Article Text
PDF
Abstract
Cytokines are small proteins that regulate the growth and functional activity of immune cells, and several have been approved for cancer therapy. Oncolytic viruses are agents that mediate antitumor activity by directly killing tumor cells and inducing immune responses. Talimogene laherparepvec is an oncolytic herpes simplex virus type 1 (oHSV), approved for the treatment of recurrent melanoma, and the virus encodes the human cytokine, granulocyte-macrophage colony-stimulating factor (GM-CSF). A significant advantage of oncolytic viruses is the ability to deliver therapeutic payloads to the tumor site that can help drive antitumor immunity. While cytokines are especially interesting as payloads, the optimal cytokine(s) used in oncolytic viruses remains controversial. In this review, we highlight preliminary data with several cytokines and chemokines, including GM-CSF, interleukin 12, FMS-like tyrosine kinase 3 ligand, tumor necrosis factor α, interleukin 2, interleukin 15, interleukin 18, chemokine (C-C motif) ligand 2, chemokine (C-C motif) ligand 5, chemokine (C-X-C motif) ligand 4, or their combinations, and show how these payloads can further enhance the antitumor immunity of oHSV. A better understanding of cytokine delivery by oHSV can help improve clinical benefit from oncolytic virus immunotherapy in patients with cancer.
- Cytokine
- Immune modulatory
- Oncolytic virus
This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See http://creativecommons.org/licenses/by-nc/4.0/.
Statistics from Altmetric.com
Introduction
Cancer immunotherapy is an established approach to cancer treatment that uses the patient’s immune system, both innate and adaptive, to effectively recognize and eliminate cells. One promising avenue of cancer immunotherapy involves using oncolytic viruses (OVs).1–5 Among all OVs, the oncolytic herpes simplex virus (oHSV) stands as the most advanced in clinical development.3 6 oHSVs are designed to replicate selectively within cancer cells, triggering tumor cell elimination through virus-induced oncolysis while spreading in the tumor microenvironment to infect and lyse tumor cells without harming healthy cells and tissues (figure 1A). Further, the local release of tumor-associated antigens and viral-mediated danger signals induces systemic antitumor immunity that can further drive immune-mediated tumor clearance at distant sites of infection, a so-called “abscopal” effect.6
(A) Mechanism of action of cytokine/chemokine-armed oHSVs. Armed oHSVs can enter healthy or normal host cells. However, the deletion or mutation of viral genes (such as ICP34.5, ICP6) and the presence of cellular antiviral defense mechanisms in healthy cells (such as the antiviral PKR-eIF2α pathway) prevent oHSV replication. As a result, healthy cells remain non-permissive to oHSV replication and infection (upper panel). In contrast, in cancer cells, antiviral mechanisms like the PKR-eIF2α pathway are often defective, allowing viral replication (ie, permissiveness). This leads to the killing of cancer cells and the release of tumor (and viral) antigens. These antigens provoke an antitumor immune response. Through oncolysis, oHSVs spread to surrounding cancer cells, a process known as in-situ amplification and spread. This further augment the immune response, often referred to as the in-situ vaccine effect (lower panel). The virus-induced in-situ vaccine effect can be enhanced by the viral expression of cytokines/chemokines from the armed oHSVs in the tumor microenvironment. The cytokine/chemokine expression results in the further orchestration of the immune response, including the recruitment of various cells into the tumor. These recruited cells contribute to the production of a cancer-killing effect. This figure was created with BioRender.com. (B) Cytokine/chemokine-expressing oHSVs in the tumor microenvironment and their antitumor immune effects. (1) GM-CSF-armed oHSVs demonstrated significant antitumor efficacy not only in the injected tumor but also in non-injected tumors in preclinical and clinical studies. GM-CSF stimulated granulocyte and monocyte differentiation, enhanced T-cell recruitment at the tumor site, locally reduced regulatory T cells (Tregs) and myeloid-derived suppressor cells (MDSCs), and induced TH9 cell expansion and NK cell activation, along with dendritic cell (DC) maturation. (2) IL-15-armed oHSVs significantly enhance infiltration and activation of NK and CD8+ T cells in tumor and inhibit tumor growth and prolong survival. (3) IL-12-armed oHSVs significantly boosted antitumor immune responses, promoting intratumoral CD8+ T-cell infiltration, inducing Th1 (T-bet+) cell differentiation, reducing CD4+FoxP3+ regulatory T cells (Tregs) and Th2 cell activity, polarizing macrophages towards antitumor M1 phenotypes, increasing release of IFN-γ, TNF-α, GM-CSF, and IL-2, and decreasing CD31+ tumor vascularity. (4) IL-2-armed oHSVs greatly improved antitumor efficacy compared with unarmed oHSV. Elevated IL-2 levels drove differentiation and expansion of activated T cells toward Th1/CTLs, enhancing TNF-α and IFN-γ release, while potentially increasing Treg cells and suppressing IL-17A. (5) IL-18-armed oHSVs did not exhibit improved antitumor efficacy compared with unarmed oHSVs. However, when combined with vHsv-IL-12, vHsv-IL-18 significantly enhanced antitumor immunity synergistically with IL-12 expression, increasing IFN-γ and GM-CSF release, and promoting the infiltration of CD4+ and CD8+ T lymphocytes and NK cells. (6) FLT3L-armed oHSVs significantly improved long-term survival rates in glioma-bearing mice by enhancing IFN-α-expressing plasmacytoid and conventional DC infiltration in the tumor microenvironment. (7) CCL-5-armed oHSVs enhance antitumor immune responses by recruiting CD8+, CD4+, and NK cells into tumor microenvironment. (8) TNF-α-armed oHSVs effectively inhibited injected and non-injected tumors, stimulating T-cell mediated antitumor immunity, promoting M1 polarization of tumor-associated macrophages, inhibiting Treg cells and facilitating the recruitment of antigen-presenting cells and neutrophils. (9) CCL2-armed oHSVs demonstrated inconsistent antitumor efficacy, but the combination therapy of CCL2 expression with IL-12 expression showed superior antitumor effects characterized by recruiting CD4+ and CD8+ T lymphocytes to the tumor site. CCL2 expression also activated monocytes, macrophages, memory T lymphocytes, and NK cells, leading to the release of pro-inflammatory cytokines, such as IL-1, IL-6, and TNF-α. (10) CXCL4-armed oHSVs significantly outperformed unarmed viruses in controlling tumor growth, resulting in a substantial extension of median survival. CXCL4 expression activated DCs, leading to CD4+ T-cell proliferation/activation and IL-17 production. This figure was created with BioRender.com. APC, antigen-presenting cells; CCL2, chemokine (C-C motif) ligand 2; CCL5, chemokine (C-C motif) ligand 5; CTLs, cytotoxic T lymphocytes; CXCL4, chemokine (C-X-C motif) ligand 4; DC, dendritic cell; ECM, extracellular matrix; Fib, fibroblasts; FLT3L, FMS-like tyrosine kinase 3 ligand; GM-CSF, granulocyte-macrophage colony-stimulating factor; IFN, interferon; IL, interleukin; MΦ, macrophage; NK, natural killer; PKR, protein kinase R; oHSV, oncolytic herpes simplex virus; TNF, tumor necrosis factor.
OHSV-induced antitumor immunity occurs through two interconnected mechanisms. First, the direct oncolysis of tumor cells can eliminate at least a segment of viable tumor cells. Tumor cell lysis may be optimized by selecting native viral species that exhibit high levels of in vitro cytopathic or cytolytic effects against the tumor cell of interest. Further improvements in tumor-selective replication and lysis can be genetically engineered by deleting non-essential viral genes or by expressing pro-lytic gene(s) that help promote tumor cell lysis. The second mechanism of antitumor activity with oHSV occurs through immunogenic cell death, a process in which cell lysis leads to the release of tumor-associated antigens and danger factors. This allows the presentation of tumor antigens (and viral antigens) to the immune system, and the tumor antigens, in turn, may prime tumor-reactive T-cell responses.7 8 The local infection with oHSV also induces a chemokine cascade that can recruit tumor-specific CD8+ T cells to the tumor microenvironment (TME).9 This phenomenon is commonly called the “in-situ cancer vaccine effect”.3 Further, the induction of oHSV-specific T-cell responses may lead to a “bystander” effect in which virus-specific T cells eliminate oHSV-infected tumor cells. The therapeutic potential of oHSV, both the in-situ cancer vaccine and the bystander effect, can be further improved by arming the virus with cytokine transgenes that can modulate the immune response (figure 1A).3 9
Cytokines are a general family of small proteins and glycoproteins (generally around 5–25 kD) that bind to specific receptors resulting in intracellular signaling and ultimately mediating immune function. Cytokines are thought to act in an autocrine, paracrine, and endocrine fashion allowing rapid regulation of local and systemic host immune responses.10 Cytokines have also exhibited potent single-agent antitumor effects against a variety of human cancers.10 Therapeutic cytokines typically require very high supraphysiologic doses, resulting in on-target, off-tumor systemic toxicity. Thus, much of the current drug development with cytokine agents has focused on modifying the protein construct to allow for more specific receptor binding or for higher local retention of the cytokine to avoid systemic toxicity. Due to their small size and potent immune activity, encoding human cytokine gene(s) in various OVs has gained significant attention for further optimizing the antitumor activity of OVs, including oHSVs.2
Many genetically modified oHSV expressing various immunostimulatory cytokines have been described. These cytokines include granulocyte-macrophage colony-stimulating factor (GM-CSF), interleukin 12 (IL-12), FMS-like tyrosine kinase 3 ligand (FLT3L), tumor necrosis factor α (TNF-α), interleukin 2 (IL-2), interleukin 15 (IL-15), interleukin 18 (IL-18), chemokine (C-C motif) ligand 2 (CCL2), chemokine (C-C motif) ligand 5 (CCL5), and chemokine (C-X-C motif) ligand 4 (CXCL4) (figure 1B; table 1). Of 22 oHSVs that are/were in clinical trials, 10 of 12 of the armed expressed a cytokine. Preclinical studies indicate that cytokine expression, such as IL-2 and IL-12, from oHSV, is limited in time and to the tumor,11 12 compared with other OVs where systemic cytokines (eg, IL-2) are detected, as well as associated toxicity.13 14 Based on the published clinical evidence, oHSV expression of a cytokine, such as GM-CSF15–19 or IL-12,20 demonstrated a favorable safety profile following intratumoral injections in patients or following intracerebral injections in non-human primates, respectively. Among all cytokine-expressing oHSVs, oHSV talimogene laherparepvec (T-VEC, IMLYGIC; previously named OncoVEXGM-CSF and JS1/34.5−/47−/GM-CSF), expressing the GM-CSF cytokine, has achieved Food and Drug Administration (FDA) approval in the USA in 2015 for treating melanoma that recurs after surgery,6 17 marking a significant milestone in oncolytic immunovirotherapy for cancer. In addition, a non-oncolytic adenovirus encoding interferon (IFN)-α, nadofaragene firadenovec, was FDA-approved in 2022 to treat non-muscle invasive bladder cancer.21
Oncolytic herpes simplex viruses encoding cytokines or chemokines
This review will discuss the effects of the aforementioned cytokines when expressed by various oHSVs, encompassing preclinical studies evaluating antitumor efficacy, and clinical studies to provide valuable insights into the use of these cytokine-encoding oHSVs as promising agents to treat cancer.
GM-CSF expressing oHSVs
GM-CSF, a hematopoietic growth factor, is produced from various cell types on immune stimulation, including macrophages, T cells, epithelial cells, fibroblasts, and tumor cells.22 Recombinant GM-CSF was approved by the FDA in 1991 for neutrophil recovery following induction chemotherapy in patients with acute myeloid leukemia. GM-CSF is also thought to act as an immune adjuvant, enhancing humoral and cellular antitumor responses by stimulating differentiation of granulocytes and monocytes, modulating interactions between T cells and antigen-presenting cells (APCs), maturing dendritic cells (DCs), and activating and proliferating T cells.22 GM-CSF-primed DCs trigger long-lasting tumor-specific immune responses by activating CD4+ and CD8+ T cells to recognize tumor-associated antigens (figure 2). However, GM-CSF, in some instances, can promote tumor growth and metastasis by recruiting neutrophils to the TME, which suppresses antitumor immunity by inhibiting the activity of CD8+ T cells.22 Nonetheless, GM-CSF remains one of the most common cytokines included in OV development.23
(A) oHSV-infected tumor microenvironment. GM-CSF-armed or IL-12-armed oHSVs infect and lyse cancer cells, resulting in the release of GM-CSF or IL-12 in the tumor microenvironment (TME), respectively, for induction of antitumor immunity. (B) GM-CSF expression and its associated signaling pathways in antitumor immunity. GM-CSF released from oHSVs binds to its receptor, which is a heterodimer, consisting of α chains that specifically bind GM-CSF and βc chains that are responsible for signaling. Neither the α chains nor the βc chains of the GM-CSF receptor contains a tyrosine (Y) kinase catalytic domain. The βc chain of the GM-CSF receptor links to tyrosine kinase JAK2. Once the α chain binds to GM-CSF, it subsequently binds to the βc chain to form a multimer. The polymerization of the receptor leads to the activation of JAK2, and the activated JAK2 consequently phosphorylates the tyrosine residues on the βc chain of the receptor. The phosphorylated tyrosine on the βc chain recruits STAT-5 by binding to the SH2 domain of STAT5. JAK2 further phosphorylates STAT-5 and activates the JAK-STAT pathway, which induces pro-inflammatory cytokines that activate T lymphocytes within the TME. Furthermore, JAK2 activates PI3K and initiates the PI3K-Akt signaling pathway, which promotes the proliferation and survival of monocytes, macrophages, and granulocytes within the TME. The phosphorylated tyrosine on the βc chain of the GM-CSF receptor also recruits the adaptor protein SHC, which in turn activates RAS/MEK/ERK signaling pathway to induce signaling in the nucleus, enhancing the differentiation and proliferation of macrophages, neutrophils, and dendritic cells within the TME. (C) IL-12 expression and its associated signaling pathways in antitumor immunity. IL-12 (IL-12p70) is composed of two covalently linked subunits, p35 and p40. The released IL-12 from oHSVs binds to IL-12 receptor, which is composed of two subunits, IL-12Rβ1 and IL-12Rβ2. JAK2 and TYK2 become trans-phosphorylated after the association of IL-12p40 and IL-12p35 with IL-12Rβ1 and IL-12Rβ2, respectively. Phosphorylated IL-12Rβ2 binds to STAT4, which then dimerizes with another STAT4 molecule. STAT4 homodimers translocate to the nucleus and promote transcription of the IFN-γ gene. The IL-12 and IFN-γ induce the activity and proliferation of macrophages, NK cells, and T lymphocytes in TME. This figure was created with BioRender.com. DC, dendritic cell; GM-CSF, granulocyte-macrophage colony-stimulating factor; IFN, interferon; IL, interleukin; Mφ, macrophage; Neu, neutrophil; NK, natural killer cell; oHSV, oncolytic herpes simplex virus.
Preclinical studies with GM-CSF-armed oHSVs
Antitumor effects of an oHSV armed with GM-CSF (JS1/34.5−/47−/GM-CSF; OncoVEXmGM-CSF)
Liu and colleagues constructed oHSV JS1/34.5−/47−/GM-CSF armed with human or mouse (m)GM-CSF on HSV-1 clinical isolate JS1, which exhibited improved killing against human tumor cell lines compared with laboratory strain 17+.24 Deletion of ICP47 (JS1/34.5−/47−) increased expression of HSV-1 US11, improved antitumor activity, and increased major histocompatibility complex class I on the surface of infected cells. Similar antitumor effects were observed in virus-injected tumors with or without mGM-CSF, while non-infected contralateral tumors were somewhat smaller with GM-CSF.24 The complete remission of non-injected tumors signifies systemic antitumor immunity, the so-called abscopal effect, and these animals also rejected tumor re-challenge, indicating the development of tumor-specific immunologic memory.24 The antitumor efficacy of OncoVEXmGM-CSF was further enhanced by immune checkpoint inhibitors (ICIs), such as anti-cytotoxic T lymphocyte antigen 4 (anti-CTLA-4) or anti-programmed death 1 (anti-PD-1), with the combination therapy eliciting significant inhibition of B16F10-mNectin tumor growth and lung metastasis, and extension of median survival compared with monotherapies.8
Antitumor effects of other oHSVs armed with GM-CSF (R-121, NV1034)
Lucia and colleagues examined the antitumor efficacy of the next-generation targeted oHSV armed with GM-CSF (R-121) in hHER2-transgenic C57BL/6 mice bearing Lewis lung carcinoma(LLC1) cells expressing human HER2 (HER2-LLC1).25 The GM-CSF expression (R-121) was not statistically better than the unarmed oHSV (R-113) in eliciting antitumor efficacy.25 Similar to R-121, NV1034 (oHSV armed with GM-CSF) was not statistically different from its control counterpart (ie, NV1023, an oHSV without GM-CSF expression) in inducing antitumor effects in murine prostate tumor models26 or mediating a whole-body antitumor memory response in a murine squamous cell carcinoma model.27 Overall, these three preclinical studies question the role of GM-CSF expression in inducing antitumor activity. In contrast, NV1034 treatment significantly reduced CT26 liver metastasis and subcutaneous tumor growth of CT26 and Hepa 1–6 tumors compared with unarmed NV1023.28 29 Thus, the role of GM-CSF as an immune adjuvant for OVs remains speculative.
Immune activity of T-VEC
Bommareddy and colleagues investigated the mechanism of action of T-VEC (with human GM-CSF) in melanoma models. T-VEC infection of human melanoma cell lines in vitro induced significant immunogenic cell death with the release of damage-associated molecular patterns, including HMGB1, ATP, and translocated ecto-calreticulin. The sensitivity of melanoma cells to T-VEC infection was inversely correlated with stimulator of interferon genes (STING) expression.7 Protein kinase R (PKR) and cyclic guanosine monophosphate-adenosine monophosphate synthase (cGAS)/STING pathways are two critical anti-viral mechanisms in host cells that resist infection with DNA viruses.30 Their high expression in melanoma cells contributes to resistance against T-VEC infection and oncolysis. Gene knockouts demonstrated that STING, but not cGAS or PKR, contributes to resistance against T-VEC-mediated oncolysis.
Prior studies indicated that low expression of STING in cancer contributes to resistance to PD-1 blockade.7 31 In vivo, T-VEC induced significant tumor regression in low STING mouse melanoma tumors (D4M3A melanoma), while no significant effect was observed with anti-PD-1 antibody. In a bilateral D4M3A model, T-VEC treatment significantly reduced the growth of both injected and non-injected contralateral tumors compared with mock-treated mice.7 Flow cytometry revealed significant increases in CD3+ and CD8+ T cells in both injected and non-injected tumors following T-VEC treatment. While the levels of HSV-1 gB-specific CD8+ T cells were similar in injected and non-injected tumor lesions, melanoma antigen-specific CD8+ T cells and CD45+ PD-1+ cells were significantly higher in injected tumors. Gene expression analysis demonstrated upregulation of genes involved in antigen presentation, costimulation, chemokines, and CD8+ T-cell activation in both injected and non-injected tumors, indicating that T-VEC can induce systemic inflammation and recruit virus-antigen and tumor-antigen specific CD8+ T cells into the tumors.7 Overall, this study shows the effectiveness of T-VEC in low STING-expressing melanoma cells refractory to anti-PD-1 treatment and suggests that STING status may be a potential biomarker for T-VEC response. Further clinical studies are needed to better validate this finding.
Clinical studies with GM-CSF-armed oHSV (oHSVGM-CSF/T-VEC)
Phase I study of oHSVGM-CSF
Hu and colleagues conducted a Phase I clinical trial of OncoVEXGM-CSF (T-VEC) in patients with refractory cutaneous or subcutaneous metastases from breast, gastrointestinal adenocarcinoma, malignant melanoma, or epithelial cancer of the head and neck, with 13 patients in the single-dose group and 17 patients in the multidose group. On enrollment, 11 patients were seronegative for HSV, while 19 were seropositive.15 In the single-dose group, the leading side effects were grade 1 fever and associated constitutional symptoms, which were more pronounced among HSV-seronegative patients.15 Other common side effects included fatigue, low-grade anorexia, nausea, and vomiting. Most patients experienced inflammation at the injected lesion, with more pronounced inflammation and erythema observed in seronegative patients. This led to a decrease in the starting virus dose in the multidosing group to 106 pfu/mL in order to seroconvert patients. The exact viral dose varied from patient-to-patient based on the size of the tumor lesions. For instance, for lesions with a diameter of ≤1.5 cm, >1.5 to ≤2.5 cm, or >2.5 cm, up to a maximum of 1 mL, 2 mL, or 4 mL per dose, respectively, were used.15 Similar side effects were observed in the multidose and single-dose groups. 14 of 19 post-treatment biopsies with tumor displayed tumor necrosis, and all of those stained for HSV. After treatment, three patients had stable disease, six patients had tumors flattened (injected and/or un-injected lesions), and four demonstrated inflammation in both injected and non-injected tumors.15 In summary, OncoVEXGM-CSF was well tolerated even with multidosing, and exhibited biological activity, paving the way for multiple Phase II clinical trials in various tumor types.15
Phase II study of oHSVGM-CSF
Building on Phase I data,15 Senzer and colleagues conducted a Phase II clinical trial with JS1/34.5−/47−/GM-CSF (T-VEC) for patients with unresectable stages IIIc and IV metastatic melanoma.16 This study enrolled a total of 50 patients, with 74% having prior non-surgical treatments, such as dacarbazine/temozolomide or IL-2. Patients received 24 intratumoral injections, beginning with up to 4 mL of 106 pfu/mL and 3 weeks later, up to 4 mL of 108 pfu/mL every 2 weeks. Adverse effects related to oHSV treatment were observed in 85% patients, and were mostly grade 1–2 symptoms, such as mild influenza-like syndrome, chills, fatigue/malaise, nausea, vomiting, and headache.16 Treatment resulted in an overall objective response rate (ORR) of 26%, with eight complete (CR), five partial (PR), and two surgical complete responses (sCR), including regressions of both injected and distant visceral lesions. Remarkably, 92% of these responses persisted for 7–31 months. The 1-year survival rate for patients with PR, CR, or sCR was 93%.16 Biopsies from regressing lesions revealed a link between therapeutic response and the presence of IFN-γ-producing MART-1-specific CD8+ T cells and reduced regulatory T cells (Tregs), indicating that OncoVEXGM-CSF was able to induce antitumor immunity.32
Phase III study with T-VEC
In a Phase III clinical trial (OPTiM trial), T-VEC was compared with GM-CSF in patients with advanced melanoma.17 18 A total of 436 patients were randomized in a 2:1 ratio to receive intratumoral T-VEC (n=295) or subcutaneous recombinant GM-CSF (n=141). T-VEC demonstrated a significantly higher durable response rate (16.3% vs 2.1%), ORR (26.4% vs 5.7%), and longer median overall survival (OS) (23.3 months vs 18.9 months) than GM-CSF.17 Notably, 16.9% of T-VEC patients achieved CR, compared with only 0.7% with GM-CSF. The efficacy of T-VEC was more pronounced in stage IIIB–IVM1a melanoma, with a 5-year survival rate of 48.9%, compared with 15.1% in stage IVM1b/c disease. Common adverse effects included fatigue, chills, fever, nausea, and influenza-like illness, with grade 3/4 treatment-related adverse effects occurring in 11.3% T-VEC-treated and 4.7% GM-CSF-treated patients. Vitiligo was the most frequently reported immune-related adverse effect observed in T-VEC-treated patients.18 These promising results with T-VEC justify continued efforts to improve its efficacy in melanoma, particularly as a combination partner with other immunotherapies.18 What is not clear is the role, if any, that GM-CSF played in the therapeutic outcomes. However, based on these results, T-VEC was approved by the FDA in 2015 followed by approvals in Europe, Australia, and Israel.
T-VEC for non-melanoma skin and other cancers
Intralesional treatment with T-VEC is being currently investigated in a Phase II clinical trial for solid tumors with skin metastasis, including Merkel cell carcinoma (MCC), both as monotherapy and in combination with hypo-fractionated radiotherapy (NCT02819843). Another Phase II study assessing the antitumor efficacy of T-VEC in combination with nivolumab (anti-PD-1) for advanced or refractory non-melanoma skin cancers (NCT02978625). Additionally, a Phase I single-center study is evaluating the efficacy of T-VEC in locally advanced non-melanoma skin cancers (NCT03458117).33 T-VEC clinical trials in other cancers include breast (NCT03802604, NCT02779855, NCT03554044), head and neck (NCT02626000), liver (NCT02509507), pancreatic (NCT00402025, NCT03086642), rectal (NCT03300544), and sarcomas (NCT03069378, NCT03921073, NCT02923778), many involving combinations with other therapeutics.
oHSVGM-CSF in combination with chemoradiotherapy
Harrington and colleagues conducted a Phase I/II study using JS1/34.5−/47−/GM-CSF in combination with radiotherapy and cisplatin for untreated stage III/IV squamous cell cancer of the head and neck.34 All 17 enrolled patients received uninterrupted full-dose radiotherapy and chemotherapy after virus injection. Patients were divided into four cohorts: (1) 106 pfu/mL on four occasions; (2) 106 pfu/mL on one occasion followed by 107 pfu/mL on three occasions; (3) 106 pfu/mL on one occasion followed by 108 pfu/mL on three occasions; and (4) 106 pfu/mL on one occasion followed by 108 pfu/mL on three occasions, and (1–3) up to a maximum of 4 mL per dose and (4) up to a maximum of 8 mL per dose.34 End-of-treatment CT scans confirmed a response in 14 patients, with 4 experiencing CR and 10 PR. Combination therapy resulted in all patients remaining free of locoregional disease, compared with 50–70% historically after radiotherapy/chemotherapy, and a 70.5% OS rate.34 Unfortunately, a planned Phase III clinical trial was never initiated.
T-VEC in combination with ICI
A Phase II trial compared T-VEC plus ipilimumab (anti-CTLA-4) to ipilimumab alone (1:1) in patients with advanced melanoma. Among 198 patients, 38 patients in the combination arm and 18 patients in the ipilimumab arm showed an objective response. Responses extended beyond injected lesions, with 52% in the combination arm and 23% in the ipilimumab arm showing decreases in visceral lesions.35 This combination demonstrated greater antitumor activity without added safety concerns compared with ipilimumab alone.35
Ribas et al reported results from a Phase 1b study of a combination T-VEC and pembrolizumab in 21 patients with advanced melanoma.31 In this study, patients received two doses of T-VEC and then started pembrolizumab after the second T-VEC dose. The treatment was well tolerated with no dose-limiting adverse events and the most common side effects were fatigue, fevers, and chills. There was a confirmed ORR of 62% with 33% of patients achieving a CR. In addition, responding patients were found to have increased local CD8+ T cells and programmed death ligand 1 (PD-L1) expression with a high baseline IFN-γ gene expression profile. This study justified a larger prospective, global, randomized clinical trial to evaluate the benefit of a combination T-VEC and pembrolizumab in patients with melanoma.
The KEYNOTE-265 clinical trial (NCT02263508) equally randomized 692 anti-PD-1-naïve patients with stage IIIB–IVM1c unresectable melanoma to treatment with T-VEC and pembrolizumab or intratumoral placebo and pembrolizumab. T-VEC was given at the approved dosing and scheduled for five doses and then every 3 weeks thereafter. Pembrolizumab was given at the standard approved dosing and schedule and was started concurrently with T-VEC. There were two primary endpoints of the study including progression-free survival (PFS) and OS based on modified response evaluation criteria in solid tumors (RECIST) V.1.1 criteria. T-VEC and pembrolizumab did not improve PFS or OS. The ORR was 48.6% for the combination treatment compared with 41.3% for placebo and pembrolizumab.36 The reason for the negative outcome of this study is not clear but may relate to an underestimation of the response rate of pembrolizumab alone since all patients were required to have accessible lesions or may have been related to concurrent drug delivery since PD-1 blockade may promote earlier clearance of the virus. Another possible explanation is that in the Phase 1 study, T-VEC was administered for two doses prior to pembrolizumab, whereas in the Phase III trial, they were given concurrently starting on day 1. It is possible that early checkpoint blockade could result in more intense or more rapid anti-viral immune responses and premature viral clearance. HSV-1 encodes immune suppressive genes, and it is possible these may block effective anti-PD-1-mediated antitumor responses. Further investigation is needed to better understand the interaction of OV and ICI.
Ongoing trials assess the combination of T-VEC with PD-1/PD-L1 inhibitors for treating MCC, melanoma, and other cutaneous malignancies (NCT02978625, NCT02965716). Of 54 T-VEC clinical trials, 16 are/were in combination with different ICIs (2 with anti-PD-L1, 3 with anti-CTLA-4, and 11 with anti-PD-1) beginning in 2013. Several other GM-CSF-armed oHSVs like OrienX010, RP1, and OH2 are in clinical trials as monotherapy or in combination for melanoma, skin cancer, bladder cancer, and other solid tumors (table 1). So far, despite the strong rationale for the combination, the clinical results have been modest.
IL-12 expressing oHSVs
IL-12 is a pro-inflammatory cytokine that bridges innate and adaptive immunity. It is a heterodimeric protein (IL-12-p70) with subunits p35 and p40.37 The IL-12 receptor, composed of IL-12R-β1 expressed constitutively or IL-12R-β2 inducibly in natural killer (NK) cells, T cells, and B cells.37 IL-12 has diverse activities, promoting T and NK cell proliferation, CD4+ T-cell differentiation towards Th1, and IFN-γ production (figure 2), which exerts T-cell-mediated killing of tumor cells and inhibits angiogenesis, establishing IL-12 as a master regulator of antitumor immunity.
Early clinical trials demonstrated clinical activity of IL-12, but were hindered by severe side effects such as hepatotoxicity and neutropenia.38 These issues, coupled with IL-12’s narrow therapeutic window, tempered enthusiasm for its systemic use in cancer patients.37 Nonetheless, IL-12 remains a subject of interest, particularly in immunovirotherapy where localized expression by oHSV can mitigate side effects.37 As such, many IL-12 expressing oHSVs have been developed as anticancer agents, which we have presented below.
oHSV-IL-12 (M002)
M002 is an ICP34.5-deleted oHSV that contains the insertion of two copies of a murine IL (mIL)-12 expression cassette, which is transcriptionally regulated by the murine early growth response 1 promoter.39
The antitumor potential of M002 was evaluated in the Neuro-2a neuroblastoma model in immunocompetent A/J mice, sensitive to HSV infection.39 In vitro, infection of Neuro-2a cells resulted in efficient production of IL-12 at a physiologically relevant level (1 ng/mL). M002 treatment significantly extended median survival compared with parental R3659 virus, and was associated with increased infiltration of CD4+ and CD8+ T cells and macrophages into tumors.39 In a vaccine study, irradiated Neuro-2a cells infected with M002 were administered before and after tumor implantation, a prime-boost strategy, that led to long-lasting antitumor immunity and significant survival benefits. The lack of an unarmed oHSV-infected irradiated tumor cell control makes it unclear whether the effect was due to oHSV, IL-12 expression, or a combination of both.40
In a murine breast cancer brain metastases model, M002 significantly extended median survival compared with R3659 or mock treatments,41 indicating the importance of IL-12 in the efficacy. Two new murine sarcoma models, SARC-28 and SARC-45, with most cells expressing viral entry receptors, were sensitive to M002-mediated cytotoxicity in vitro.42 Intratumoral M002 treatment of athymic mice bearing SARC-28 or SARC-45 tumors significantly extended median survival, with 30–75% long-term survivors compared with none in the mock group, an effect not seen with treated tumors in immunocompetent BALB/c mice where M002 was no better than R3659.42 However, M002 was significantly better than R3659 in inducing beneficial immune responses, increased infiltration of CD8+ T cells and activated monocytes, reduced myeloid-derived suppressor cells (MDSCs), and improved CD8:MDSC and CD8:Tregs ratios.42 M002 was efficacious against a spontaneous ovarian cancer model in MISIIR-TAg mice, where it was better at controlling metastases (81.2% vs 18.2% in the phosphate-buffered saline (PBS) group; p=0.008), leading to a significant extension of OS. This enhanced efficacy of M002 was linked to a significant induction of ovarian tumor antigen-specific CD8+ T-cell responses.43
A human version of M002 was prepared, M032, which expresses human IL-12.20 Safety evaluation in non-human primates showed that direct injection of M032 into the brain was safe.20 M032 is currently undergoing clinical testing in patients with glioblastoma (GBM)44 (NCT02062827). Omar and colleagues reported a median OS of 151 days in pet dogs with sporadic gliomas treated with intracranial injection of M032, with no significant adverse events or dose-limiting toxicities observed.45 These studies will offer insights for designing concurrent human clinical trials (NSC 733972).
oHSV-IL-12 (NV1042)
NV1042 is derived from HSV-1 R7040, also known as NV1020. R7020 was developed as an HSV-1 and HSV-2 vaccine. R7020 contains deletions in UL56 and the joint region (one copy of ICP0, ICP4, and γ34.5) and insertion of an HSV-2 fragment containing glycoproteins D, J, G, and I in the deleted joint region. It underwent extensive safety studies and a human clinical trial where it did not elicit sufficient immunogenicity,46 and was repositioned as oHSV NV1020.47 NV1020 has been administered by intrahepatic artery infusion in two clinical trials.48 To create NV1042, mIL-12, regulated by a hybrid α4-TK promoter, was inserted adjacent to the HSV-2 sequence.27
Intratumoral delivery of NV1042
NV1042 outperformed NV1023 (no cytokine expression) or NV1034 (GM-CSF expression) in reducing the growth of squamous cell carcinoma (SCC) and prostate tumors,26 27 and protecting the cured mice from SCC re-challenge27; highlighting the superiority of IL-12 over GM-CSF in inducing antitumor immunity. However, in a metastatic colorectal liver model, both NV1042 and NV1034 were equally effective and T-cell dependent.28 In the CT26 colorectal cancer model, NV1042 was significantly better than NV1023.49 Combining NV1042 with vinblastine significantly reduced prostate xenograft tumor burden and decreased CD31+ endothelial cells compared with monotherapy or NV1023 plus vinblastine, demonstrating the anti-angiogenic activity of IL-12.50
Intravenous delivery of NV1042
For disseminated metastatic tumors, systemic delivery of oHSV should enable the targeting of multiple tumor sites or those not accessible to direct injection. Intravenously administered NV1042 was evaluated in two prostate cancer models, implanted lung metastases and spontaneously arising tumors and metastases in TRAMP transgenic mice, where it significantly inhibited primary tumor growth and lung metastases compared with NV1023.51 52 Following intravenous delivery, NV1042 was found transiently in vital organs lacking metastases; however, no visible abnormalities were observed,52 affirming the safety of intravenous NV1042. These studies underscore the importance of IL-12 expression and intravenous delivery of oHSV-IL-12 in controlling metastatic burden. Despite the promising results in preclinical studies, NV1042 was not translated to the clinic.
oHSV-IL-12 (G47Δ-mIL-12)
G47Δ-mIL-12, a triple-mutated oHSV lacking ICP6, ICP47, and γ34.5, that contains an insertion of mIL-12 subunits p35 and p40 connected by a bovine elastin motif in the ICP6 region12 (table 1).
G47Δ-mIL-12 as monotherapy
In vitro, G47Δ-mIL-12 efficiently infected and replicated in mouse 005 GBM stem-like cells (005 GSCs), similarly to unarmed G47Δ-empty.12 In vivo, G47Δ-mIL-12 treatment modestly extended survival of C57BL/6 mice bearing orthotopic 005 GBM, which was associated with increased IFN-γ production, angiogenesis inhibition, and reduced 005 GSCs and Tregs in the tumor compared with G47Δ-empty.12 In a separate study with the 005 model,53 G47Δ-mIL-12 treatment also led to increased infiltration of CD3+ T cells and M1-like macrophages and decreased percentages of 005 GSCs and Tregs, and an increased CD8+ T effector cell to Tregs ratio (figure 3). The antitumor effects of G47Δ-mIL-12 were dependent on T cells.12 G47Δ-mIL-12 treatment of orthotopic mouse malignant peripheral nerve sheath tumors (MPNSTs) in the sciatic nerve significantly extended survival and reduced neurologic deficits compared with G47Δ-empty.54
Immunological alterations in the tumor microenvironment following oHSV-IL-12 treatment as monotherapy and combination therapy. (1) oHSV-12 monotherapy enhances antitumor immune responses characterized by increased intratumoral infiltration of CD8+ T cells, enhanced Th1 (T-bet+) cell differentiation, decreased CD4+FoxP3+ regulatory T cells (Tregs), increased polarization of macrophages toward an antitumor M1 phenotype, elevated levels of IFN-γ and IL-12, and reduced CD31+ tumor vascularity. (2) Immune checkpoint inhibitors significantly enhance the effectiveness of oHSV-IL-12 monotherapy. The combination therapy markedly increases infiltration of CD3+ and CD8+ T cells, reduces infiltration of CD4+FoxP3+ Tregs, enhances Th1 differentiation (T-bet+), and augments influx of M1-like antitumoral macrophages. (3) oHSV-IL-12 in combination with angiogenic inhibitors leads to enhanced tumor oncolysis, decreased tumor vascularity (CD34+ vessels), and greater infiltration of macrophages (CD68+). (4) However, oHSV-IL-12 in combination with a chemotherapy agent, temozolomide (TMZ), yields no significant survival benefit in the combination group when compared with oHSV-IL-12 monotherapy, rather TMZ treatment abrogates the antitumor efficacy of oHSV-IL-12. This is due to TMZ treatment, which substantially impairs immune responses, characterized by the reduction of T cells (CD4+ and CD8+) and macrophages (CD68+) in the tumor microenvironment, along with CD4+ T-cell reduction systemically in the spleen. This figure was created with BioRender.com. IFN, interferon; IL, interleukin; oHSV, oncolytic herpes simplex virus; TAM, tumor-associated macrophage.
Due to the promising antitumor effects of IL-12 expression, an oHSV expressing human IL-12 (G47Δ-hIL-12) is currently in the pipeline to be tested in the clinic. In fact, the non-IL-12-expressing parent oHSV (G47Δ) has been approved in Japan for the treatment of recurrent GBM.55
G47Δ-mIL-12 in combination with anti-angiogenic agents
The anti-angiogenic activity of G47Δ-mIL-12 was evaluated in combination with axitinib, a vascular endothelial growth factor receptor (VEGFR) tyrosine kinase inhibitor, in human and mouse GSC-derived GBM models.56 In an immunodeficient human MGG123 GBM model, both G47Δ-mIL-12 and axitinib individually improved survival. Combining them further extended survival, reducing tumor vascularity, inducing apoptosis and necrosis, increasing macrophage infiltration, and inhibiting the PDGFR/ERK pathway (figure 3).56 In the mouse 005 GBM model, monotherapies (G47Δ-mIL-12 or axitinib) significantly extended survival compared with mock, and the combination further improved survival.56 However, CTLA-4 inhibition did not improve the efficacy of the combination therapy.56 This lack of improvement might be attributed to axitinib inhibiting T-cell infiltration into the tumor,56 counteracting the beneficial effects of G47Δ-mIL-12 and anti-CTLA-4 in stimulating T cell-mediated antitumor immune responses.12
Due to potential renal toxicities associated with systemic anti-angiogenic therapy, such as axitinib,57 Zhang and colleagues combined G47Δ-mIL-12 with G47Δ-mAngio (which expresses angiostatin, an anti-angiogenic polypeptide) in human U87 and MGG4 GSC GBM models.58 In the U87 model, both G47Δ-mIL-12 and G47Δ-mAngio monotherapies significantly extended survival compared with G47Δ-empty or PBS, with the combination further improving survival. This was attributed to increased viral spread, reduced VEGF expression, decreased CD31+ blood vessels, and decreased macrophages within the tumor.58 Construction of oHSV expressing both IL-12 and angiostatin would eliminate the need for FDA approval of two separate viruses.
G47Δ-mIL-12 in combination with ICI
Anti-CTLA-4 enhanced the efficacy of G47Δ-mIL-12 in the 005 GBM model; increasing CD3+ T-cell infiltration and elevating the CD8+ T effector to CD4+FoxP3+ Treg ratio in the TME. However, this combination only resulted in 13% long-term survivors. Similar modest survival extensions were observed with other ICIs like anti-PD-1 or anti-PD-L1 when combined with G47Δ-mIL-12.53 A triple combination therapy was explored to enhance antitumor immunity: G47Δ-mIL-12 plus anti-PD-1 plus anti-CTLA-4, resulting in 89% long-term survivors.53 These cured mice developed immunological memory and remained tumor-free for up to 9 months.53
Efficacy of the triple combination therapy was associated with significant beneficial effects (compared with dual antibody or mock treatments); reduced 005 GSCs, increased CD8+ T-cell infiltration and proliferation, decreased Tregs, elevated CD8+ T to CD4+FoxP3+ ratio, and increased M1-like antitumoral macrophages (figure 3).53 Depletion of immune cells revealed that CD4+ T cells were essential and CD8+ T cells and macrophages contributed to the synergistic curative effects of triple combination therapy.53 The promising therapeutic outcomes of the triple combination therapy provide a solid basis for its application in the clinic.
G47Δ-mIL-12 in combination with a chemotherapeutic agent
Temozolomide (TMZ) chemotherapy is the standard of care for GBM, but its efficacy is limited due to resistance from the expression of O6-methylguanine DNA methyltransferase (MGMT).59 Saha et al conducted a study combining TMZ with G47Δ-mIL-12 in the MGMT-positive 005 GBM model.60 In vitro, TMZ plus G47Δ-mIL-12 effectively killed 005 GSCs. In vivo, G47Δ-mIL-12 monotherapy extended median survival significantly. Low-dose TMZ alone (7.5 mg/kg) or the combination (TMZ and virus were given concurrently) did not improve survival outcomes, likely due to MGMT expression.60 The use of regular-dose TMZ (25 mg/kg, equivalent to 75 mg/m2/day dosing regimen in humans) surprisingly abrogated the therapeutic benefits of G47Δ-mIL-12. Immune response analysis revealed reduced T cells and macrophages in the brain and spleen after TMZ treatment (figure 3).60 Overall, TMZ, when given concurrently, negatively affected immunovirotherapy. Thus, it is important to carefully consider the timing and dosing of combination therapies when planning clinical trials.
oHSV-IL-12 (R115)
Introducing genetic deletions and/or mutations in the oHSV genome somewhat reduces replication efficiency and viral virulence. To overcome these deficits, Menotti and colleagues developed R-LM113, a fully virulent oHSV with no genome deletions but instead featuring an insertion of an anti-HER2 antibody fragment.61 R-LM113 was further modified to express IL-12, creating R-115.61 62
In the LLC1 (HER2-LLC1) model,62 HER2-IL-12-armed R115 outperformed unarmed R-LM113 in reducing tumor growth, indicating the importance of IL-12 expression. Long-term survivors demonstrated virus-induced immunological memory.62 R115’s superior antitumor efficacy correlated with T-cell polarization toward Th1 and increased IFN-γ expression in the TME.62
In an orthotopic brain tumor model derived from mHGGpdgf-hHER2 cells,63 R115 (and its unarmed control R-LM113) significantly extended median survival compared with mock. Although no significant differences were observed between R115 and R-LM113, R115 yielded 27% long-term survivors compared with none with R-LM113. These survivors rejected two consecutive tumor re-challenges with HER2-expressing and non-HER2-expressing tumor cells, demonstrating R115’s ability to induce immunological memory, irrespective of HER2 expression.63 R115 treatment resulted in significant production of hHER2 receptor-specific antibodies in vivo compared with control but not R-LM113 treatment. Additionally, R115-mediated tumor eradication involved increased IFN-γ expressing cells and a significant localization of CD4+ and CD8+ T cells within the tumor masses compared with R-LM113 treatment.63
In summary, these two independent studies suggest that the receptor-retargeted IL-12-expressing oHSV, which is fully virulent and non-attenuated, can safely be used as oncolytic virotherapy for cancer.
Other IL-12-expressing oHSVs
In addition to the aforementioned oHSV-IL-12s, several other oHSVs encoding IL-12 and often other transgenes, such as VT1092M (+anti-PD-1),64 T2850 (+anti-PD-1),65 T-mfIL-12,66 and vHsv-IL-12,67 were developed and tested in various cancer models. IL-12, alongside GM-CSF, is a leading cytokine in the context of engineered oHSV therapy for cancer, consistently demonstrating safety, superior anticancer efficacy, and beneficial antitumor immunity compared with non-IL-12-expressing oHSVs. This underscores the potential of IL-12-expressing oHSVs for cancer treatment. Notably, both murine (M002) or human (M032) IL-12-expressing oHSVs have shown safety when administered intracerebrally in non-human primates,20 with M032 currently undergoing clinical evaluation.44
FLT3L expressing oHSVs
FLT3 ligand (FLT3L) is a cytokine and growth factor, which plays a vital role in the development and expansion of DCs, NK cells, and B cells.68 FLT3L can control tumor growth by primarily increasing the number of DCs at the tumor site.68 Barnard and colleagues constructed G47Δ-Flt3L (an oHSV expressing FLT3L) and evaluated its efficacy in a murine glioma model.69 Compared with G47Δ-empty (lacking FLT3L expression), G47Δ-Flt3L significantly improved the long-term survival rate of glioma-bearing mice from 0% to 40%. G47Δ-Flt3L’s efficacy relies on oHSV’s oncolytic effect, release of tumor-associated antigen and danger signals, and FLT3L-mediated increase of plasmacytoid and conventional DCs in the TME. This impressive anti-glioma efficacy of G47Δ-Flt3L warrants its clinical investigation.69
TNF-α expressing oHSVs
TNF-α is a pro-inflammatory cytokine primarily secreted by monocytes, macrophages, NK, mast, and T cells, as well as non-immune cells like fibroblasts and endothelial cells.70 TNF-α exerts antitumor effects by directly causing tumor cell apoptosis and/or necrosis, disrupting neo-angiogenesis, and collapsing tumor vasculature.70 Additionally, it promotes M1 macrophage polarization, activates T cells by inhibiting Tregs, and recruits/stimulates DCs, neutrophils, monocytes, and macrophages to tumor sites.70 71 However, in vitro studies have reported that low-dose TNF-α can promote the proliferation of some malignant cells, and in vivo, it may act as an autocrine tumor growth factor.72 Despite its potent antitumor effects, clinical use of TNF-α has been limited due to its high systemic toxicity.73 To overcome this, Han and colleagues developed a second-generation oHSV for localized tumor delivery of TNF-α under the cytomegalovirus (CMV) or HSV Us11 promoter.74 In a murine lymphoma model, Us11-driven TNF-α demonstrated similar efficacy in injected tumors and enhanced efficacy in un-injected tumors compared with CMV-driven TNF-α, however, the use of the Us11 promoter reduced TNF-α’s toxic effects in vivo.74 This study illustrates how oncolytic activity and regulated expression for localized TNF-α delivery is safe and effective.
IL-2 expressing oHSVs
IL-2 plays a pivotal role in promoting proliferation and survival of T and NK cells.75 On stimulation, IL-2 is predominantly synthesized by CD4+ T cells and, to a lesser extent, by CD8+ T cells.75 IL-2 can also be produced by NK cells, natural killer T (NKT) cells, monocytes, and DCs.2 Notably, IL-2 was the first cytokine approved by the FDA for metastatic melanoma and renal cell carcinoma. However, systemic IL-2 therapy can have significant adverse effects, limiting its use in cancer treatment. Carew and colleagues harnessed oHSV G207 as a helper virus to package an HSV amplicon plasmid carrying the IL-2 gene (HSV-IL-2), designated as G207(IL-2).76 G207(IL-2) produced superior antitumor efficacy compared with G207 or HSV-IL-2 in an SCC VII carcinoma flank model. In a subsequent experiment, no significant difference in antitumor efficacy was observed between G207(IL-2) and G207 plus HSV-IL-2 treatment groups.76 Both G207 and IL-2 armed G207 triggered a significant elevation of CD4+ and CD8+ cells within the tumors.76 Altering IL-2 to avoid binding to the IL-2Rα (CD25), which is found at high levels on Tregs, has been employed to avoid Treg proliferation,77 and may improve the therapeutic effects of IL-2-armed oHSV.
IL-15 expressing oHSVs
IL-15 is a pleiotropic cytokine for developing, surviving, proliferating, and activating NK cells and CD8+ T cells. IL-15 and its receptor (IL-15Rα) are co-expressed in APCs and form a complex on the plasma membrane that can be trans-presented to NK cells and T cells bearing the IL-2 receptor βγc complex.78 In contrast to IL-2, IL-15 can stimulate only NK and effector T cells without stimulating Tregs.78 Gaston and colleagues constructed two oHSVs encoding either mIL-15 (J100) alone or with mIL-15 receptor alpha (mIL-15Rα, J100D) to determine whether co-expression of these proteins is required for the production of bioactive mIL-15 from oHSV. Enhanced mIL-15 production was observed in J100D as compared with J100. Soluble mIL-15/IL-15Rα was detected from J100D infected cells and was bioactive, which stimulated NK cell proliferation and reduced the viability of GL261 and CT-2A cells. No neuropathologic effects were observed after direct injection of J100 and J100D into the brains of CBA/J mice; however, these oHSVs were not tested for in vivo antitumor efficacy.79
An oHSV type 2 virus (oHSV2-mIL-15CherryFP) expressing murine IL-15 was constructed using the CRISPR/Cas9 system. Murine IL-15 released from the cell culture supernatant enhanced T-cell killing of CT26-GFP tumor cells. Intratumoral injection of oHSV2-mIL-15CherryFP inhibited tumor growth in the CT26-iRFP and BGC823-iRFP models.80 An oHSV type 1 expressing human IL-15/IL-15Rα sushi domain fusion protein (OV-IL-15C) was more efficacious in inhibiting tumor growth and prolonging survival of GBM-bearing mice compared with parental oHSV without transgene.81 The combination of OV-IL15C and EGFR-CAR NK cells synergistically inhibited GBM growth and significantly enhanced survival compared with either monotherapy, which correlates with increased intracranial activation and infiltration of NK and CD8+ T cells.81
Hu and colleagues constructed five armed oHSV2 expressing IL-12, IL-15, GM-CSF, PD-1v, or IL-7×CCL19.82 These oHSVs have deletions of ICP34.5 and ICP47 with the insertion of transgenes into the deleted ICP34.5 region. The cocktail therapy (quintuplet combination of five oHSVs) exhibited the highest tumor inhibition efficacy compared with mono or dual combination treatments in the CT26 or 4T1 tumor models. The study suggested that the combined therapy of multiple oncolytic HSV-2 vectors carrying different immunostimulatory genes would be a potential strategy for cancer therapy.82
IL-18 expressing oHSVs
IL-18, a member of the IL-1 cytokine family, is crucial in stimulating innate and acquired immunity. It enhances IFN-γ production from various immune cells, including NK cells, CD4+ NKT cells, B cells, DCs, and macrophages, in the presence of IL-12. When coupled with IL-2, IL-18 triggers Th2 cytokine production from CD4+ NKT and NK cells.83 However, IL-18 can also contribute to NK cell exhaustion, differentiation toward immunosuppressive subsets, and the expansion of MDSCs in the TME, promoting its pro-tumorigenic effects.84 On the other hand, its ability to synergize with IL-12 in stimulating IFN-γ production suggests that IL-18 may augment the effector function of Th1 and NK cells, contributing to its antitumor activities.83
Ino and colleagues generated four oHSVs: vHsv-null, vHsv-B7.1-Ig, vHsv-IL-12, and vHsv-IL-18, with deletions of ICP34.5 and insertion of immune transgenes into the deleted ICP6 locus.67 In a subcutaneous Neuro2a neuroblastoma model, vHsv-B7.1-Ig and vHsv-IL-18 monotherapies demonstrated modest tumor growth inhibition relative to mock, while vHsv-IL-12 resulted in significantly greater tumor growth inhibition compared with other treatments. In a bilateral subcutaneous Neuro2a model, where the left-sided tumors received treatments and the right-sided tumors were untreated, the combination of vHsv-IL-12 plus vHsv-IL-18 demonstrated significantly superior antitumor efficacy compared with other dual combinations (vHsv-B7.1-Ig plus vHsv-IL-18 or vHsv-B7.1-Ig plus vHsv-IL-12). This suggests that IL-12 expression (vHsv-IL-12) enhances the antitumor effects of vHsv-IL-18,67 illustrating IL-18’s potential to synergize with IL-12 in stimulating antitumor immunity.
CCL2 expressing oHSVs
Human CCL2 is a small cytokine belonging to the CC chemokine family.85 Both activated peripheral blood mononuclear leukocytes and cancer cells can produce CCL2, which recruit monocytes, T cells, and NK cells to the tumor site and induces macrophage-mediated killing of cancer cells.85 By binding to G protein-coupled receptors CCR2, CCR4, and CCR5 on cells,86 CCL2 activates monocytes, macrophages, memory T cells, and NK cells.85 However, recent research indicates that the CCL2-CCR2 axis functions as a tumor-promoting signaling pathway in various cancers.87
Antitumor efficacy of a CCL2-expressing oHSV lacking ICP34.5 (M010) was evaluated in a flank Neuro-2a model.88 Intratumoral M010 showed inconsistent results when compared with the parent R3659.88 However, when M010 was combined with IL-12-expressing M002, the combination therapy exhibited superior tumor growth inhibition compared with monotherapy. The combination significantly increased IFN-γ levels in tumors during the later stages of treatment compared with M002 monotherapy.88 These findings suggest that oHSV co-expression of CCL2 plus IL-12 would be a better strategy than oHSV expression of CCL2 only.
CCL5 expressing oHSVs
The potent chemoattractive chemokine CCL5 is secreted by APCs and activated T cells, and binds to the receptors CCR1, CCR3, and CCR5 expressed on several types of effector and regulatory T cells. CCL5 enhances antitumor immunity by recruiting CD4+, CD8+, and NK cells into the TME.89 Tian and colleagues generated an oHSV-1 expressing a single-chain variable fragment of the epidermal growth factor receptor (EGFR) antibody cetuximab linked to CCL5 by an Fc knob-into-hole strategy that produces heterodimers (OV-Cmab-CCL5).90 The study exhibited that OV-Cmab-CCL5 treatment reduced tumor size and prolonged survival of GBM-bearing mice. Additionally, the continuous release of CCL5 from EGFR+ GBM cells in TME induces migration and activation of NK cells, macrophages, and T cells.90 Liu and colleagues constructed T7011 expressing the extracellular domains of CD19 and B-cell maturation antigen (BCMA), and CCL5, IL-12, and anti-PD-1 antibody. Combining therapy with T7011 and CD19-specific or BCMA-specific CAR T cells demonstrated significant antitumor responses compared with monotherapies in several solid tumor models.91
CXCL4 expressing oHSV (G47Δ-platelet factor 4)
CXCL4, previously known as platelet factor 4 (PF4), is an immunomodulatory cytokine synthesized by various immune cells. It possesses the remarkable ability to target almost all cells within the vasculature.92 CXCL4 stimulates IL-17 production in human CD4+ T cells activated by CD3/CD28 and indirectly induces IL-17 production in co-cultures of CD4+ T cells, monocytes, and myeloid DCs.93 It also primes DCs derived from monocytes, leading to CD4+ T-cell proliferation/activation and IL-17 production. These IL-17-secreting CD4+ T cells can also co-produce IL-22, amplifying the immune response at the inflammation site. Thus, CXCL4 is considered a driver for Th17 function.93
G47Δ-PF4, an oHSV similar in design to G47Δ-mIL-12 but with PF4 inserted into the ICP6 deleted region, was significantly superior in controlling tumor growth and extending median survival in athymic nu/nu mice bearing human U87 glioma or mouse MPNST tumors, compared with G47Δ-empty (no PF4 expression) or PBS.94 This was associated with a significant inhibition of tumor angiogenesis.94 In an immunocompetent orthotopic mouse MPNST model, the expression of PF4 resulted in a significant extension of survival compared with G47Δ-empty or PBS, highlighting a role for immunomodulatory CXCL4 in therapy.54
Dual or multiple cytokine expressing oHSVs
oHSV expressing IL-12 plus GM-CSF (R-123)
HER-2-retargeted oHSV R-113 (or R-LM113),61 lacking cytokine expression, was modified by integrating IL-12 and/or GM-CSF into the US1-US2 and UL26-UL27 intergenic regions, respectively, resulting in R-123 (expressing IL-12 plus GM-CSF), R-115 (expressing only IL-12), and R-121 (expressing only GM-CSF).25 61
Intratumoral delivery of R-123
R-123 demonstrated superior antitumor efficacy against subcutaneous HER2-LLC1 tumors, with tumor regression observed in 40% of treated animals, compared with 0–20% for the other virus constructs.25 Importantly, the combination of R-123 plus anti-PD-1 resulted in 100% long-term survivors, while R-115 plus anti-PD-1, R-121 plus anti-PD-1, R-113 plus anti-PD-1, or anti-PD-1 treatment groups achieved 75%, 50%, 60%, or 0% long-term survivors, respectively.25 The survivors from the R-123+anti-PD-1 combination successfully rejected tumor re-challenge, indicating the induction of a memory response.25 Immune cell depletion studies suggest that CD4+ and CD8+ T cells and NK cells played crucial roles in treatment efficacy, with CD4+ T cells being the major player.25 Furthermore, the depletion of IFN-γ also significantly abrogated the therapeutic benefits of R-123+anti-PD-1. This suggests that IFN-γ, secreted from the combination therapy stimulated CD4+ T cells (and other immune cells), played a critical role in the combination therapy.25
Intravenous delivery of R-123
The antitumor efficacy of intravenous R-123 in combination with anti-PD-1 was tested in an HER2-LLC1 pulmonary metastatic model. Combination therapy significantly inhibited the growth of lung tumor nodules compared with anti-PD-1 alone (R-123 monotherapy was not included in this study).25 Surprisingly, the presence of pre-existing anti-HSV antibodies, a challenge for intravenous oHSV therapy in humans,9 did not negatively impact the efficacy of systemic R-123 therapy. In fact, the antitumor effects of intravenous R-123 (+intraperitoneal anti-PD-1) in HSV-immune mice were similar to those observed in HSV-naïve mice.25
In summary, dual expression of IL-12 and GM-CSF by HER2-targeted R-123 appeared more effective in eradicating tumors than the expression of either cytokine in the HER2-expressing tumor model. However, whether R-123 can achieve a similar level of antitumor efficacy in other cancer models with low or no HER2 expression remains to be determined.
OHSV expressing IL-12 plus IL-15 plus IL-15Rα (VG2025 or VG161)
VG2025 expresses IL-15/IL-15Rα and IL-12 using a transcriptional and translational dual regulation system. A tumor-specific carcinoembryonic antigen promoter was used to control ICP27 protein transcriptionally and multiple microRNAs to control ICP34.5 expression. VF2025 exhibited strong in vivo antitumor efficacy following intratumoral or intravenous administration.95 Increased immune cell infiltration and abscopal antitumor effects suggested a robust antitumor immunity.95 VG161, armed with IL-12, IL-15, and IL-15Rα and a peptide fusion protein disrupting PD-1/PD-L1 interactions, exhibited superior antitumor efficacy in syngeneic CT26 and A20 tumor models compared with parent virus expressing no transgenes.96 Moreover, intratumoral injection of VG161 induced abscopal responses and immune memory. Its superior antitumor efficacy is associated with the infiltration of NK and T cells in tumors and an increased frequency of splenic tumor-specific T cells.96
OHSV expressing GM-CSF plus other immunomodulator(s)
Thomas et al developed several new gibbon ape leukemia viral (GALV) fusogenic protein-expressing oHSVs. They are similar in design to T-VEC (deletion of ICP34.5 and ICP47, with murine (mRP1) or human GM-CSF (RP1)) plus GALV expression (which is functionally inactive in mice). RP1 plus anti-human or anti-murine CTLA-4 is RP2/mRP2,97 and RP3 is plus CD40L and 4-1BBL.98 In vivo, intratumoral mRP1 treatment resulted in an enhanced antitumor effect and immune response in both treated and untreated tumors, and its efficacy was further enhanced by anti-PD-1.97 RP1, RP2, and RP3 (developed by Replimune) are now under clinical investigation in various cancers, either as monotherapy or in combination with other immunotherapeutic agents.99
Conclusions
Cancer immunotherapy employing cytokine-expressing oHSVs provides a feasible and beneficial strategy for harnessing the immune-modulating effects of cytokines to promote antitumor activity while limiting systemic toxicity. To date, preclinical studies using a spectrum of cytokines and chemokines, including GM-CSF, IL-12, FLT3L, TNF-α, IL-2, IL-15, IL-18, CCL2, CCL5, and CXCL4, have been explored to augment and fortify the immunomodulatory attributes of oHSVs (figure 1B, table 1). Of particular significance is that the only FDA-approved oncolytic virus, oHSV T-VEC, expresses the human GM-CSF cytokine.17 Remarkably, approximately 53 clinical studies of T-VEC have either been concluded or are currently underway, examining its efficacy as monotherapy or in combination with other immunotherapeutic or anticancer agents across diverse malignancies. Likely based on the success of T-VEC, GM-CSF is the most commonly encoded cytokine in other OVs in development.
Following GM-CSF, IL-12 has undergone extensive preclinical exploration as an oHSV-expressed cytokine for cancer immunotherapy. Multiple IL-12-expressing oHSVs, including R115, G47Δ-IL-12, M002, NV1042, T-mfIL-12, and vHsv-IL-12, have been generated and rigorously evaluated in various mouse cancer models. In particular, M032, akin in design to M002 but expressing human IL-12, has advanced to clinical trials for malignant glioma, either as monotherapy or in combination with pembrolizumab (anti-PD-1) (NCT05084430, NCT02062827).44 45 Another noteworthy IL-12-expressing oHSV, G47Δ-IL-12, has exhibited promising antitumor effects across diverse malignancies, most notably in combination with dual immune checkpoint inhibitors for GBM. These findings support the clinical translation of G47Δ-IL-12. Notably, the parent virus of G47Δ-IL-12, G47Δ (lacking IL-12 expression), has obtained approval in Japan for GBM treatment and is currently undergoing clinical evaluation in patients with other cancers, including prostate cancer, olfactory neuroblastoma, and malignant pleural mesothelioma.55 100
While GM-CSF and IL-12 expressing oHSVs have been the most widely studied, the overarching aim of cytokine expression is to invigorate and orchestrate antitumor immunity, a goal that could be amplified through the co-expression of multiple cytokines. Notably, oHSV R-123, re-targeted to the HER2 receptor and expressing both IL-12 and GM-CSF, showcases robust antitumor efficacy in preclinical tumor models, whether as monotherapy or in conjunction with an ICI, stimulating antitumor immune responses surpassing those elicited by single cytokine transgenes.25 Future exploration encompassing other dual and multiple cytokine-expressing oHSVs is anticipated. Incorporating multiple cytokines within the oHSV genome holds promise in promoting antitumor immunity while circumventing systemic toxicity as a general approach for optimizing immunotherapy for patients with cancer.
Moreover, intravenous administration of cytokine-armed oHSVs is promising for simplifying and enhancing cancer treatment.25 To achieve this, a range of critical considerations and strategies must be applied to safeguard the oHSV from serum neutralization. These strategies include previously used techniques for effective virus-like structure or oHSV delivery, such as nanoparticle encapsulation, liposomal technology surface modifications, stealth methodologies, selective targeting, and stem cells.101–103 Additional research must be conducted to fine-tune these strategies, ultimately advancing their readiness for widespread clinical utilization.
Ethics statements
Patient consent for publication
Ethics approval
Not applicable.
References
Footnotes
Contributors HW: Wrote/edited the manuscript, prepared tables, and figures. MB: Prepared figures. HLK: Wrote/edited the manuscript and provided clinical perspective. UL: Provided expertise on oHSV delivery. HJN: Wrote/edited the manuscript and provided expertise as a virologist and HSV biologist. SDR: Wrote/edited the manuscript and provided preclinical HSV immunotherapy expertise. DS: Conceptualization, wrote/edited the manuscript, edited tables/figures, funding acquisition. All authors contributed to the article and approved the submitted version.
Funding DS was supported in part by a fund from the DOD (W81XWH-20-1-0702), and SDR was supported in part by a grant from the NIH (R01 CA160762) and the Thomas A Pappas chair in Neurosciences.
Competing interests SDR is a co-inventor on patents relating to oncolytic herpes simplex viruses, owned, and managed by Georgetown University and Massachusetts General Hospital, which have received royalties from Amgen and Acti\Vec Inc., and acted as a consultant and received honoraria from Replimune, Cellinta, and Greenfire Bio, and honoraria and equity from EG 427. HLK is an employee of Ankyra Therapeutics and has received honoraria for participating on advisory boards for Castle Biosciences, Midatech Pharma, Marengo Therapeutics, and Virogin. The remaining authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
Provenance and peer review Commissioned; externally peer reviewed.