Adenovirus (Ad) is non-enveloped viruses, 90-100 nm in size, with a genome contained within an icosahedral capsid (49). There are 57 known Ad serotypes, with Ad2 and Ad5, both from subtype C, the most widely used for OVs (50, 51). Ad is considered a promising oncolytic virus due to its wide range of serotypes and receptors, high titer production, genomic stability, feasibility of genetic modification and well-characterized replication (52).

The coxsackie and adenovirus receptor (CAR) is a 46 kDa transmembrane glycoprotein in the junction adhesion molecule (JAM) family and serves as a common mediator for both coxsackieviruses B and most Ads (53) ( Figure 2 ). CAR plays a crucial role in epithelial cell adhesion and signal transduction, as well as cancer development (44). Notably, CAR is rarely expressed on normal lung cells but shows variable levels of expression on lung cancer cells (54). H101, with E1b55K and partial E3 deleted, infects cells by the binding of the viral fiber knob with CAR (55). Deletion of E1b55K allows Ad preferentially replicate in p53-deficient cancer cells (56) and E3 genes mainly participate in host anti-virus immune response ( Figure 3 ), however, the mechanism by which H101 selectively replicates in cancer cells remains uncharacterized. A preclinical study confirmed that the lung adenocarcinoma cell line XWLC-05 from Xuanwei highly expresses CAR by RT-PCR and immunocytochemistry staining, thereby the oncolytic adenovirus H101 is able to efficiently infect XWLC-05 and lead to oncolysis in vivo (57). However, the hypoxic environment of some solid tumors is often associated with CAR downregulation, and the RAS-MEK signaling pathway has also been linked to reduced CAR levels (58). Stecker et al. proposed that CAR expression in tumor cells may vary by stage and correlate with tumor aggressiveness, suggesting the need to assess CAR expression and the tumor microenvironment before selecting a viral therapy (59).

CD46, a transmembrane protein, serve as an inhibitor of complement activation and negatively regulates the complement system, as well as the primary receptor for most species B Ad types (60). While CD46 is widely expressed in normal tissues, it is often overexpressed in lung cancer, potentially due to abnormal signal transducers and activators of transcription 3 (STAT3) activation and p53 mutations (61). Additionally, CD46 also protects cancer cells from complement-mediated cell death (62). Studies have shown that CD46 is upregulated in lung adenocarcinomas (A549, Z793) more than in squamous lung cancers (QG56, NCI-H520), but the latter has a relatively higher levels of CAR (46). Other complement inhibitory proteins, such as decay-accelerating factor (DAF, or CD55), also play a role in protecting tumor cells from immune surveillance (63).

Desmoglein-2 (DSG2), another major receptor for Ad, a transmembrane glycoprotein belonging to the cadherin family (20). DSG2 has been shown to be involved in cell-cell adhesion and tumorigenesis, and is also overexpressed in NSCLC (64). Sun et al. analyzed lung adenocarcinoma (LUAD) patients and corresponding normal tissues to assess DSG2 expression. Combining their results with the data from TCGA and Oncomine, showed that high DSG2 expression positively correlates with tumor size, lymph node metastasis and TNM stage (65). A meta-analysis demonstrated that high DSG2 expression is associated with poor overall survival (OS) in NSCLC patients (66). Another preclinical and clinical study found that DSG2 overexpression promoted LUAD cell proliferation and migration, potentially through the regulation of EGFR and Src phosphorylation, activation of the PAK1 signaling pathway, and alterations in the tumor microenvironment (TME). DSG2 overexpression was also associated with increased resistance to the EGFR tyrosine kinase inhibitor Osimertinib. However, subsequent studies confirmed that DSG2 expression was not statistically associated with tumor size, differentiation, lymph node metastasis or stages (67), which contradicts the fundings of Sun et al.

Additionally, Ad can use integrins as entry receptor, others like SARs, CD80 and CD86 (61), MHC-I (68), etc. The receptors mentioned above are representative examples. Ad uses a variety of receptors for cell entry, lacking inherent specificity for tumor cells.

Herpesvirus (HSV) is large enveloped dsDNA virus, with 9 known types, including the commonly studied HSV-1 and HSV-2 from the α-Herpesviridae family (74). HSV replicates in cellular nucleus without integrating into the host genome (75), a feature that enhances its safety profile as an OV.

HSV-1 glycoprotein D and B mediate viral entry by binding to specific receptors on the surface of cancer cells (76). gD serves as a ligand protein for most α-Herpesviridae receptors, which binds to 3 primary receptors: herpes virus entry medium (HVEM), nectin-1, and 3-O-sulfated heparan sulfate (3-OS-HS) (77, 78). gB is a trigger protein that is responsible for viral fusion. Reported receptors binding to gB and mediate entry including paired immunoglobulin-like type 2 receptor-α (79), myosin-9 (80), myelin-associated glycoprotein (81). However, the exact mechanisms by which gB receptors mediate viral entry remain incompletely understood and only a few studies have characterized myosin-9 and myelin-associated glycoprotein in NSCLC (82, 83).

HVEM, belonging to the tumor necrosis factor receptor (TNFR), is expressed predominantly on immune cells, and functions as a primary receptor for HSV-1 and HSV-2, excluding other α HSVs (84, 85). Notably, Ren et al. evaluated 527 NSCLC samples and 56 NSCLC cell lines, suggest that HVEM is overexpressed in NSCLC patients (positive rate 18.6% & 48.2%) with N2 lymph node metastasis or advanced stages, but its expression levels is not capable to predicting OS (86, 87). HVEM is a key immune checkpoint, interacting with B and T lymphocyte attenuator (BTLA) and CD160 (BY55) to trigger inhibitory signals (47). It is independent from PD-1/PD-L1 network and may contribute to immune evasion in lung cancer, making it a promising therapeutic target (88).

Nectin-1 is a cell adhesion protein belonging to the immunoglobulin superfamily, which functions as an entry receptor for a big part of α HSVs (89). The nectin family mainly consists of nectins 1-4, but not much research has been done on how HSV infects cells with the help of nectin-1 or its expression in lung cancer cells. In contrast, nectin-4, a receptor for Measles virus, has demonstrated significant predictive and applied value (90). Nectin-2 also facilitates the entry of certain HSV strains and it might be used in lung cancer diagnosis since high nectin-2 expression in LUAD has been found to be associated with recurrence after surgery (91–93).

HSV-1 was the first oncolytic virus to be genetically engineered for therapeutic use. T-VEC, based on HSV-1 JS-1 strain, is modified to the deletion of neurovirulence factor ICP34.5 and ICP47 genes, also armed with GM-CSF gene, which enhancing the recruitment of APCs and immune filtration of TME (94). OH2 is a novel OV derived from HSV-2 HG52 strain, with the same modification strategy as T-VEC (95). HSV-2-based OVs for lung cancer remain in preclinical development, with no evidence distinguishing HSV-1 from HSV-2 in this context (96, 97).

Vaccinia virus (VV) is a large dsDNA virus from the poxviridae family, approximately 192 kb in length, with a characteristic asymmetric brick-like complex structure (69). It can enter the host cells through membrane fusion or via receptor-mediated endocytosis in acidic environments, though the involvement of a specific host cell receptor remains unclear (26, 98). VV is considered a safe oncolytic agent, as demonstrated by its successful use in smallpox vaccines and its cytoplasmic replication, preventing host genome integration (98).

VV’s tumor selectivity depends largely on the thymidine kinase (TK) gene, which supports viral replication. Since TK is overexpressed in cancer cells and minimally expressed in normal somatic cells, oncolytic VVs with TK deletions have been engineered. Additionally, VV-infected tumor cells secrete viral proteins that activate the EGFR-RAS pathway, further promoting TK synthesis and enhancing VV replication (27).

In addition, VV’s large genome can accommodate substantial exogenous DNA without impairing its replication capacity (99). JX-594, the most well-known VV derivative, has been engineered with a GM-CSF gene insertion and TK deletion, showing promise for intravenous administration (73). Moreover, VV holds significant potential for future tumor vaccine development, particularly due to its ability to deliver therapeutic genes and stimulate robust immune responses (100).

Coxsackievirus (CV), a member of the Picornaviridae family, is a positive-sense single-stranded RNA virus with a genome of approximately 7.4 kb, encapsulated by icosahedral capsid proteins (101). CV possesses several features that make it a promising candidate for lung cancer therapy, including multiple receptor targets, ease of genetic modification, cytoplasmic replication, and the ability to specifically target hypermutated molecular pathways in lung cancer (102).

Kirsten rat sarcoma homolog (KRAS) mutations are present in approximately 30% of NSCLC patients, with 20%-40% in LUAD and only 5% in squamous NSCLC (5, 61). CV-B3, a well-studied oncolytic virus, effectively targets KRAS-mutant LUAD cells (A549, H23, H2030) with minimal effects on normal lung epithelial and EGFR-mutant LUAD cells (H1975, PC-9, H3255, H4006) (103, 104). However, CV-B3 can cause severe viral myocarditis and pancreatitis, limiting its therapeutic potential, and further genetic modifications are needed to reduce its toxicity (105).

CVs primarily enter cells via receptors such as CAR, intercellular adhesion molecule 1 (ICAM-1), decay accelerating factor (DAF, or CD55), KRM1 and SCARB2 (scavenger receptor class B member 2, also known as lysosomal integration membrane protein-2, LIMP-2) (32, 33, 106, 107).

As mentioned above, CAR is a common receptor for Ad and CV, and lung cancer cells that express CAR on their surface would be attacked by both of them theoretically. Future studies may need to find the expression of this receptor on more lung cancer cell lines and patient-derived tumor tissue or primary cells to maximize the oncolytic effect of CV that use this receptor to infect lung cancer cells. Deng et al. found that KRAS mutations downregulate CAR expression by activating the ERK1/2 signaling pathway (103).

ICAM-1 (CD54) is an inducible glycoprotein involved in cell adhesion during immune and inflammatory responses (108). CV-A21, is the most researched recently with ICAM-1 (109). Infection with CVB upregulates ICAM-1 expression and increases production of the pro-inflammatory cytokines, including IL-6, IL-8 and TNF-α (110). NSCLC cell lines with high ICAM-1 expression are sensitive to CV-A11-mediated cytotoxicity, while DAF expression levels do not correlate with cytotoxic effects (32, 111). Preclinical and clinical trials showed that the CV-A21-based OVs product V937, which preferentially lyses ICAM-1 upregulated NSCLC cells, is currently in Phase I clinical trials, but the clinical efficacy intravenously administered V937 with pembrolizumab does not appear to be superior to that of monotherapy (34, 112).

DAF/CD55 is a co-receptor that is expressed in both lung cancer and normal lung fibroblast cell lines (113). It has been shown that DAF assists in the entry of ICAM-1 and CAR (114). DAF can act as a lower affinity attachment site, enhancing virus presentation, or as a virus binding site for subsequent higher affinity binding to ICAM-1 and CAR (63, 111). Overexpression or mutations of epidermal growth factor receptor (EGFR) can be detected in 10%-20% NSCLC patients (115). Shao et al. treated H1395 and H322M NSCLC cells with EGF for 24h, and suggest that EGFR activation increases the expression of CD55, but not CD46, upregulated CD55 expression inhibited the complement system and cytokine secretion required for CD8⁺ T cell activation, and CD55 levels were negatively correlated with infiltration of M1 macrophages and CD8⁺ T cells in human lung cancer specimens (n=24), which indirectly promoted tumor growth and could use predicted patient prognosis (116). Therefore, EGFR mutations may enhance CV infection by upregulating CD55, but this also contributes to immune suppression within the tumor microenvironment (TME), which may facilitate tumor progression.

KRM1 is a widespread membrane-anchoring protein located on the cell surface and in the intracellular membrane which is recently identified as an important entry receptor of a major subset of CV-As (117). Most studies on KRM1 focus on its interaction with CV-A10, yet its expression in lung cancer cells and the oncolytic effects of KRM1-targeting CVs remain unexplored, likely due to incomplete understanding of CV-A10’s infection mechanisms and pathogenesis (117, 118).

SCARB2 is a type-III transmembrane protein which mediates the translocation and reorganization of the endosomal/lysosomal compartment membranes (119). However, the expression conditions and the mechanism of how coxsackieviruses use SCARB2 to infect lung cancer cells needs to be further investigated.

Seneca Valley virus (SVV) is a ss (+) RNA virus that selectively infects and kills neuroendocrine SCLC cells (35, 36, 120). Preclinical studies showed that repeated passaging of SVV in SCLC cell cultures enhances its cytolytic activity over time, suggesting increased anti-tumor efficacy (121). Reovirus, an enveloped dsRNA virus (122), infects host cells via receptor-mediated endocytosis, primarily binding to junction adhesion molecule A (JAM-A), which is overexpressed in NSCLC and thus makes it a target for oncolytic virotherapy (31). Additional OVs, including Measles virus (MV) (123), Newcastle disease virus (NDV) (38), Vesicular stomatitis virus (VSV) (124), and Semliki Forest virus (SFV) have been explored for their potential in lung cancer therapy in preclinical settings (125).

In conclusion, OVs can target lung cancer cells by binding to overexpressed surface receptors. Different viruses show varying potential and limitations in lung cancer therapy, and optimizing viral genetic modifications and receptor selection is key to improving efficacy. Ads offer strong tumor-targeting potential due to their multiple serotypes, ease of genetic modification, and ability to use various receptors for cell entry. However, they lack tumor specificity, are prone to be cleared by immune responses, and show off-target effects, including liver accumulation in vivo. HSVs avoid host genome integration risk, their large genomes allow for gene insertions and effective immune activation. However, limited expression of key receptors like HVEM and Nectin-1 in lung cancer and potential neurotoxicity remain concerns for clinical application. VV have large genomes capable of carrying exogenous genes, and their cytoplasmic replication avoids integration into the host genome. Their safety is supported by widespread vaccine use, and deletion of the thymidine kinase (TK) gene enhances tumor cell selectivity. However, unidentified specific receptors limit their targeting, and replication depends on the high metabolic state of tumor cells, reducing efficacy in low-proliferation tumors. CVs can target multiple receptors and certain CVs show selective efficacy in KRAS-mutant lung adenocarcinomas. Its small genome allows easy modification, and cytoplasmic replication reduces integration risks. However, it can cause toxic effects like myocarditis and pancreatitis, requiring further modification to minimize side effects. Additionally, receptor expression (e.g., CAR, ICAM-1) varies across lung cancer types, affecting oncolytic efficacy. Other viruses, such as Seneca Valley virus (SVV), have potential for targeting specific lung cancer subtypes (e.g., small-cell lung cancer) and have shown antitumor activity in early studies. However, clinical data are scarce, mechanisms are unclear, and most are in early development, necessitating further validation of their efficacy and safety.

Future research should focus on understanding the molecular mechanisms of viral entry into tumor cells to enhance specificity and reduce toxicity. Additionally, the therapeutic potential of OVs in clinical settings remains underexplored ( Table 3 ) and warrants extensive clinical trials to validate efficacy and safety.

Notably, OVs are usually used in combination with other therapies, and cancer cells often undergo molecular and phenotypic changes in response to treatment. Changes in surface receptor expression have been observed after radiotherapy or chemotherapy, though results are inconsistent across studies (126–128).

Harrington et al. assessed the combining effects of oncolytic Ad and the external beam radiotherapy, suggested that CAR and integrins were upregulated in colorectal (HCT116) and head-neck (SIHN-5B) cancer cell lines after radiation (129), but Geoerger et al. (130) reported that radiation does not upregulate CAR and integrins expression in glioma cell lines. that radiation CAR and integrins expression in glioma cell lines. Another experimental research also showed radiotherapy fails to increase CD46 receptor expression in glioblastoma for measles virus (131). Wu et al. (126) demonstrated that pre-medication of camptothecin or doxorubicin downregulated the CAR expression in tumor lines including H1299 (lung), HCT 116 (colon) and BxPC-3 (pancreas), this chemotherapy-induced downregulation was observed in patients undergoing chemoradiotherapy prior to colorectal cancer resection. However, Sakhawat et al. showed that CAR expression was enhanced by using cisplatin which improving the infection of Ad in breast cancer cells lines (127). Further research is needed to elucidate receptor expression changes in cell lines and primary tumors following neoadjuvant chemoradiotherapy, and the intracellular signaling pathways mediating these changes.

Therefore, assessing receptor expression before and after combination therapy is crucial to enhancing its clinical benefit, additional studies are required to clarify the impact of combination therapy on OVs receptor expression across different tumor types.