen with antibody-drug conjugates.
BioNTech recently launched a phase 1/2 study pairing pumitamig with its DualityBio-licensed B7-H3 ADC candidate BNT324 in lung cancer, including SCLC and non-small cell lung cancer.
The German biotech has also started teaming pumitamig up with DualityBio-partnered TROP2-targeting ADC BNT325 in another phase 1/2 trial across multiple solid tumors, including NSCLC, for which the phase 2/3 Rosetta Lung-02 trial is evaluating BNT325 plus chemo in the first-line setting. The company released some early clinical data for the TROP2 combo from the global phase 1/2 trial conducted by DualityBio in April.
To round out the initial wave of potential indications for pumitamig, the phase 3 Rosetta Breast-01 study in triple-negative breast cancer is expected to start this year.
The DualityBio collaboration also includes the HER2 ADC BNT323, which recently entered a phase 1/2 trial in combination with pumitamig in breast cancer. And a phase 1/2 to evaluate the PD-L1xVEGF bispecific with MediLink Therapeutics-partnered HER3 ADC BNT326 is being planned for 2025.
“We think that it is an obvious step that, at some point, chemotherapies in some indications will be replaced by ADCs, and we want to be, with our partner BMS, early movers in this space,” Türeci said, adding that the pair has not determined what specific ADC combinations to move into late-stage testing.
“One of the real beauties of this partnership is both companies have broad pipelines,” Campbell said. “In terms of the next wave of combinations, we have potential to combine with a number of different assets, including ADCs, which we’re excited about.”
Elsewhere, Pfizer is looking to pursue ADC combinations for its own PD-1xVEGF bispecific licensed from 3SBio.
Part of Pfizer’s clinical development plan for its agent will include potential internal combination opportunities, “particularly with some of our ADCs, where we’ve been generating evidence that combinations with anti-PD-1 agents can deliver potentially practice-changing results,” Pfizer’s chief oncology officer, Jeff Legos, Ph.D., said during a webinar in July.
There has been some ongoing debate about which component—PD-L1 or PD-1—is better for a VEGF bispecific molecule. As monoclonal antibodies, PD-1 inhibitors have had better outcomes than their anti-PD-L1 peers in most tumor types. But for a bispecific construct, Türeci argued that the PD-L1 portion does not only have the purpose to restore T-cell immunity against tumors but also help anchor the entire molecule to the tumor microenvironment. And, because PD-L1 is expressed on tumors—versus PD-1, which is expressed on activated T cells—targeting it would theoretically lead to better anchoring.
However, PD-1xVEGF developers such as Akeso have argued that because T cells are only activated and express PD-1 in the tumor microenvironment, their anchoring ability is just as potent, and the more therapeutic success with PD-1 monoclonal antibodies makes PD-1xVEGF the superior construct versus PD-L1xVEGF.