PD-1 x VEGF bispecific antibodies: the competitive landscape.

A PD-1 x VEGF bispecific antibody puts checkpoint blockade and anti-angiogenesis into one molecule. The class went from a single Chinese asset to the most contested area in solid-tumour immuno-oncology in under three years, and every large pharmaceutical company that did not originate one has now licensed one.

This page is a permanent reference, not a news item. It lists the programmes, the trials that will decide them and the regulatory actions already taken, each linked to its primary record. It is maintained as those records change.

The programmes, at a glance

Pumitamig is included because it competes for the same clinical position, but note the target difference: it binds PD-L1, not PD-1.

Ivonescimab — the one with a regulatory date

Ivonescimab is a tetravalent bispecific engineered by Akeso, licensed to Summit Therapeutics for North America, South America, Europe, the Middle East, Africa and Japan. Summit's filed 8-K exhibit describes cooperative binding, with higher affinity to PD-1 in the presence of VEGF, and a half-life of six to seven days after the first dose rising to roughly ten days at steady state.

The FDA accepted the Biologics License Application and set a PDUFA goal action date of 14 November 2026. The application covers ivonescimab plus platinum-doublet chemotherapy in EGFR-mutated, locally advanced or metastatic non-squamous NSCLC after progression on a third-generation EGFR TKI, and rests on the global Phase 3 HARMONi study. The same exhibit states that fourteen Phase 3 studies of ivonescimab are ongoing or complete, ten run by Akeso in China and four sponsored by Summit globally.

In the separate Chinese Phase 3 HARMONi-6, Akeso reported that ivonescimab plus chemotherapy improved overall survival against tislelizumab plus chemotherapy in first-line squamous NSCLC: median OS 27.9 months versus 23.7, hazard ratio 0.66 (95% CI 0.50–0.87), p = 0.0017, in 532 patients at a 27 February 2026 data cutoff, presented in an ASCO 2026 Plenary Session and published in The Lancet. The 24-month OS rate was reported as 64.7% versus 48.6%.

PF-08634404 (SSGJ-707) — the broadest Phase 3 programme

Pfizer completed an ex-China licence from 3SBio in July 2025 for SSGJ-707, a PD-1 x VEGF bispecific built on 3SBio's CLF2 (common light chain, linear-Fabs-IgG) platform. It is now the widest Phase 3 footprint in the class: lung, colorectal, small cell and endometrial cancer, all registered inside twelve months.

The pivotal lung study is NCT07222566 (Symbiotic-Lung-01): Phase 3, double-blind, 1,410 patients, PF-08634404 plus chemotherapy against pembrolizumab plus chemotherapy in first-line NSCLC, split into a squamous Part 1 and a non-squamous Part 2. In China, 3SBio's affiliate runs the parallel NCT06980272 against pembrolizumab in PD-L1-positive first-line NSCLC, 420 patients.

The Phase 2 monotherapy data that justified this expansion come from NCT06361927, a first-line PD-L1-positive NSCLC study at 10 mg/kg every three weeks, reported at ASCO with a confirmed objective response rate of 67.6% and median progression-free survival of 12.4 months. Two caveats belong with that number and are stated rather than smoothed: it is single-arm Phase 2, and the registry record for that study carried a status of "unknown" at the review date, meaning the sponsor has not updated it within the registry's expected window.

Pumitamig (BNT327 / BMS-986545) — the PD-L1 variant

BioNTech's pumitamig, partnered with Bristol Myers Squibb, is the structural outlier: it targets PD-L1 x VEGF-A rather than PD-1 x VEGF. The Phase 3 programme spans three tumour types with different comparators — NCT07361510 (ROSETTA Lung-202, monotherapy against pembrolizumab in PD-L1 ≥ 50% first-line NSCLC, 750 patients), NCT06712355 in first-line small cell lung cancer against atezolizumab plus chemotherapy, 621 patients, and NCT07173751 in PD-L1-negative metastatic triple-negative breast cancer, 558 patients — alongside the NCT06712316 Phase 2/3 first-line NSCLC master protocol at 1,580 patients.

MK-2010 (LM-299) — the different geometry

Merck licensed LM-299 from LaNova Medicines for $588 million upfront, closing in December 2024, with up to $2.7 billion in milestones. Merck's own description of the construct is unusual for the class: an anti-VEGF antibody linked to two C-terminal single-domain anti-PD-1 antibodies, rather than a symmetric IgG-scFv fusion.

The asset is still early. The first-in-human study is NCT06650566, sponsored by LaNova Medicines, Phase 1/2, 108 patients planned, recruiting. First clinical data appeared at AACR 2026 as abstract CT057, reporting unconfirmed objective response rates of 55% at 20 mg/kg and 44% at 30 mg/kg in treatment-naive PD-L1-positive patients, with VEGF class adverse events in 51% and 49% of patients at those doses respectively.

CR-001 — the deliberate replica

Crescent Biopharma's CR-001 is a tetravalent construct — an anti-VEGF IgG carrying stabilised anti-PD-1 scFvs — designed explicitly to reproduce ivonescimab's geometry. The NCT07335497 ASCEND study is Phase 1/2, 290 patients planned, recruiting, with the first patient dosed in February 2026 and proof-of-concept data guided by the company to the first quarter of 2027.

Every registered late-stage trial in the class

What actually decides this class

Landscape last reviewed 1 September 2026. This page is a permanent reference on PD-1 x VEGF and PD-L1 x VEGF bispecific antibodies, maintained continuously as registries, regulatory actions and congress disclosures change; it is not a news briefing and does not claim to be exhaustive. Every fact links to the primary record behind it, and trial figures are read from the registry itself on the review date rather than from a release describing it. Enrolment figures are planned targets unless the record states otherwise. No claim of superiority or inferiority is made about any programme: the differences described are structural. Prognyx has no affiliation with any company named here and holds no position in their securities. Names and trademarks belong to their owners. Nothing here is medical or investment advice. Found an error? Write to hello@prognyx.com and it will be corrected.