Computational design of Vascular Endothelial Growth Factor Receptor 2 (VEGFR2)-Targeted Oncolytic Virus Therapy to In-hibit Angiogenesis in Non-Small Cell Lung Cancer

ABSTRACT

Non-Small Cell Lung Cancer (NSCLC) accounts for approximately 85% of all lung cancer cases and is strongly associated with dysregulated angiogenesis mediated by Vascular Endothelial Growth Factor Receptor 2 (VEGFR2). Activation of VEGFR2 by its ligand VEGF promotes endothelial cell proliferation, tumor vascularization, and cancer progression. Targeting VEGFR2 therefore represents a promising strategy to inhibit tumor growth. In this study, we computationally designed a VEGFR2-targeted oncolytic virus therapy aimed at suppressing angiogenesis in NSCLC. The amino acid sequence of VEGFR2 was obtained from UniProt, and its three-dimensional structure was modeled using AlphaFold 3. Potential ligand-binding regions were predicted using ScanNet and validated against known crystal structures. Multiple antibody Fab fragments were screened through protein–protein docking using HDOCK to evaluate their binding affinity toward the VEGFR2 extracellular domain. Binding interactions were analyzed using PLIP, RING, and PRODIGY to assess interaction networks and calculate binding energies. Among the candidates tested, Fab 8ZCA demonstrated the strongest predicted binding affinity, indicating its potential as a competitive inhibitor of VEGF binding. Additionally, NanoHub-based simulations were performed to model oncolytic viral spread within the tumor microenvironment, demonstrating effective viral replication and tumor cell lysis over time. Collectively, these findings suggest that integrating a high-affinity VEGFR2-binding Fab into an engineered oncolytic virus may enhance tumor selectivity, inhibit angiogenesis, and promote targeted NSCLC therapy. Future work will focus on experimental validation and viral genome engineering to translate these computational findings into therapeutic applications.

INTRODUCTION.

In the U.S. each year, about 226,650–229,410 new cases of lung cancer are diagnosed (lung and bronchus combined)..(1) Lung cancer results from the malignant transformation of lung epithelial cells due to accumulated genetic and epigenetic alterations, leading to uncontrolled proliferation and impaired apoptosis. Non–small cell lung cancer (NSCLC) is a heterogeneous group of tumors associated with risk factors such as tobacco exposure, environmental carcinogens, and oncogenic mutations (e.g., EGFR, KRAS, ALK, TP53) that disrupt normal cellular signaling and growth regulation.Symptoms are often nonspecific and may include coughing up blood, rust-colored sputum, unexplained weight loss, shortness of breath, and a prior history of lung cancer. Current treatments for NSCLC include surgery, radiation therapy, chemotherapy, targeted therapy, and immunotherapy. However, these treatments can cause side effects such as pain, nausea or vomiting, constipation or diarrhea, anemia, and, in some cases, collapsed lungs.(2) NSCLC is classified into three main subtypes: adenocarcinoma, squamous cell carcinoma, and large cell carcinoma. Adenocarcinoma is the most common subtype and typically develops in the outer regions of the lungs; it is also frequently observed in non-smokers.(1, 3) Squamous cell carcinoma generally arises in the central airways and is strongly linked to smoking. Large cell carcinoma can occur in any part of the lung and tends to behave more aggressively. Treatment decisions depend on the stage of the cancer stages and may involve a combination of surgery, radiation, chemotherapy, targeted therapy, or immunotherapy.(1)

Since NSCLC tumors depend on new blood vessel formation for survival and growth, angiogenesis plays a critical role in their progression. Angiogenesis is the formation of new blood vessels that supply tumors, allowing them to grow larger and become more difficult to remove (Figure 1).(4) It promotes cancer progression by providing essential nutrients and supporting tumor expansion. Angiogenesis is regulated by vascular endothelial growth factor (VEGF), which coordinates the development of new blood vessels. Many cancer therapies aim to block VEGF or its receptors to inhibit tumor growth.(1) Under normal conditions, angiogenesis plays a crucial role in wound healing, growth, and development. However, in diseases such as cancer, angiogenesis becomes abnormal. Tumors release signaling molecules—primarily VEGF—that stimulate nearby blood vessels to grow toward them. These new vessels deliver oxygen and nutrients to the tumor, enabling it to grow and potentially spread (metastasize) to other parts of the body. As a result, anti-angiogenic therapies have been developed to block new blood vessel formation, effectively “starving” the tumor and slowing disease progression.

Given that angiogenesis is essential for sustaining NSCLC tumor growth, targeting this process with novel therapeutic strategies—such as oncolytic virus therapy—offers a promising approach for disrupting tumor progression. Oncolytic virus therapy is an advanced cancer treatment in which genetically engineered or naturally occurring viruses selectively infect and destroy cancer cells. These viruses are modified to recognize and bind to tumor cells while sparing healthy tissue. Once inside the cancer cell, the virus replicates, eventually causing the cell to burst (lysis) and releasing new viral particles that spread to nearby cancer cells. This process also exposes tumor antigens, activating immune cells such as cytotoxic T lymphocytes to further attack and eliminate cancer cells. The first FDA-approved oncolytic virus therapy is talimogene laherparepvec (T-VEC), a genetically modified herpes simplex virus type 1 used to treat advanced melanoma.(5) T-VEC is engineered to express the GM-CSF gene, which enhances immune activation and improves anti-tumor responses. Another example is Oncorine (H101), an E1B-deleted adenovirus approved in China for head and neck cancers, illustrating how targeted gene deletions can make viruses selectively replicate in tumor cells.(6) More recently, teserpaturev (G47Δ), another modified herpes virus, was approved in Japan for malignant glioma, demonstrating the growing potential of oncolytic virotherapy for brain cancers.(7) Overall, oncolytic viruses not only destroy cancer cells directly but also transform tumors into in-situ vaccines, stimulating powerful systemic anti-tumor immunity, Figure 2.

An oncolytic virus is a specialized virus designed to selectively infect and kill cancer cells while leaving normal cells unharmed.(8) It also stimulates the immune system to recognize and attack remaining tumor cells, enhancing overall anti-cancer activity. The herpes simplex virus–based therapy T-VEC demonstrated durable responses in advanced melanoma, establishing the dual oncolytic and immunostimulatory roles of viral therapy.(9) The p53-selective adenovirus Onyx-015 provided early clinical proof that genetically engineered viruses can selectively replicate in tumors and induce measurable tumor regression in head and neck cancer.(10) Reovirus-based therapy demonstrated selective activity in Ras-activated tumors, with clinical trials reporting meaningful biological responses in metastatic colorectal cancer.(11) In hepatocellular carcinoma, the vaccinia virus vector JX-594 produced dose-dependent tumor necrosis and survival benefits, illustrating the therapeutic value of highly immunogenic viral platforms.(12) Collectively, these studies show that oncolytic viruses exert antitumor effects through both direct tumor lysis and by reprogramming the tumor microenvironment toward a stronger immune response, supporting their continued development in precision oncology. The proposed mechanism of VEGFR-mediated activation of angiogenesis and cancer cell formation is shown in Figure 3.  VEGFR2 binding to VEGF activates the receptor and promotes cancer cell proliferation. This increased proliferation leads to hypoxic, low-oxygen conditions that trigger angiogenesis, and the resulting angiogenesis further accelerates cancer cell growth. VEGFR2-specific oncolytic viruses can directly infect and lyse NSCLC cells that overexpress VEGFR2, minimizing off-target toxicity. By binding and blocking VEGFR2, this therapy could prevent tumor angiogenesis, effectively starving cancer cells of the nutrients and oxygen they need to survive. In addition, oncolytic viruses hold strong potential for combination therapy, as they can be used synergistically with radiotherapy, chemotherapy, or immune checkpoint inhibitors to enhance overall treatment response.(8) Furthermore, the viral lysis of tumor cells releases tumor antigens, which trigger cytotoxic T lymphocytes and boost long-term anti-tumor immunity. This approach currently has several limitations, as it is based only on computational predictions with no laboratory testing yet, and viral engineering still faces challenges such as stability and potential off-target effects. Additionally, tumor biology is more complex than targeting VEGFR2 alone.

Molecular docking is a computational method that predicts how two molecules (such as proteins and antibodies) will bind.(13–15) To achieve this, molecular docking tools are applied to computationally evaluate and compare binding interactions between potential therapeutic molecules and their targets. In the current research, we have identified an antibody FAB region that can be genetically introduced into an oncolytic virus, along with its corresponding receptor, enabling the formation of a receptor on the viral surface.

METHODS

To model and analyze the VEGFR2 protein, a series of computational tools and web servers was utilized. The amino acid sequence of VEGFR2 was first obtained from the UniProt database, a comprehensive resource for protein sequences and functional annotations.(16) The 3D structure of VEGFR2 was predicted using AlphaFold 3, a state-of-the-art deep learning model for protein structure prediction.(17) ScanNet was used to identify potential VEGFR2 binding sites.(18) The binding site prediction enabled visualization and mapping of the most probable binding pockets within the VEGFR2 structure. Protein–protein docking simulations were then performed using HDOCK, an integrated web server that supports the docking of proteins, peptides, and nucleic acids.(14) HDOCK was used to evaluate the binding interactions between VEGFR2 and antibodies to target NSCLC. Post-docking interaction analysis was conducted using PLIP (Protein–Ligand Interaction Profiler), which identifies and categorizes all non-covalent interactions, including hydrogen bonds, salt bridges, hydrophobic contacts, and π–π stacking, between VEGFR2 and the FAB complexes.(19) Using molecular docking, we repurposed existing antibody Fab fragments to assess their potential binding affinity and inhibitory interactions with VEGFR2. Subsequently, PRODIGY was used to calculate binding affinity and estimate the strength of interaction between VEGFR2 and the docked antibodies, thereby determining the FAB with the highest binding potential.(20) Finally, RING (Residue Interaction Network Generator) was applied to visualize the complete network of non-covalent interactions within the VEGFR2–FAB complex, providing a detailed interaction map essential for structural and functional interpretation.

RESULTS.

Identification of VEGFR2 Binding Site.

We identified a potential antibody binder to VEGFR2 using computational modeling and molecular docking approaches. In this research, we used computational tools to identify a FAB region with the potential to bind the VEGFR receptor and prevent angiogenesis. The VEGFR2-predicted binding site, computed using the ScanNet web server, is highlighted in Figure 4A. The binding site was further confirmed by visual inspection with the X-ray structure of the VEGFR2-VEGF complex (PDB ID: 5T89).(21) VEGFR2 contains seven extracellular Ig-like domains (D1–D7), and VEGF primarily binds to domain 2 (D2), with additional stabilization provided by domain 3 (D3), to initiate receptor dimerization. As shown in Figure 4B, the VEGFR2 receptor is located on the cell surface.  Since the VEGFR2 receptor is located on the surface of cancer cells, only the extracellular domain—the portion exposed outside the cell membrane—was selected for molecular docking simulations. The transmembrane and intracellular domains were excluded because they do not participate in ligand or potential antibody (FAB) binding interactions relevant to this study.

Docking of FAB Candidates to VEGFR2.

Structural docking analysis suggests that the Fab binding site overlaps with the native VEGF-binding interface on VEGFR2, indicating potential competitive inhibition of VEGF–VEGFR2 interaction. In our docking analysis, we ensured that the FAB fragment specifically targets the receptor’s predicted binding site, which overlaps with the natural VEGF binding region. We used the HDOCK software to evaluate the interactions between VEGFR2 and the antibody candidates. As shown in Figure 5, FAB fragments corresponding to PDB IDs 1U8J, 1U8L, 3MLY, 9C2K, 9B71, 9F91, 9HB3, 8ZCA, and 9MK2 bind near or directly at the predicted VEGF-binding site.

Binding Energy Analysis of FAB–VEGFR2 Complexes.

Binding energy was calculated to assess the strength and stability of the FAB–VEGFR complex. We computed the binding energy between VEGFR2 and each FAB fragment, as shown in Figure 6. Among all candidates, 8ZCA exhibited the lowest binding energy, indicating the strongest in silico interaction with VEGFR2. A high-affinity, structurally stable Fab fragment is critical, as rapid dissociation from the receptor could significantly diminish its ability to inhibit receptor activation and downstream signaling pathways. Therefore, a high binding affinity and thermodynamically stable interaction between the Fab fragment and the VEGFR2 receptor are essential to ensure efficient and sustained antibody-mediated inhibition of receptor signaling. VEGFR2. In addition, the multiple interaction types formed between VEGFR2 and the antibody are shown in Figures 7A and 7B.

 

Oncolytic Virus Spread Simulation.

The oncolytic virus spread simulation was performed to visualize how an engineered oncolytic virus spreads through a tumor over time and to evaluate whether viral replication can effectively induce tumor cell death in a realistic microenvironment. Figure 8 illustrates the simulated progression of oncolytic virus infection and lysis within the tumor over a 138-hour period. As the simulation advances, the virus begins by infecting localized tumor regions (pink), which gradually transition into virus-infected zones (blue). These infected areas expand outward as viral replication increases, ultimately leading to widespread tumor cell lysis represented by white and yellow regions. The sequential frames reveal a continuous wave-like progression of viral spread, demonstrating that the virus is capable of penetrating deep into the tumor mass and causing substantial reduction in viable tumor cells. Overall, this model is directly relevant to the study because it provides computational evidence that a VEGFR2-targeted oncolytic virus could successfully replicate within NSCLC tumors and induce targeted tumor clearance, complementing the molecular docking results and angiogenesis inhibition findings.

DISCUSSION.

In this study, we computationally designed and evaluated a VEGFR2-targeted oncolytic virus strategy for inhibiting angiogenesis in NSCLC. Binding site prediction using ScanNet identified the VEGF-interacting extracellular region of VEGFR2 as the primary therapeutic target, which was validated against known structural data. Molecular docking via HDOCK revealed that Fab fragment 8ZCA exhibited the most favorable predicted binding affinity to the VEGFR2 extracellular domain. Interaction analysis using PLIP, RING, and PRODIGY confirmed multiple stabilizing non-covalent interactions, supporting the structural stability of the complex. Additionally, NanoHub-based tumor simulations demonstrated effective viral spread and progressive tumor cell lysis over time, suggesting that integrating a high-affinity VEGFR2-binding Fab into an engineered oncolytic virus could enable both angiogenesis inhibition and direct tumor destruction.

While Fab 8ZCA demonstrated the strongest predicted binding affinity, docking-based binding energies remain theoretical and may not fully represent in vivo biological conditions. Protein flexibility, glycosylation, receptor clustering, tumor heterogeneity, and immune system interactions are not completely captured in computational models. Furthermore, competitive inhibition of VEGF requires not only strong binding but also appropriate epitope overlap and steric blocking efficiency, which must be validated experimentally. The viral spread simulation provides conceptual support but simplifies tumor microenvironment complexity, including stromal barriers, immune infiltration, and vascular architecture. Therefore, although the computational results are promising, biological validation is essential.

Anti-angiogenic therapies targeting the VEGF–VEGFR pathway are already established in oncology; however, many current treatments (such as monoclonal antibodies and tyrosine kinase inhibitors) are associated with resistance, limited tumor penetration, and systemic side effects. Oncolytic viruses, meanwhile, have shown success in selectively lysing tumor cells and stimulating anti-tumor immunity but often lack precise receptor-specific targeting mechanisms. This study addresses a critical gap at the intersection of targeted receptor inhibition and virotherapy by proposing a dual-function system: a VEGFR2-binding viral surface modification that blocks angiogenic signaling while enabling selective viral infection and replication within VEGFR2-overexpressing tumors.

If experimentally validated, this strategy could represent a new class of precision virotherapies that simultaneously disrupt tumor vascular support and induce direct tumor lysis. By combining competitive receptor inhibition with viral replication, the therapy may reduce tumor nutrient supply while amplifying immune activation through tumor antigen release. This dual mechanism could potentially overcome limitations seen in single-modality treatments. Moreover, receptor-targeted viral platforms could be adapted to other overexpressed oncogenic receptors beyond VEGFR2, expanding the translational impact of this computational design framework.

Future work will focus on experimental validation of the VEGFR2–Fab 8ZCA interaction using in vitro binding assays such as surface plasmon resonance (SPR) or bio-layer interferometry (BLI).(22, 23) Viral genome engineering studies are required to integrate the Fab fragment into a functional viral capsid while preserving infectivity and stability. Cell culture experiments in VEGFR2-overexpressing NSCLC models should evaluate infection efficiency, angiogenesis inhibition, and cytotoxicity. In vivo tumor models will be essential to assess therapeutic efficacy, biodistribution, immune activation, and safety. Additionally, combining VEGFR2-targeted oncolytic viruses with immune checkpoint inhibitors may enhance therapeutic synergy and long-term anti-tumor immunity.

CONCLUSION

In conclusion, this study demonstrates the computational feasibility of designing a VEGFR2-targeted oncolytic virus therapy to inhibit angiogenesis in Non-Small Cell Lung Cancer (NSCLC). Through structural modeling, binding-site prediction, molecular docking, and interaction analysis, Fab fragment 8ZCA was identified as the strongest predicted binder to the VEGFR2 extracellular domain, suggesting its potential to competitively block VEGF-mediated receptor activation. By integrating a high-affinity VEGFR2-binding Fab into an engineered oncolytic virus platform, this approach may simultaneously inhibit tumor angiogenesis and promote direct tumor cell lysis. Although the findings are based on in silico analyses and require experimental validation, the results provide a strong computational foundation for the development of targeted, dual-action virotherapies aimed at improving treatment specificity and therapeutic outcomes in NSCLC.

REFERENCES

1. Zarogoulidis, K., et al., Treatment of non-small cell lung cancer (NSCLC). J Thorac Dis 5 Suppl 4, S389-396 (2013).
2. Ruano-Raviña, A., et al., Lung cancer symptoms at diagnosis: results of a nationwide registry study. ESMO Open 5, e001021 (2020).
3. Zhang, S., in Diagnostic Imaging of Lung Cancers, S. Zhang, Ed. (Springer Nature Singapore, Singapore, 2023), pp. 3-49.
4. Folkman, J., Angiogenesis. Annual Review of Medicine 57, 1-18 (2006).
5. Lehner, JM., Schacht, V., Angela, Y., Satzger, I., Gutzmer, R., Erfolgreiche Behandlung eines metastasierenden Melanoms mit Talimogen laherparepvec (T-VEC) bei einer lebertransplantierten Patientin. JDDG: Journal der Deutschen Dermatologischen Gesellschaft 18, 1495-1497 (2020).
6. Zhang, Q.-N., et al., Recombinant human adenovirus type 5 (Oncorine) reverses resistance to immune checkpoint inhibitor in a patient with recurrent non-small cell lung cancer: A case report. Thoracic Cancer 12, 1617-1619 (2021).
7. Frampton, J. F., Teserpaturev/G47Δ: first approval. BioDrugs 36, 667-672 (2022).
8. Mathis, J. M., Stoff-Khalili, M. A., Curiel,  D. T., Oncolytic adenoviruses – selective retargeting to tumor cells. Oncogene 24, 7775-7791 (2005).
9. Andtbacka, R. H., et al., Talimogene Laherparepvec Improves Durable Response Rate in Patients With Advanced Melanoma. J Clin Oncol 33, 2780-2788 (2015).
10. Khuri F. R., et al., A controlled trial of intratumoral ONYX-015, a selectively-replicating adenovirus, in combination with cisplatin and 5-fluorouracil in patients with recurrent head and neck cancer. Nature Medicine 6, 879-885 (2000).
11. Suzuki,  T., et al., A Phase II Study of Regorafenib With a Lower Starting Dose in Patients With Metastatic Colorectal Cancer: Exposure-Toxicity Analysis of Unbound Regorafenib and Its Active Metabolites (RESET Trial). Clin Colorectal Cancer 19, 13-21.e13 (2020).
12. Breitbach,  C. J., Moon, A., Burke, J., Hwang, T. H., Kirn, D. H., A Phase 2, Open-Label, Randomized Study of Pexa-Vec (JX-594) Administered by Intratumoral Injection in Patients with Unresectable Primary Hepatocellular Carcinoma. Methods Mol Biol 1317, 343-357 (2015).
13. Fan, J., Fu, A., Zhang, L., Progress in molecular docking. Quantitative Biology 7, 83-89 (2019).
14. Yan, Y., Zhang, D., Zhou, P., Li, B., Huang, S. Y., HDOCK: a web server for protein-protein and protein-DNA/RNA docking based on a hybrid strategy. Nucleic Acids Res 45, W365-w373 (2017).
15. Eberhardt, J., Santos-Martins, D., Tillack, A. F., Forli, S., AutoDock Vina 1.2.0: New Docking Methods, Expanded Force Field, and Python Bindings. Journal of Chemical Information and Modeling 61, 3891-3898 (2021).
16. Consortium, T. U., UniProt: a hub for protein information. Nucleic Acids Research 43, D204-D212 (2014).

17.Abramson, J., et al., Accurate structure prediction of biomolecular interactions with AlphaFold 3. Nature 630, 493-500 (2024).

18.Tubiana, J., Schneidman-Duhovny, D., Wolfson, H. J., ScanNet: an interpretable geometric deep learning model for structure-based protein binding site prediction. Nature Methods 19, 730-739 (2022).

19. Salentin, S., Schreiber, S., Haupt, V. J., Adasme, M. F.,Schroeder, M., PLIP: fully automated protein–ligand interaction profiler. Nucleic Acids Research 43, W443-W447 (2015).
20. Vangone, A., Bonvin, A.,  PRODIGY: A Contact-based Predictor of Binding Affinity in Protein-protein Complexes. Bio Protoc 7, e2124 (2017).
21. Markovic-Mueller, S., et al., Structure of the Full-length VEGFR-1 Extracellular Domain in Complex with VEGF-A. Structure 25, 341-352 (2017).
22. Hou, W., Cronin, S. B., A Review of Surface Plasmon Resonance-Enhanced Photocatalysis. Advanced Functional Materials 23, 1612-1619 (2013).
23. Shah, N. B., Duncan, T. M., Bio-layer interferometry for measuring kinetics of protein-protein interactions and allosteric ligand effects. J Vis Exp 15, e51383 (2014).

---

---

Posted by buchanle on Friday, June 12, 2026 in May 2026.

Tags: Non-Small Cell Lung Cancer, Oncolytic Cancer, Vascular Endothelial Growth Factor Receptor 2 (VEGFR2)