Development

GenSci registers Phase I trial of four-target retinal fusion protein across three indications

GenSci registers Phase I trial of four-target retinal fusion protein across three indications

Changchun GeneScience Pharmaceutical Co., Ltd. (GenSci) has registered a Phase I clinical trial of GenSci148, a multispecific fusion protein targeting VEGF-A, VEGF-C, VEGF-D, and angiopoietin-2 (Ang-2), in patients with neovascular age-related macular degeneration (nAMD), diabetic macular edema (DME), and retinal vein occlusion (RVO). The trial is listed as not yet recruiting, with an estimated start date of September 2026 at Shanghai General Hospital.

The open-label, dose-escalation study (NCT07817615) is structured in two parts and plans to enroll up to 60 participants. Part 1 evaluates single ascending intravitreal doses — 1 mg, 3 mg, 7 mg, and 10 mg per eye — in nAMD patients. Part 2 moves to multiple ascending doses across all three indications, with dose levels described as low, medium, and high. Participants meeting retreatment criteria during follow-up may receive additional injections. The primary completion date is listed as June 2028, with full study completion expected by October 2028.

GenSci148 is designed to simultaneously block three VEGF isoforms (VEGF-A, VEGF-C, VEGF-D) and Ang-2, a combination that, according to a GenSci press release, is intended to suppress pathological angiogenesis, reduce vascular permeability, and improve vascular stability. The molecule's broader target profile distinguishes it from faricimab (Vabysmo), Roche's approved dual VEGF-A/Ang-2 bispecific antibody, which does not address VEGF-C or VEGF-D. Whether the broader VEGF blockade translates into clinically meaningful differences in durability or anatomical outcomes remains to be established in human studies.

The molecule's development involved an AI-guided protein engineering collaboration with Shanghai Matwings Technology Co., Ltd. (Matwings). According to a Matwings press release, four iterative AI–wet lab cycles evaluated 222 protein variants, with reported preclinical improvements including up to approximately 10-fold improved VEGF-A binding affinity, up to approximately 3-fold improved VEGF-A/C/D functional blockade, and inhibitory activity maintained 84 days after dosing in an evaluated nonclinical retinal model. Under the specific experimental conditions tested, the optimized molecule also showed greater inhibition than aflibercept and faricimab. Matwings explicitly noted these findings are preclinical and do not establish comparative clinical efficacy or safety. GenSci said it received clinical trial clearance from China's National Medical Products Administration (NMPA) in August 2026.

The AllSci BriefFree, systematic R&D and deal news. Daily.

Faricimab, approved by the US FDA in January 2022 for nAMD and DME, demonstrated non-inferior visual acuity gains versus aflibercept and achieved up to 16-week dosing intervals in Phase III trials. China-based Kexing Biopharm's GB10, a bispecific antibody employing a structurally distinct VEGF-Trap fused to an anti-Ang-2 single-domain VHH antibody, dosed its first subject in a Phase I trial in China in mid-2026. GenSci148 is differentiated by its explicit simultaneous targeting of VEGF-A, VEGF-C, VEGF-D, and Ang-2, although whether broader VEGF blockade translates into greater durability or clinical efficacy remains unproven.

GenSci, a subsidiary of Changchun High-Tech Industries (Group) and headquartered in Changchun, China, has been expanding its ophthalmology pipeline alongside partnerships in other therapeutic areas, including a USD 1.365 billion global ex-China licensing deal for its anti-TSHR antibody GS-098 announced in December 2025. GenSci148 represents the company's ophthalmology lead program and, according to its pipeline page, is listed under the Evomab multispecific antibody platform.

Safety, tolerability, pharmacokinetics, and pharmacodynamics data from Part 1 are expected to inform dose selection for Part 2. The key question for progression will be whether the Phase I safety profile supports the dosing intervals GenSci has proposed, and whether preliminary efficacy signals in Part 2 — across nAMD, DME, and RVO — justify the broader development investment that would be required to compete against established agents in each indication.


Spot something wrong? Report an issue with this article