Bispecific receptor decoys: A new player to a crowded field

Main text

Abnormal vessel growth in retinopathy of prematurity (ROP), diabetic retinopathy, and age-related macular degeneration (AMD) are the leading causes of blindness across a person’s lifespan, yet modern treatments have only begun to emerge in the last two decades. The rise of anti-vascular endothelial growth factor (VEGF) therapies in the early 2000s transformed the landscape of neovascular eye disease management nearly overnight, replacing destructive or invasive therapies like thermal laser photocoagulation or macular translocation surgery with simple, in-office intraocular injections that today exceed $4 billion in Medicare spending.¹ These treatments included full-length monoclonal antibodies (bevacizumab), antibody fragments (ranibizumab), and soluble decoy receptors (aflibercept) that prevent VEGF from activating its native receptors and promoting aberrant vessel growth. Their efficacy in visual restoration was so strong that subsequent drugs have failed to surpass their performance in clinical trials, focusing instead on extending durability through non-inferiority study designs. These newer agents, which include higher-dose formulations (8 vs. 2 mg aflibercept), bispecific antibodies against VEGF and angiopoietin-2 (Ang-2; faricimab), and surgically implanted platforms (ranibizumab port-delivery system), promised to alleviate the burden of frequent eye injections by extending treatment intervals from every 4–8 weeks to every 4–6 months. However, the real-world effectiveness of these newer agents has not matched clinical trial outcomes, a discrepancy attributed in part to treatment resistance caused by failure to address the compensatory upregulation of other pro-angiogenic cytokines. For example, most current anti-VEGF therapies only target VEGF-A, VEGF-B, and/or placental growth factor (PlGF) but do not block other VEGF family members, such as VEGF-C or -D, which may increase following bevacizumab administration and continue to drive pathologic vessel growth.²

In this issue of Molecular Therapy, Sargunas and colleagues describe the development of novel bispecific VEGF receptor decoys that are engineered to block all VEGF members and sequester two ligands simultaneously, with the goal of overcoming anti-VEGF treatment resistance.³ Although bispecific antibodies have been previously developed to target multiple angiogenic factors, bispecific decoy receptors offer certain advantages. First, the binding affinity of a ligand for its receptor is typically stronger than with an antibody. For example, surface plasmon resonance showed that the binding affinity of aflibercept (0.49 pM) for VEGF is higher than that of ranibizumab (46 pM).⁴ Second, homo- or heterodimerization of receptor tyrosine kinases like VEGF receptors not only activates downstream signals but may also enhance ligand binding. By fusing the ligand-binding domains of VEGFR-1 and VEGFR-2 to the Fc domain of a human immunoglobulin (Ig)G antibody, the investigators found that the enforced dimerization of the receptor fragments enhanced binding site cooperativity, allowing the fusion protein to bind multiple VEGF family members with higher affinity. Additionally, they elegantly compared different topologies and found that binding domains positioned further from the Fc domain had higher binding affinities and that the initial occupation of one binding domain made the other site more accessible to ligands.

Despite the novelty of bispecific decoy proteins, critical questions remain: how do they compare with existing therapies in clinical use, and can they overcome treatment resistance? In their study, the authors nicely demonstrated the superior performance of their agent in mouse models of ischemic retinopathy and choroidal neovascularization (CNV), which are well-established models that reflect the pathophysiology of ROP and neovascular AMD. However, the effects were only compared with aflibercept, a monospecific decoy receptor, rather than bispecific antibodies, such as faricimab, which are also used in clinical practice. Additionally, a recent phase 3 clinical trial evaluating a combination therapy targeting both VEGF-A/B/PlGF and VEGF-C/D failed to meet primary endpoints for neovascular AMD,⁵ challenging the premise that inhibiting more VEGF members could provide better performance than VEGF-A/B blockade alone. Indeed, pathogenic angiogenesis has been largely attributed to VEGF-A isoforms such as VEGF-A₁₆₅, while the relative contribution of VEGF-C/D to disease pathophysiology remains unclear. Besides VEGF-C/D, the upregulation of other pro-angiogenic factors may also contribute to treatment resistance, including fibroblast-derived growth factor (FGF), platelet-derived growth factor (PDGF), and Ang-2. Interestingly, while combining anti-VEGF with anti-PDGF injections failed to show additional benefits in human trials,⁶ bispecific anti-VEGF/Ang-2 antibodies have demonstrated clinical efficacy across multiple neovascular eye conditions.⁷ This difference is likely attributed, however, to a careful clinical trial design that emphasized improved durability rather than visual acuity benefits.

Most clinical trials do not address treatment resistance because patients with fibrosis or chronic disease are often excluded from enrollment as study subjects. Chronic hypoxic or oxidative damage to vascular endothelium and retinal pigment epithelial cells can trigger endothelial- or epithelial-to-mesenchymal transition (EndMT or EMT), resulting in preretinal or subretinal fibrosis which can cause scarring and limit the effectiveness of anti-angiogenesis therapies. In addition to the mouse models used in their study, testing in newer disease models, such as two-stage laser CNV induction⁸ and laser-induced CNV in aged mice,⁹ could better predict their likelihood of overcoming treatment-resistant neovascularization. Measuring decreases in fibrosis markers such as collagen-1, fibronectin, or α-smooth muscle actin or the expression of pro-fibrotic or pro-inflammatory genes could also add more compelling support for the clinical translatability of these engineered proteins.

The emergence of bispecific receptor decoys introduces new contenders to an already crowded field of neovascular eye disease therapies. Small-molecule tyrosine kinase inhibitors administered via bioresorbable sustained-release polymers¹⁰ or through novel suprachoroidal intraocular injections¹¹ demonstrate better durability than conventional injections of biologic molecules in early studies. Retinal gene therapies using viral vectors to create a biofactory for continuous anti-VEGF production may offer long-term or even permanent disease suppression that could disrupt the current paradigm of repeated eye injections. Early phase trials of a dual-transgene intravitreal gene therapy encoding both aflibercept and microRNA (miRNA) sequences targeting VEGF-C have shown promise so far, providing support for the promise of pan-VEGF suppression. Amid these many emerging therapies, identifying the appropriate target patient population with unmet needs not readily addressed by other strategies may pose a challenge. However, successfully overcoming treatment resistance will help advance the fight against neovascular eye diseases and vision loss.

Declaration of interests

G.Y. received research support or personal fees from 4DMT, Abbvie, Adverum, Alcon, Bausch & Lomb, Boehringer Ingelheim, Clearside Biomedical, Endogena Therapeutics, Genentech, Iridex, Janssen, jCyte, Myrobalan Therapeutics, Nanoscope, NGM Biopharmaceutical, Novartis, Ocuphire, Opthea, Ray BioTherapeutics, RegenXBio, Samsung Bioepis, Stealth Therapeutics, Supra Medical, Surrozen, and West Pharmaceutical.