Living dossier · Competitive landscape

PD-1/VEGF bispecific in Non-small cell lung cancer

Primary identifiers: 8
Sponsors tracked: 7
Evidence types: 2 of 3
Last refreshed: 2026-07-28 · monthly

MightyRayn panel synthesis

OmniRayn’s reading of this landscape — our interpretation, not a registry fact

The first bispecific to beat a PD-1 antibody head to head, and the registry has already moved past asking whether it works. A Phase 3 reported longer median overall survival for the bispecific plus chemotherapy than for a PD-1 antibody plus the same chemotherapy. Read the trial before the enthusiasm: it ran entirely in China, the enrolled population was over 90 percent male, higher-grade treatment-related adverse events were more common in the bispecific arm, and grade 3 or higher haemorrhage ran higher — the VEGF-related risk this mechanism carries by construction. What the registry shows now is a combination land-grab: the same agent is being paired with a KRAS G12C inhibitor and with a TROP2 conjugate in this disease. Read against our KRAS G12C and TROP2 landscapes, the bispecific is positioning as the new combination backbone — the seat a PD-1 antibody used to occupy.

Programs and primary evidence

Sponsor	Asset	Phase	Mechanism	Primary source
Akeso	Ivonescimab (HARMONi-6)	3	PD-1/VEGF bispecific versus a PD-1 antibody, both with chemotherapy	NCT05840016 — Head-to-head Phase 3 against tislelizumab plus chemotherapy in first-line advanced squamous disease
Sichuan University	AK112 (malignant pleural effusion)	2	PD-1/VEGF bispecific with chemotherapy in an effusion population	NCT06769295 — Phase 2 in advanced non-squamous disease with malignant pleural effusion
Shanghai Junshi Bioscience	JS207	2	anti-PD-1/VEGF bispecific antibody with platinum doublet	NCT06944470 — Perioperative study in stage II-III disease, a second sponsor entering the same mechanism
Shanghai Chest Hospital	Fulzerasib + ivonescimab	2	KRAS G12C inhibitor combined with a PD-1/VEGF bispecific	NCT06936644 — First-line in KRAS G12C-mutant disease — the bispecific used as a combination partner rather than a monotherapy
Memorial Sloan Kettering Cancer Center	Ivonescimab + Dato-DXd or osimertinib	1/2	PD-1/VEGF bispecific paired with a TROP2 conjugate or an EGFR inhibitor	NCT07535437 — In EGFR-mutant disease that progressed on EGFR TKI therapy
Akeso	AK146D1 + AK112	2	PD-1/VEGF bispecific as a backbone for further combinations	NCT07669779 — Combination study with other anticancer therapies in advanced disease
Shanghai Jiao Tong University	HARMONi-6 interim overall-survival report	—	PD-1/VEGF bispecific versus PD-1 antibody, both with chemotherapy	10.1016/S0140-6736(26)00966-9 — Trial reported longer median overall survival for the bispecific arm; conducted entirely in China, the enrolled population was over 90 percent male, grade 3 or higher treatment-related adverse events were more frequent in the bispecific arm, and grade 3 or higher haemorrhage was more frequent
Sun Yat-sen University Cancer Centre	review	—	next-generation immunotherapy strategies against checkpoint-inhibitor resistance	10.1038/s41571-025-01035-9 — Identifies ivonescimab as a bispecific PD-1 x VEGF antibody approved in China in 2024, cited as proof-of-concept for overcoming checkpoint-inhibitor resistance

Method and limits

Programs are enumerated from primary registries (ClinicalTrials.gov, PubMed). A program appears here only if a primary record links this target to this indication as a therapeutic hypothesis — associations and risk markers are excluded by design.
Every claim above carries the identifier it rests on. We report what the record states; we do not assert efficacy, and computational findings are pending wet-lab confirmation.
Absence is not evidence of absence: unregistered, non-US and stealth-stage programs are not visible to these sources.
This page refreshes monthly (no net-new primary entity in 18 days). Mechanism classifications are our reading of the public record and may be revised.

Want the analysis behind this landscape?

The public page is the map. The full dossier adds differentiation analysis, kill-tests and the whitespace read for this target — the parts we do not publish.

Request the full dossier All landscapes