Engineering systemic oncolytic virotherapy through immune-compatible delivery and programmable tumor cell death

The challenge of oncolytic virotherapy

Oncolytic viruses (OVs) have attracted great interest as a new class of cancer therapeutics because they can both directly lyse tumor cells and stimulate antitumor immunity. Over the past two decades, multiple OV platforms—including adenovirus, herpes simplex virus, vaccinia virus, and reovirus—have advanced into clinical testing, demonstrating favorable safety profiles and the capacity to reshape the tumor immune microenvironment.¹ However, clinical success has been largely confined to intratumoral administration, as exemplified by talimogene laherparepvec (T-VEC), the first FDA-approved OV for melanoma.^2,3 For deep-seated or metastatic tumors, effective systemic delivery is essential but remains a major barrier. Neutralizing antibodies, innate immune clearance, and rapid loss of viral particles from the circulation substantially limit bioavailability following systemic administration.⁴ Even when systemically administered OVs reach tumor sites, productive infection and intratumoral spread are often limited by dense stromal architecture—particularly in desmoplastic tumors such as pancreatic ductal adenocarcinomas—as well as by intrinsic antiviral defenses within tumor cells.⁵ These limitations have driven a number of efforts to engineer OVs with improved immune evasion, tumor targeting, and immunostimulatory payloads; however, a broadly effective strategy that integrates systemic delivery with efficient intratumoral amplification has yet to be established. Collectively, these challenges highlight a critical question for the field: how can an OV platform be designed to circulate systemically, evade immune attack, selectively accumulate in tumors, and subsequently expand its activity while reprogramming the local immune environment?

Engineering immune-compatible cloaking for systemic OV delivery

In a recent study, our group addressed this challenge through an immune-compatible cloaking strategy that enables systemic OV administration while minimizing off-target immune recognition.⁶ Donor HEK293 cells were genetically edited to knock out B2M and Class II major histocompatibility complex transactivator (CIITA) (ablating HLA-I and HLA-II expression) and to knock in CD47 and HLA-E, thereby reducing T cell recognition, macrophage phagocytosis, and natural killer cell-mediated clearance. These donor cells were further engineered to express tumor-specific chimeric antigen receptors (CARs). Upon viral infection and mechanical extrusion, the resulting immunoengineered nanovesicles (iNVs) encapsulate genetic engineered OVs (GOV), generating particles termed iNV-GOV.

This cloaking strategy provides multiple layers of protection and targeting. The host-derived membrane shields viral capsids from neutralizing antibodies and innate immune sensors, while CD47 and HLA-E confer resistance to phagocytic and cytotoxic clearance. In addition, CAR expression enables active tumor targeting based on antigen recognition. Together, these features extend viral circulation time, reduce systemic immunogenicity, and enhance selective accumulation within tumor tissues following systemic administration (commonly intraperitoneally in PDAC models), while intravenous (i.v.) feasibility was also demonstrated in other tumor settings.

Controlled intratumoral amplification and immune remodeling

Systemic delivery alone is insufficient if viral replication and dissemination remain restricted within the tumor microenvironment. To address this limitation, we introduce an inducible intratumoral amplification mechanism based on pyroptotic tumor cell death. Pyroptosis is a lytic and inflammatory form of programmed cell death that rapidly disrupts cellular membranes, facilitating the release of intracellular contents, including newly produced viral particles. By coupling a genetically encoded pyroptotic effector to an inducible control system, this strategy enables spatial activation of pyroptosis within infected tumor cells. Controlled activation promotes efficient virion release and secondary infection of neighboring tumor cells, thereby overcoming physical and biological barriers that limit intratumoral viral spread. Importantly, external regulation of this process provides temporal precision and an additional safety layer, minimizing the risks associated with constitutive expression of cytolytic payloads during systemic administration.

In addition to enhancing OV infection, pyroptotic tumor cell death reshapes the tumor immune microenvironment. In humanized orthotopic patient-derived xenograft models, iNV-GOV activated by ultrasound to trigger pyroptosis expression then accelerated OV release, leading to increased infiltration and activation of cytotoxic CD8⁺ T cells, enhanced dendritic cell maturation, and a shift in macrophage polarization toward an M1-like phenotype. Regulatory T cells were reduced, collectively reversing the immunosuppressive features of the tumor environment. These immune changes translated into robust systemic antitumor immunity, as evidenced by increased tumor-specific T cell responses and durable tumor growth control. Together, these effects link intratumoral amplification with immune activation, positioning controlled pyroptosis as both a mechanistic and immunological amplifier of systemic oncolytic virotherapy.

Significance

A key implication of this work is that the iNV-GOV platform can be readily combined with existing cancer therapies. By inducing immunogenic tumor cell death and increasing T cell infiltration, iNV-GOV enhances tumor sensitivity to immune checkpoint blockade. In preclinical models, combination treatment with anti-PD-1 antibodies resulted in improved tumor growth control and prolonged survival compared with anti-PD-1 monotherapy, supporting the use of oncolytic virotherapy in combination with immune checkpoint blockade. In addition to its effects on immunotherapy, iNV-GOV also improved the response to chemotherapy in desmoplastic tumors such as pancreatic ductal adenocarcinoma. These tumors are often resistant to treatment because dense stromal barriers limit drug penetration. Remodeling of the stromal and immune microenvironment by pyroptosis-accelerated virotherapy may help reduce these barriers and improve the activity of conventional chemotherapies. Together, these findings suggest that pyroptosis-accelerated virotherapy may function as a versatile foundation for rational combination treatment strategies to enhance their effectiveness.

Despite these promising results, several important questions remain to be addressed before this strategy can be fully translated. The durability of immune remodeling following viral clearance, the balance between antitumor and antiviral immune responses, and the scalability and consistency of manufacturing immune-engineered vesicle-coated viruses will require careful evaluation in immunocompetent and clinically relevant settings. Moreover, optimal selection of tumor-targeting ligands and treatment parameters for different tumor types will likely be necessary to maximize benefit while minimizing off-target effects. Addressing these challenges will be critical to determine whether pyroptosis-accelerated systemic virotherapy can mature into a broadly applicable, combinatorial backbone for cancer treatment.

Translational perspectives

In this work, we illustrate a generalizable strategy to address two long-standing barriers in oncolytic virotherapy: systemic delivery and intratumoral amplification. The modular iNV cloaking design can be adapted to different viral backbones and tumor-targeting ligands, while the pyroptosis payload may be replaced or complemented by alternative inducible effector programs. A clinically established external trigger enables spatial and temporal control of intratumoral activity, which is particularly relevant for systemically administered virotherapies. Clinical translation will require careful evaluation of long-term safety in immunocompetent, antigenically relevant models. Notably, gasdermin-driven pyroptosis is intrinsically pro-inflammatory; off-tumor activation could increase the risk of local tissue injury and systemic inflammatory toxicities,⁷ underscoring the need for stringent tumor-restricted control and dosing strategies. Comprehensive testing in models that account for stromal barriers, antigen heterogeneity, and chronic inflammation will be essential to translational potential.

Conclusion

This study demonstrates that immune-compatible iNV cloaking can enable systemic oncolytic virotherapy while preserving effective intratumoral amplification through controllable gasdermin-mediated pyroptosis. Important questions remain regarding the durability of immune remodeling, the balance between antitumor and antiviral immunity, manufacturing consistency, and inflammatory safety. The work sets the stage for further exploration of combining systemic delivery with controllable effector programs to enhance virotherapy. Collectively, this study represents a step toward expanding the therapeutic reach of OVs in cancer immunotherapy.

Declaration of interests

The authors declare no competing interests.