Cancer frenemy pathways: Inducing immunogenic cell death by exploiting apoptosis-interferon antagonism with oncolytic viruses Authors: Vladimir Bozhdansky, Eran Bacharach, Marcelo Ehrlich Published: 2026 Exploiting frenemy biological programs for synthetic immunogenic lethality in cancer therapy The term “frenemy,” a portmanteau of “friend” and “enemy,” describes a relationship that blends cooperation and antagonism, with overt collaboration often masking underlying tension.1 Analogously, many signaling pathways and biological programs that preserve tissue integrity in normal contexts assume ambivalent roles in cancer, simultaneously sustaining tumor survival while encoding vulnerabilities exploitable therapeutically. Within this framework, synthetic lethality, originally defined in classical genetics as a condition where combined disruption of two genes is lethal whereas perturbation of either alone is tolerated, has expanded from discrete gene pairs to encompass interactions among pathways, functional states, and higher-order processes. In oncology, this broader view has enabled strategies that capitalize on vulnerabilities emerging at the intersection of distinct biological programs altered in malignant cells. More recently, the concept has been extended from simply inducing cell death to engineering synthetic immunogenic cell death, in which combined perturbations not only eradicate tumor cells but also activate antitumor immunity. Importantly, the relevance, selectivity, and robustness of such synthetic lethal relationships do not arise de novo in tumors; rather, they recapitulate native crosstalk and feedback among evolutionarily conserved signaling networks. Consequently, the reciprocal influences that underpin synthetic lethality emerge from co-evolved architectures integrating stress responses, tissue homeostasis, and immune surveillance, such that synthetic lethality reflects a network-level behavior of an adaptive cellular system that ultimately determines whether a given set of perturbations tips the balance toward immunogenic cell death. Apoptosis and type I interferon signaling, a frenemy pair In many biological contexts, apoptosis and type I interferon responses occupy opposite poles of the inflammation-immune activation spectrum. Apoptosis, triggered by extrinsic ligands or intrinsic stress signals, typically supports sterile cell clearance with minimal inflammation, whereas type I interferon responses, which orchestrate the first line of defense against viral infection, promote inflammatory signaling and the activation of immune effector cells that recognize, kill, and remove infected cells. Notably, although a subset of interferon-stimulated genes directly promotes apoptosis, apoptotic caspases in turn cleave and inhibit multiple factors required for interferon induction and downstream signaling,2,3 thereby antagonizing interferon responses during the execution of cell death. Viruses differ in their potential to induce or benefit from apoptosis Viruses differ substantially in both their propensity to trigger or inhibit programmed cell death, including apoptosis, and the extent to which they benefit from it.4 Some viruses actively induce apoptosis late in infection to facilitate the release and dissemination of progeny virions, using viral proteins that engage or mimic pro-apoptotic pathways. Others evolve mechanisms to suppress or delay apoptosis, thereby prolonging host cell survival and maximizing viral replication. As such, in certain cases, apoptosis primarily serves as a host defense strategy that limits viral spread, whereas in others, the controlled and timely induction of apoptotic death becomes a viral ally, contributing to tissue tropism, pathogenesis, and the delicate balance between lytic infection and immune containment. Our recent study In our recent study published in Cell Communication and Signaling,5 we established a strategy to selectively suppress interferon (IFN) signaling in malignant cells by activating caspases with Abt737, a BH3-mimetic compound, to enhance the efficacy of oncolytic virotherapy (OVT) in tumors resistant to viral oncolysis. Using the bladder cancer (BC) cohort from The Cancer Genome Atlas (TCGA), we identified clusters of antiviral IFN-stimulated genes (ISGs) with positively correlated expression, delineating distinct tumor subsets with high or low ISG activity. A similar expression pattern was observed in the urinary tract cohort of the Cancer Cell Line Encyclopedia (CCLE), suggesting that this ISG regulation pattern is cell autonomous. To model this intrinsic heterogeneity, we employed two BC cell lines, T24 and UMUC3, representing high and low ISG expression states, respectively, along with SKMEL3 melanoma cells, which shared phenotypic features with T24. Consistent with elevated constitutive ISG expression, T24 and SKMEL3 cells displayed Janus kinase (JAK)-dependent resistance to the oncolytic orbivirus epizootic hemorrhagic disease virus Tel Aviv University (EHDV-TAU). These findings parallel and extend earlier reports demonstrating the selective tropism of EHDV-TAU for tumor cells with impaired IFN signaling6,7 and its safety and therapeutic efficacy in a syngeneic murine model characterized by epigenetic silencing of IFN-regulated pattern recognition receptors (PRRs).8 Treatment with the Abt737 suppressed IFN-mediated antiviral responses and enabled productive EHDV-TAU infection in tumor cells. The concomitant enhancement of EHDV-TAU infection and caspase activation is shown in Figure 1, which presents maximum-intensity projections of three-dimensional confocal micrographs of T24 cells under different treatment and infection conditions, or their combination, and a volumetric rendering of infected and treated cells. Importantly, Abt737 did not enhance EHDV-TAU infection in normal human foreskin fibroblasts. Mechanistically, Abt737 induced caspase-dependent cleavage of signal transducer and activator of transcription 1 (STAT1) in both T24 and SKMEL3 cell lines and interferon regulatory factor 3 (IRF3) in T24 cells, disrupting both IFN/JAK/STAT signaling and IFN induction. In these models, the combination of EHDV-TAU and Abt737 increased plasma membrane calreticulin exposure (Figure 2) and extracellular ATP release, indicating immunogenic cell death. In contrast, combining Abt737 with the mutant vesicular stomatitis virus VSVΔ51M8 reduced viral titers, underscoring virus-specific effects. These findings showed that OVs such as EHDV-TAU, which was derived from a virus that induces and replicates efficiently in an apoptotic environment,9 benefit from coupling intrinsic apoptosis pathways with targeted inhibition of antiviral signaling. Abt737 enhances caspase activation, EHDV-TAU infection, and T24 cell death T24 cells were left uninfected or infected with EHDV-TAU and treated for 72 h with Abt737, the pan-caspase inhibitor QVD-OPH, or their combination. After staining with DAPI (blue), anti-EHDV-TAU NS3 (green), and cleaved caspase-3 (magenta), confocal stacks were acquired using a spinning disk confocal microscope (63×). The top images show maximum-intensity projections of representative fields under the indicated conditions (scale bars, 40 μm), and the bottom image shows a volumetric rendering of the central image in the second row (50 μm grid). The combination of Abt737 and EHDV-TAU infection induces immunogenic T24 cell death T24 cells were infected and treated as described in Figure 1. Live cells were subsequently stained for surface-exposed calreticulin (ecto-CRT) and with LIVE/DEAD fixable dead cell stain. Stained cells were analyzed by fluorescence-activated cell sorting (FACS). The graphs depict the distribution of cells exhibiting single- or double staining under the indicated conditions, with the upper-right-hand quadrant indicating immunogenic cell death. Conclusions and perspectives Although our study suggests that modulating tumor apoptosis and IFN resistance can broaden the therapeutic range of OVT and improve responses in previously resistant cancers, the broader concept of combining OVs with small molecules to induce synthetic immunogenic cell death may extend beyond BH3 mimetics. Compounds that emulate the intracellular environment of virus-infected cells, such as those activating stress response pathways, could enable OVs to more effectively exploit these cellular contexts (as in De et al.10). The inherent heterogeneity among OVs and the distinct intracellular environments they generate, together with the expanding palette of molecularly targeted therapies, may ultimately form the foundation for developing personalized OVT approaches tailored to individual tumor immunogenic and signaling profiles. Acknowledgments This work was supported by research grants from: the Cancer Biology Research Center (CBRC) of Tel Aviv University (M.E.), Israel Science Foundation (1966/18; M.E.), and German Cancer Research Fund-Israel Ministry of Science and Technology (DKFZ-MOST, CA184/2711; M.E. and E.B.). Declaration of interests The authors declare no competing interests. Declaration of generative AI and AI-assisted technologies in the writing process During the preparation of this work the authors used Perplexity.ai in order to revise/edit the text. After using this tool/service, the authors reviewed and edited the content as needed and take full responsibility for the content of the published article. References