Triple‐Mutated HSV‐1 Expressing Soluble B7‐1 Plus CTLA‐4 Blockade Suppresses Lymph Node Metastasis in Tongue Cancer

ABSTRACT

Cervical lymph node metastasis is a major predictor of poor prognosis in patients with oral squamous cell carcinoma (OSCC), underscoring the need for therapies that effectively target metastatic spread. We previously reported that G47Δ, a third‐generation oncolytic herpes simplex virus type 1 (HSV‐1), rapidly traffics from tongue tumors to cervical lymph nodes and suppresses metastasis in murine models. Based on this lymphotropic property and the immunological significance of CD28–B7‐1 interactions for T‐cell activation, we pursue the potential of T‐mB7‐1, a new G47Δ‐derived oncolytic HSV‐1 expressing soluble murine B7‐1. T‐mB7‐1 secreted soluble B7‐1 in vitro without compromising cytopathic activity. In vivo, intratumoral T‐mB7‐1 suppressed tumor growth in the resistant KLN205‐MUC1 subcutaneous model and prolonged survival in SCCVII orthotopic tongue tumors. In the KLN205‐MUC1 orthotopic tongue cancer that exhibits rapid lymphatic metastasis, a single dose of T‐mB7‐1, but not the control virus T‐01, significantly reduced cervical lymph node metastasis and improved survival. Repeated low‐dose T‐mB7‐1 further caused near‐complete suppression of lymph node metastasis and markedly extended survival. Flow cytometry showed both T‐01 and T‐mB7‐1 induced CD69⁺ T‐cell activation in cervical lymph nodes at 18 h, which was reduced with T‐mB7‐1 at 24 h. Importantly, combining T‐mB7‐1 with systemic CTLA‐4 blockade markedly prolonged survival of mice harboring orthotopic KLN205‐MUC1 tumors, with all treated mice surviving beyond 80 days in the otherwise fatal model. These findings demonstrate that soluble B7‐1‐expressing oncolytic HSV‐1 combined with CTLA‐4 blockade durably suppresses lymph node metastasis, providing a promising and minimally invasive strategy for regionally metastatic OSCC.

In murine tongue cancer models, T‐mB7‐1, a triple‐mutated oncolytic herpes simplex virus type 1 (HSV‐1) engineered to express soluble B7‐1, suppresses cervical lymph node metastasis and improves survival, addressing a major prognostic determinant in oral squamous cell carcinoma (OSCC). A combination with systemic CTLA‐4 blockade further unlocks potent and durable antitumor immunity, causing long‐term survival in all treated mice. These findings establish a powerful rationale for combining soluble B7‐1‐expressing oncolytic HSV‐1 with checkpoint inhibition as a minimally invasive strategy against regionally metastatic OSCC.

Abbreviations

DMEM

Dulbecco's Modified Eagle Medium

DPBS

Dulbecco's phosphate‐buffered saline

FCS

fetal calf serum

HSV‐1

herpes simplex virus type 1

ICIs

immune checkpoint inhibitors

IgG

immunoglobulin G

OSCC

oral squamous cell carcinoma

pfu

plaque‐forming units

1. Introduction

Oral squamous cell carcinoma (OSCC) is one of the most common malignant tumors of the oral cavity. Despite advances in surgery, radiotherapy, and chemotherapy, the prognosis for patients with metastatic disease remains poor, particularly in the presence of cervical lymph node metastases, which are recognized as one of the most significant predictors of unfavorable clinical outcomes [1, 2]. In recent years, immune checkpoint inhibitors (ICIs) have emerged as promising therapeutic options for advanced OSCC. Although anti–PD‐1 therapy provides improved survival compared with conventional chemotherapy in inoperable cases [3, 4], its clinical benefit remains constrained by a relatively low response rate and the risk of severe immune‐related adverse effects [3, 4, 5, 6].

There is a pressing need for safer and more effective therapeutic strategies, particularly for the control of metastasis progress. As an emerging modality, oncolytic virus therapy has attracted considerable attention as a minimally invasive treatment option. Talimogene laherparepvec (T‐VEC), a second‐generation oncolytic herpes simplex virus type 1 (HSV‐1) armed with granulocyte–macrophage colony‐stimulating factor, became the first oncolytic virus approved by the U.S. Food and Drug Administration for the treatment of melanoma in 2017 [7, 8]. More recently, a triple‐mutated third‐generation oncolytic HSV‐1, G47Δ, was approved in Japan for the treatment of malignant glioma [9]. With its enhanced tumor‐selective replication capacity and engineered immune‐stimulatory modifications, G47Δ has demonstrated both safety and therapeutic efficacy in various cancers, preclinically and clinically, including glioblastoma, prostate cancer, and mesothelioma [10, 11].

In our previous study, we demonstrated that intratumoral administration of G47Δ into murine tongue tumors led to rapid viral dissemination to the cervical lymph nodes and suppression of metastatic progression [12]. Considering the pivotal role of T‐cell activation in antitumor immunity—particularly the engagement of CD28 by its co‐stimulatory ligand B7‐1 within regional lymph nodes—we hypothesized that local expression of soluble B7‐1 during armed virus replication could potentiate immune activation within the metastatic niche. In this study, we utilize T‐mB7‐1, a G47∆‐derived, triple‐mutated oncolytic HSV‐1 engineered to express soluble murine B7‐1 [13, 14, 15]. While the use of oncolytic HSV‐1 armed with soluble B7‐1 has been demonstrated in murine models [16, 17], the present study extends this strategy using a G47Δ‐based platform and applies it to a clinically relevant OSCC model with cervical lymph node metastasis. Here, we evaluate the therapeutic efficacy and safety of T‐mB7‐1 in models of murine tongue cancer with cervical lymph node metastasis and further assess its potential synergy with a systemic CTLA‐4 blockade.

2. Materials and Methods

2.1. Cells

African green monkey kidney (Vero) cells were obtained from the American Type Culture Collection (ATCC, Rockville, MD, USA). The SAS human oral squamous cell carcinoma (OSCC) cell line was acquired from the Japanese Collection of Research Bioresources (JCRB) Cell Bank (Osaka, Japan). The murine lung squamous cell carcinoma cell line KLN205‐MUC1 was provided by the Riken BioResource Research Center (Ibaraki, Japan). The murine OSCC cell line SCCVII was kindly provided by Prof. Yoshiaki Yura (Graduate School of Dentistry, Osaka University, Japan). Vero cells were cultured in Dulbecco's Modified Eagle Medium (DMEM) supplemented with 10% fetal calf serum (FCS). SAS cells were maintained in DMEM/F12 medium with 10% FCS. SCCVII and KLN205‐MUC1 cells were grown in Minimum Essential Medium (MEM) supplemented with 10% FCS. All cell lines were maintained at 37°C in a humidified atmosphere containing 5% CO₂.

2.2. Construction of T‐mB7‐1

T‐mB7‐1 was constructed in the Todo laboratory at the University of Tokyo using the T‐BAC system, a recombinant HSV‐1 generation system constructed based on the concept of our previous G47∆‐BAC system [16], but with modifications for improvements [15]. Briefly, the gene encoding the extracellular domain of murine B7‐1 fused to the Fc portion of human immunoglobulin G (IgG), derived from the B7.1‐pIg construct [17], was cloned into the shuttle vector plasmid SV‐01, yielding SV‐01‐mB7. This plasmid was then inserted into the T‐BAC plasmid by Cre‐mediated recombination (New England Biolabs, MA, USA). T‐BAC/SV‐01‐mB7 was transfected to Vero cells for Flp‐mediated recombination, and from selected progeny viruses, a single clone was picked after 3 rounds of limiting dilutions.

2.3. In Vitro Cytotoxicity Studies

Tumor cells were seeded in six‐well plates at a density of 2.5 × 10⁵ cells per well and allowed to adhere for 18 h. The cells were then infected with either T‐mB7‐1 or control viruses at multiplicities of infection of 0.1 or 1.0 for 1 h. Following infection, cells were incubated at 34.5°C for 24–96 h. Viable cells were subsequently quantified using either the Countess II FL (Thermo Fisher Scientific, MA, USA) or the Coulter Counter (Beckman Coulter, CA, USA).

2.4. ELISA

Vero, SCCVII, and SAS cells were seeded in six‐well plates at a density of 2.0 × 10⁵ cells per well and incubated for 18 h. The cells were then infected with viruses at a multiplicity of infection of 1.0 for 1 h. After overnight incubation at 34.5°C, the concentration of soluble B7‐1 in the culture supernatant was determined using a mouse CD80 ELISA kit (RayBiotech, GA, USA). Absorbance was measured with an iMark Microplate Reader (Bio‐Rad, CA, USA), and data were analyzed using Microplate Manager 6 software (Bio‐Rad, CA, USA).

2.5. Animal Studies

Female DBA/2, C3H/He and A/J mice (5–6 weeks old) were obtained from Japan SLC (Shizuoka, Japan). All animal experiments were performed in accordance with institutional guidelines for the care and use of laboratory animals and were approved by the Committees for Ethics of Animal Experimentation at both the University of Tokyo and the University of Osaka.

2.6. Subcutaneous Tumor Models

Tumor cells (1 × 10⁶ per mouse) were subcutaneously injected into the left flank of syngeneic mice. When tumors reached approximately 5 mm in diameter, mice received intratumoral injections of T‐01 (control virus), T‐mB7‐1 (5.0 × 10⁵ plaque‐forming units [pfu] for SCCVII tumors; 2.5 × 10⁶ pfu for KLN205‐MUC1 tumors), or mock in 20 μL of 10% glycerol in Dulbecco's phosphate‐buffered saline (DPBS) on Days 0 and 3. Tumor diameters were measured using a caliper, and tumor volume was calculated as length × width × depth. Mice were euthanized when the tumor diameter exceeded 24 mm or if they became moribund, in accordance with ethical guidelines.

2.7. Tongue Cancer Models

Orthotopic tongue tumors were established by injecting SCCVII cells (2.0 × 10⁵ per mouse) into the left side of the tongue of syngeneic C3H/He mice. After 3 days, tumors were inoculated with oncolytic HSV‐1 (2.5 × 10⁵ pfu per mouse) or mock in 20 μL of 10% glycerol in DPBS. The KLN205‐MUC1 orthotopic tongue tumor model was generated in syngeneic DBA/2 mice using the same procedure. Three days after tumor cell injection, mice received a single intratumoral administration of T‐01, T‐mB7‐1 (1.0 × 10⁶ pfu per mouse), or mock in 20 μL of 10% glycerol in DPBS, followed by surgical tumor resection [12]. For the multiple‐administration protocol, mice were treated with T‐01, T‐mB7‐1 (2.5 × 10⁵ pfu per mouse), or mock via intratumoral injection every other day for a total of three doses. Tongue tumors were surgically resected after the third administration. Cervical lymph nodes were measured using a caliper, volume was calculated as longest diameter × (shortest diameter)² ×1/2, and metastatic nodes were defined as those > 10 mm³. Moribund mice were euthanized in accordance with protocols approved by the Institutional Animal Care and Use Committee and were recorded as surviving until the following day for survival analysis.

2.8. Cell Count Test

In the KLN205‐MUC1 orthotopic tongue tumor model, cervical lymph nodes were harvested 14 days after virus administration and placed in 500 μL of culture medium. Tissues were homogenized through a 70‐μm cell strainer and centrifuged. The resulting cell pellets were washed with medium containing 1% penicillin–streptomycin and cultured at 37°C for 18 h. After incubation, the number of viable cells was determined.

2.9. Real‐Time PCR Analysis

Cervical lymph nodes were collected 24 h after intratumoral virus injection. Viral DNA was extracted using the QIAamp DNA Mini Kit (Qiagen, Venlo, The Netherlands) according to the manufacturer's instructions. Quantification was performed using specific primers and a probe targeting the glycoprotein B gene of HSV‐1 (forward primer 5′‐GGCGCGGTCCTCAAAGAT‐3′; reverse primer 5′‐AGAACATCGCCCCGTACAAG‐3′; probe 5′‐CACACCTGCGAAACGGTGACGTCTT‐3′). Viral genome copy numbers were determined by generating a standard curve with a plasmid containing the glycoprotein B sequence as a positive control. Real‐time PCR was performed using the 7500 Fast Real‐Time PCR System (Thermo Fisher Scientific, MA, USA).

2.10. Safety Study Following Intralingual Virus Administration

T‐01, T‐mB7‐1, or strain F (wild‐type HSV‐1) was diluted to 1 × 10⁷ pfu in 20 μL of DPBS containing 10% glycerol and injected into the tongue of 6‐week‐old female A/J mice (strain F, n = 6; other groups, n = 7). Mice were monitored for body weight changes and clinical signs, including appearance, spontaneous activity, and responsiveness to external stimuli, for a period of two months [18]. As in the tongue cancer model, moribund mice were euthanized in accordance with institutional guidelines and were recorded as surviving until the following day for survival analysis.

2.11. Flow Cytometry

Single‐cell suspensions were prepared from cervical lymph nodes harvested from mice bearing orthotopic tongue tumors. Red blood cells were lysed with 1× RBC Lysis Buffer Solution (eBioscience, San Diego, CA, USA), and viable cells were identified using Propidium Iodide Staining Solution (eBioscience). Lymphocytes were first incubated with mouse anti‐CD16/32 antibody (eBioscience) to block Fc receptors, followed by staining with the following fluorophore‐conjugated antibodies: APC‐conjugated anti‐CD3 (clone 17A2, eBioscience), FITC‐conjugated anti‐CD4 (clone RM4‐5, eBioscience), APC‐eFluor 780‐conjugated anti‐CD8a (clone 53‐6.7, eBioscience), PE‐Cyanine7‐conjugated anti‐CD69 (clone H1.2F3, BioLegend, San Diego, CA, USA). Fluorophore‐conjugated isotype control antibodies were used in parallel. Flow cytometric analysis was conducted using a CytoFLEX flow cytometer (V5‐B5‐R3 configuration, Beckman Coulter, Fullerton, CA, USA). Data acquisition and analysis were performed using CytoExpert software (Beckman Coulter) as previously described [12].

2.12. Combination Therapy

Anti‐mouse CTLA‐4 antibody (clone 9H10, Bio X Cell, NH, USA) or control polyclonal Syrian hamster IgG (Bio X Cell) was administered intraperitoneally at 25 μg per mouse in 200 μL of DPBS.

In the KLN205‐MUC1 subcutaneous tumor model, antibody or isotype control was administered intraperitoneally on Days 0, 3, 6, and 9, beginning when tumors reached approximately 5 mm in diameter (designated as Day 0). Concurrently, T‐01, T‐mB7‐1 (2.5 × 10⁶ pfu per mouse), or mock was injected intratumorally in 20 μL of 10% glycerol in DPBS on days 0 and 3 using 30G needles.

In the KLN205‐MUC1 orthotopic tongue tumor model, tumors were allowed to develop for 3 days after transplantation. Anti‐CTLA‐4 antibody or isotype control was administered intraperitoneally on Days 0, 3, 6, and 9. On Day 0, mice also received a single intratumoral injection of T‐01, T‐mB7‐1 (1.0 × 10⁶ pfu per mouse), or mock in 20 μL of 10% glycerol in DPBS. Tumor growth was monitored in the subcutaneous tumor model by measuring tumor diameters. In the orthotopic tongue tumor model, survival time was recorded to evaluate the efficacy of cervical lymph node metastasis suppression.

2.13. Statistical Analysis

One‐way analysis of variance (ANOVA) was used to compare three or more groups with normally distributed data, followed by the Tukey–Kramer post hoc test. For datasets that did not meet normality assumptions, the nonparametric Steel–Dwass test was applied. Survival curves were generated using the Kaplan–Meier method, and statistical significance was evaluated by log‐rank test. For comparisons involving more than two groups, the log‐rank test was followed by Holm–Bonferroni correction, where appropriate. Statistical analyses were performed using Statcel 4 software (OMS, Tokyo, Japan).

For evaluation of cervical lymph node metastasis, lymph nodes with volumes > 10 mm³ were defined as metastatic, and group differences were assessed using Fisher's exact test. p‐values < 0.05 were considered statistically significant. Significance levels are indicated in the figures as follows: *p < 0.05, **p < 0.01, and ***p < 0.001.

3. Results

3.1. T‐mB7‐1 Induces B7‐1 Secretion Without Compromising Viral Cytopathic Activities In Vitro

T‐mB7‐1 is a triple‐mutated oncolytic HSV‐1 expressing soluble murine B7‐1. It has the G47∆ backbone with the gene encoding murine B7‐1 fused to the Fc portion of human IgG inserted into the deleted ICP6 locus (Figure 1). We first examined whether T‐mB7‐1–infected cells secreted soluble B7‐1. ELISA confirmed detectable B7‐1 release from Vero, SAS, and SCCVII cells 2 days after infection with T‐mB7‐1, but not with the control virus T‐01 (Figure 2a). Importantly, T‐mB7‐1 retained cytopathic activities comparable to those of T‐01 in murine SCC cell lines (SCCVII and KLN205‐MUC1) (Figure 2b,c).

FIGURE 1.

Genome structures of T‐01 and T‐mB7‐1. Both T‐01 and T‐mB7‐1 are based on the G47Δ backbone, which contains triple deletions in both copies of the γ34.5 gene, as well as in the ICP6 and α47 genes. T‐mB7‐1 additionally carries a cDNA encoding the extracellular domain (ECD) of murine B7‐1 fused to the Fc region of human IgG1, inserted into the ICP6 locus under the control of the immediate‐early CMV promoter. T‐01 serves as a control virus, containing an empty cassette inserted at the same locus. Abbreviations: TR_L, terminal repeat long; U_L, unique long; IR_L, inverted repeat long; IR_S, inverted repeat short; U_S, unique short; TR_S, terminal repeat short; ICP6, infected cell protein 6; lacZ, Escherichia coli LacZ; pA, poly A; p‐CMV, immediate‐early cytomegalovirus promoter.

FIGURE 2.

Expression of soluble B7‐1 and cytopathic activity of T‐mB7‐1 in human and murine squamous cell carcinoma cell lines. (a) Concentrations of soluble B7‐1 secreted into the culture supernatant following infection of Vero, SAS (human OSCC), and SCCVII (murine OSCC) cells with T‐mB7‐1 or T‐01, as determined by ELISA. (b, c) Cytopathic effects of T‐mB7‐1 and T‐01 in the murine SCC cell lines SCCVII (b) and KLN205‐MUC1 (c). Cell viability is expressed as a percentage relative to mock‐infected controls. Error bars indicate standard deviation (SD). MOI, multiplicity of infection.

3.2. Antitumor Efficacy of T‐mB7‐1 In Vivo

To evaluate the therapeutic efficacy of T‐mB7‐1 in immunocompetent hosts, T‐01 or T‐mB7‐1 was inoculated intratumorally on Days 0 and 3 in syngeneic mice harboring subcutaneous tumors (Figure 3a). Both T‐01 and T‐mB7‐1 (5.0 × 10⁵ pfu per dose) significantly suppressed the growth of subcutaneous SCCVII tumors (p < 0.01 for both versus mock), although no significant difference was observed between T‐01 and T‐mB7‐1 (Figure 3b). On the other hand, T‐mB7‐1, but not T‐01 (2.5 × 10⁶ pfu per dose), significantly suppressed the growth of subcutaneous tumors of KLN205‐MUC1, a squamous cell carcinoma cell line less susceptible to HSV‐1 than SCCVII (p < 0.05 versus mock; Figure 3c).

FIGURE 3.

Antitumor efficacy of T‐01 and T‐mB7‐1 in immunocompetent mouse models. (a) Experimental timeline for subcutaneous tumor models. Syngeneic mice bearing SCCVII or KLN205‐MUC1 subcutaneous tumors were intratumorally injected with T‐01, T‐mB7‐1, or mock on days 0 and 3 (n = 7 per group). (b) Tumor growth curves of subcutaneous SCCVII tumors in C3H/He mice. Both T‐01 and T‐mB7‐1 significantly inhibited the tumor growth in SCCVII‐bearing mice, with no significant difference between the two viruses. (c) Tumor growth curves of subcutaneous KLN205‐MUC1 tumors in DBA/2 mice. T‐mB7‐1, but not T‐01, significantly inhibited the tumor growth. Bars, standard error of the mean (SEM). Tumor volume was calculated as length × width × depth. *p < 0.05; **p < 0.01; NS, not significant. (d) Experimental timeline for the orthotopic tongue tumor model using SCCVII cells. (e) Kaplan–Meier survival analysis of mice with SCCVII orthotopic tongue tumors treated with mock, T‐01, or T‐mB7‐1 on day 3 post‐implantation (n = 11). Both T‐01 and T‐mB7‐1 significantly prolonged the survival compared with controls. **p < 0.01; NS, not significant.

The in vivo efficacy was further evaluated in C3H/He mice harboring orthotopic SCCVII tongue tumors (Figure 3d). Both T‐01 and T‐mB7‐1 significantly prolonged the survival compared with mock (p < 0.01 for both versus mock; Figure 3e), with T‐mB7‐1 causing significantly more surviving animals than T‐01 (p < 0.05, Fisher's exact test), although no significant difference was observed by log rank test (p = 0.168).

3.3. Safety of Intralingual Administration of T‐mB7‐1

To assess the safety, T‐01, T‐mB7‐1, or wild‐type HSV‐1 strain F (1 × 10⁷ pfu) were injected into the tongue of HSV‐1‐susceptible A/J mice. Whereas strain F caused rapid weight loss and mortality, neither T‐mB7‐1 nor T‐01 affected the body weight or the survival of mice receiving intralingual injections (Figure 4a,b). Behavioral observations revealed that, whereas strain F caused rapid deterioration and deaths, T‐01 and T‐mB7‐1 caused no or minimal influences (Figure 4c,d), indicating that intralingual administration of T‐mB7‐1 at a high dose is safe.

FIGURE 4.

Safety evaluation of T‐01 and T‐mB7‐1 following intralingual administration in A/J mice. (a, b) Changes in individual body weight (a) and survival rates (b) of A/J mice after receiving intralingual injections of mock, T‐01, T‐mB7‐1, or wild‐type HSV‐1 strain F (1 × 10⁷ pfu) (strain F, n = 6; other groups, n = 7). (c) Behavioral assessment based on three clinical parameters: Appearance, spontaneous activity, and response to external stimuli. Each parameter was scored on a five‐point scale ranging from 4 (normal) to 0 (death). (d) Time course of behavioral scores of individual A/J mouse after intralingual virus administration (1 × 10⁷ pfu) (strain F, n = 6; other groups, n = 7).

3.4. T‐mB7‐1 Suppresses Lymph Node Metastasis and Prolongs Survival in Orthotopic Tongue Cancer Models

Based on our previous finding of the lymph node tropism of oncolytic HSV‐1 [12], we investigated the effect of T‐mB7‐1 on cervical lymph node metastasis in DBA/2 mice harboring orthotopic KLN205‐MUC1 tongue cancer, a model that exhibits rapid lymphatic metastasis (Figure 5a). Viral genomes from both T‐01 and T‐mB7‐1 were similarly detected in cervical lymph nodes 24 h after injection (Figure 5b). However, only T‐mB7‐1, and not T‐01, significantly prolonged survival compared with mock treatment (p < 0.05; Figure 5c). Cervical lymph node sizes 36 days after tumor implantation were also significantly reduced in T‐mB7‐1 treated mice compared with mock (p = 0.028, Figure 5d,e; p < 0.001, Figure S1). Also, T‐mB7‐1, but not T‐01, significantly reduced the number of viable metastatic cells 14 days after implantation (Figure 5f). The results indicate that T‐mB7‐1 is superior to T‐01 in targeting metastatic cells in cervical lymph nodes after being injected into tongue cancer.

FIGURE 5.

Antitumor efficacy of T‐01 and T‐mB7‐1 against cervical lymph node metastasis in the KLN205‐MUC1 orthotopic tongue cancer model. (a) Experimental timeline of tumor implantation, viral administration, and assessment. (b) Quantification of viral DNA in cervical lymph nodes 24 h after intratumoral injection of T‐01 or T‐mB7‐1 (n = 6), measured by real‐time PCR. Bars, standard error of the mean (SEM). NS, not significant. (c) Kaplan–Meier survival curves of KLN205‐MUC1–implanted DBA/2 mice treated with vehicle control, T‐01, or T‐mB7‐1 (n = 8). *p < 0.05; NS, not significant (log‐rank test with Holm–Bonferroni correction). (d, e) Representative images (d) and volumes (e) of cervical lymph nodes on day 36 after tumor implantation (mock: n = 3 mice, 6 lymph nodes; T‐01 and T‐mB7‐1: n = 4 mice, 8 lymph nodes). Metastatic nodes were defined as those > 10 mm³. Numbers of metastatic versus nonmetastatic nodes: Mock, 6/0; T‐01, 6/2; T‐mB7‐1, 3/5. (f) Viable cell counts in cervical lymph nodes on Day 14 after tumor implantation (n = 7 mice, 14 lymph nodes per group). **p < 0.01; NS, not significant (Steel–Dwass test).

3.5. Multiple Low‐Dose Injections of T‐mB7‐1 Enhance Therapeutic Efficacy

For further optimization for therapeutic efficacy, we investigated the effect of multiple low‐dose administrations of T‐mB7‐1 in the orthotopic KLN205‐MUC1 tongue cancer model (Figure 6a). Mice were treated with intratumoral injections of T‐01, T‐mB7‐1, or mock every other day for a total of three administrations using a quarter dose compared with the single injection regimen (2.5 × 10⁵ pfu per dose). This regimen achieved near‐complete suppression of lymph node metastasis, with T‐mB7‐1 significantly reducing the lymph node size compared with both T‐01 (p = 0.004) and mock treatment (p = 0.0003; Figure 6b,c). The survival was markedly prolonged by T‐mB7‐1 treatments compared with mock treatments (p < 0.01), with six of seven mice surviving beyond 100 days (Figure 6d). These findings indicate that repeated low‐dose T‐mB7‐1 administrations potentiate the therapeutic efficacy.

FIGURE 6.

Therapeutic efficacy of multiple intratumoral injections of oncolytic virus in the orthotopic KLN205‐MUC1 tongue cancer model. (a) Treatment schedule. DBA/2 mice harboring KLN205‐MUC1 tongue tumors received intratumoral injections of T‐01, T‐mB7‐1 (2.5 × 10⁵ pfu/20 μL), or mock every other day for a total of three administrations. Tongue tumors were surgically resected after the third administration. Cervical lymph node volumes and survival were subsequently evaluated. (b, c) Assessment of cervical lymph nodes 50 days post‐tumor implantation. Representative images of excised lymph nodes (b). Measurements of lymph node volumes (c; mock: n = 6 mice, 12 lymph nodes; T‐01 and T‐mB7‐1: n = 7 mice, 14 lymph nodes). Metastatic nodes were defined as those > 10 mm³. Numbers of metastatic versus nonmetastatic nodes: Mock, 8/4; T‐01, 6/6; T‐mB7‐1, 0/14. Bars represent group averages. **p < 0.01; ***p < 0.001 (Fisher's exact test). (d) Kaplan–Meier survival curves of mice treated with three‐time, low‐dose, intratumoral injections of mock, T‐01, or T‐mB7‐1 (n = 11 per group). **p < 0.01; NS, not significant.

3.6. Soluble B7‐1 Modulates T‐Cell Activation in Lymph Nodes

Flow cytometric analysis of lymphocytes isolated from cervical lymph nodes at 18 and 24 h after a single intratumoral injection of mock, T‐01 or T‐mB7‐1 in the KLN205‐MUC1 orthotopic tongue cancer model revealed early T‐cell activation, as indicated by increased CD69 expression on CD3⁺ and CD8⁺ cells (Figure 7a,b). Notably, the percentage of CD69⁺ T cells was significantly lower in the T‐mB7‐1 group at 24 h than T‐01, suggesting that local soluble B7‐1 expression results in shortening the CD69+ T cell presence. This may reflect the higher expression levels of soluble B7‐1 at 24 h than at 18 h after infection (Figure S2). Despite this reduction in the early activation marker, IFN‐γ levels were significantly higher in the T‐mB7‐1 group at 24 h than T‐01 (Figure S3). Soluble B7‐1 levels in the lymph nodes at 24 h tended to be higher in the T‐mB7‐1 group than in the mock and T‐01 groups, although the differences were not statistically significant (Figure S4).

FIGURE 7.

Flow cytometry analysis of T‐cell activation in cervical lymph nodes after viral inoculation. (a, b) Proportions of CD69⁺ T cells in cervical lymph nodes at 18 h (a, n = 5) and 24 h (b, n = 7) after intratumoral injection of mock, T‐01 or T‐mB7‐1 in the KLN205‐MUC1 orthotopic tongue cancer model. Flow cytometry was used to quantify activation markers in lymphocyte populations. Bars represent group averages. **p < 0.01; ***p < 0.001; NS, not significant.

3.7. Combination With Systemic CTLA‐4 Blockade Augments Therapeutic Efficacy

We next investigated whether systemic CTLA‐4 blockade could augment the efficacy of intratumorally administered T‐mB7‐1 in subcutaneous and orthotopic KLN205‐MUC1 models (Figure 8a,c). In the subcutaneous tumor model, the combination of T‐mB7‐1 and anti‐CTLA‐4 antibody caused significant suppression of tumor growth compared with anti‐CTLA‐4 alone or T‐01 alone (p < 0.01 for both), but showed no significant difference between T‐mB7‐1 alone or T‐01 + anti‐CTLA‐4 antibody (Figure 8b). In the orthotopic KLN205‐MUC1 model, the combination of T‐mB7‐1 and anti–CTLA‐4 antibody significantly augmented the efficacy compared with either therapy alone (p < 0.05 and p < 0.001, respectively; Figure 8d). Notably, all mice receiving the combination of T‐mB7‐1 and anti‐CTLA‐4 antibody survived beyond 80 days, whereas control mice succumbed within 60 days.

FIGURE 8.

Antitumor efficacy of T‐mB7‐1 combined with systemic CTLA‐4 blockade in KLN205‐MUC1 tumor models. (a) Treatment schedule for the subcutaneous KLN205‐MUC1 tumor model. Mice received intratumoral injections of mock, T‐01, or T‐mB7‐1 (2.5 × 10⁶ pfu) on days 0 and 3, and intraperitoneal injections of anti–CTLA‐4 antibody or isotype control (25 μg) on days 0, 3, 6, and 9 (n = 8 per group). (b) Tumor growth curves of subcutaneous KLN205‐MUC1 tumors. The combination of intratumoral T‐mB7‐1 and systemic anti–CTLA‐4 antibody significantly suppressed the tumor growth compared with anti‐CTLA‐4 alone, T‐01 alone, or mock. Bars represent standard error of the mean (SEM). *p < 0.05; **p < 0.01. (c) Treatment schedule for the orthotopic KLN205‐MUC1 tongue tumors. Mice received intratumoral injections of mock, T‐01, or T‐mB7‐1 (1.0 × 10⁶ pfu) on days 0, and intraperitoneal injections of anti–CTLA‐4 antibody or isotype control (25 μg) on days 0, 3, 6, and 9 (n = 9 per group). Tongues were resected immediately after virus injections. (d) Kaplan–Meier survival analysis. The combination of T‐mB7‐1 and anti–CTLA‐4 antibody significantly prolonged the survival compared with either therapy alone. *p < 0.05; **p < 0.01; ***p < 0.001.

4. Discussion

Oncolytic virus therapy represents a promising and minimally invasive strategy for cancer treatment, with nearly 100 clinical trials conducted between 2000 and 2020 [19]. The approval of the third‐generation HSV‐1 G47Δ for malignant glioma in Japan marked a significant milestone in this field [20, 21]. A major component of the therapeutic efficacy of oncolytic viruses lies in their capacity to stimulate host antitumor immunity [16]. In line with this, arming G47Δ with immunomodulatory genes such as IL‐12 or TSP‐1 has demonstrated considerable promise in preclinical models [22, 23].

B7‐1 is a critical co‐stimulatory ligand that activates T cells through CD28 engagement, particularly in lymph nodes [24]. Previous studies have shown that B7‐1 expression by recombinant viruses, such as adenovirus and vaccinia virus, induces T cell‐mediated antitumor responses [25, 26]. In situ expression of soluble B7‐1 in the context of or by oncolytic HSV‐1 has been shown to induce potent antitumor immunity in poorly‐immunogenic tumor models [16, 17]. Building on these prior works, the ideas of G47∆‐based platform and the soluble B7‐1 expression are integrated, which is investigated in this study for the first time in head and neck cancer, a tumor type where lymph‐node control is clinically decisive. Whereas the in vivo therapeutic effects of T‐mB7‐1 on injected tumors were comparable to those of the control virus T‐01 in subcutaneous and orthotopic tongue cancer models, a single intratumoral administration of T‐mB7‐1, but not T‐01, significantly suppressed the lymph node metastasis of orthotopic KLN205‐MUC1 tongue cancer that exhibits rapid lymphatic metastasis. Multiple intratumoral injections of low‐dose T‐mB7‐1 further enhanced the suppression of lymph node metastasis, resulting in significantly smaller cervical lymph nodes compared with T‐01. As translational significance, these findings suggest that repeated dosing may be desirable to fully exploit the immunological benefits of soluble B7‐1 expression. Importantly, intralingual administrations of T‐mB7‐1 are found safe in HSV‐1 susceptible A/J mice, which establishes an initial safety profile that may facilitate future clinical translation.

An intriguing finding was that both T‐mB7‐1 and T‐01 treatments caused activation of CD69⁺ T cells within lymph nodes of the KLN205‐MUC1 orthotopic tongue cancer model at 18 h, but the percentages of CD69⁺ T cells decreased at 24 h only in T‐mB7‐1 treated mice, possibly reflecting the higher expression levels of soluble B7‐1 at 24 h. T‐mB7‐1 significantly increased IFN‐γ expression in metastatic lymph nodes at 24 h, indicating enhanced effector cytokine production in the local immune microenvironment. Besides T cell stimulation through CD28 binding, soluble B7‐1 can also bind to CTLA‐4 and generate inhibitory signals to T cells, although less dominant than CD28 signals [27, 28]. Therefore, combining T‐mB7‐1 with systemic CTLA‐4 blockade is a reasonable therapeutic strategy, unlocking its full immunostimulatory potential (Figure 9). Consistent with prior studies demonstrating synergism between oncolytic viruses and ICIs [29, 30, 31, 32], our results showed that T‐mB7‐1 in combination with anti‐CTLA‐4 therapy outperformed either monotherapy in prolonging the survival of mice harboring orthotopic KLN205‐MUC1 tongue cancers, causing all mice in the group to survive beyond 80 days. Mechanistically, CTLA‐4 blockade may attenuate regulatory T‐cell (Treg) activity, consistent with reports that CTLA‐4‐deficient Tregs fail to suppress antitumor immunity in vivo [33]. The relative contributions of CD28‐mediated co‐stimulation, CTLA‐4 antagonism, and other potential pathways are yet to be determined. Currently, ICIs such as anti‐PD‐1 antibodies are approved for OSCC but their clinical benefits remain limited by low response rates and the risk of severe immune‐related adverse effects. Our study provides preclinical evidence that the combination of a functionally enhanced oncolytic HSV‐1, such as T‐mB7‐1, with CTLA‐4 blockade represents a potent and safe therapeutic strategy for controlling lymph node metastases in OSCC.

FIGURE 9.

Illustration of potential mechanism of antitumor effects by T‐mB7‐1 in combination with CTLA‐4 blockade in primary and metastatic tongue cancer lesions.

While CTLA‐4 blockade was selected in the present study based on the known interactions of soluble B7‐1 with CD28 and CTLA‐4 [27, 34], soluble B7‐1 may also modulate the PD‐L1 pathway. As anti‐PD‐1 antibodies are currently approved for the treatment of OSCC, combination strategies incorporating PD‐1 pathway inhibition remain an important subject for future investigation. In addition to CD8⁺ T cells, other immune cell populations, including CD4⁺ T cells, natural killer cells, macrophages, Tregs, and conventional type 1 dendritic cells (cDC1s), may also contribute to antitumor immunity in this setting [35, 36, 37]. In particular, cDC1s can cross‐present tumor antigens and prime CD8⁺ T cells in tumor‐draining lymph nodes. Oncolytic virus therapy has been reported to induce systemic immune activation partly via tumor cell death and the release of danger‐associated molecular patterns (DAMPs) [38, 39]. These processes may promote immunogenic cell death and enhance antigen presentation, thereby contributing to systemic antitumor immune responses. Further studies are warranted to elucidate the precise mechanisms underlying the observed effects and to validate these findings in additional tumor models. Nevertheless, our results provide a strong rationale for the development of new lines of oncolytic HSV‐1 engineered to express immunomodulatory molecules, and further in combination with immune checkpoint blockade, particularly as a minimally invasive therapeutic strategy for regionally metastatic OSCC.

Author Contributions

Akinari Sugauchi: data curation, formal analysis, investigation, methodology, writing – original draft. Toshihiro Uchihashi: conceptualization, formal analysis, funding acquisition, methodology, supervision, writing – review and editing. Hirotaka Ito: data curation, formal analysis, methodology, writing – review and editing. Miwako Iwai: data curation, formal analysis, investigation, methodology, writing – review and editing. Yuzuri Tsurumaki Sato: investigation. Kyoko Kurioka: investigation. Sena Zeynep Usta: investigation. Minoru Tanaka: supervision. Mikihiko Kogo: conceptualization. Susumu Tanaka: conceptualization. Hiroshi Fukuhara: supervision. Tomoki Todo: conceptualization, formal analysis, funding acquisition, methodology, supervision, writing – review and editing.

Funding

This study was supported in part by the Japan Society for the Promotion of Science (JSPS) Grant‐in‐Aid for Scientific Research under grant number JP21K10137 to T.U. and the Grant‐in‐Aid for Early‐Career Scientists under grant numbers JP19K19194 to T.U. and JP20K18721 to A.S., and the Japan Agency for Medical Research and Development (AMED) under grant numbers JP18ck0106416, JP20lm0203140, JP24bm1223021 and JP24am0521005 to T.T.

Ethics Statement

All were approved by the Committee for Ethics of Animal Experimentation and conducted in accordance with the Guideline for Animal Experiments at the University of Tokyo and the University of Osaka.

Consent

The authors have nothing to report.

Conflicts of Interest

Tomoki Todo and Toshihiro Uchihashi are editorial board members of Cancer Science. Minoru Tanaka is a member of a division endowed by Denka Company. The other authors declare no conflicts of interest.

Supporting information

Figure S1: Antitumor efficacy of T‐01 and T‐mB7‐1 against cervical lymph node metastasis in the KLN205‐MUC1 orthotopic tongue cancer model.

Figure S2: Expression of soluble B7‐1 in KLN205‐MUC1 cells infected with T‐mB7‐1.

Figure S3: IFN‐γ expression in cervical lymph nodes.

Figure S4: Soluble B7‐1 levels in the cervical lymph nodes 24 h after intratumoral virus administration.

Data Availability Statement

The data that support the findings of this study are available from the corresponding author on reasonable request.