Article Text

Original research
Efficacy of LCMV-based cancer immunotherapies is unleashed by intratumoral injections of polyI:C
  1. Celia Gomar1,
  2. Claudia Augusta Di Trani1,
  3. Angela Bella1,
  4. Leire Arrizabalaga1,
  5. Jose Gonzalez-Gomariz1,
  6. Myriam Fernandez-Sendin1,
  7. Maite Alvarez1,
  8. Joan Salvador Russo-Cabrera1,
  9. Nuria Ardaiz1,
  10. Fernando Aranda1,
  11. Timo Schippers2,
  12. Marisol Quintero3,
  13. Ignacio Melero1,4,
  14. Klaus K Orlinger2,
  15. Henning Lauterbach2 and
  16. Pedro Berraondo1
  1. 1Program of Immunology and Immunotherapy, Cima Universidad de Navarra, Pamplona, Spain
  2. 2Hookipa Pharma Inc, New York, New York, USA
  3. 3Highlight Therapeutics, Valencia, Spain
  4. 4Departments of Immunology and Oncology, Clínica Universidad de Navarra, Pamplona, Spain
  1. Correspondence to Dr Pedro Berraondo; pberraondol{at}unav.es

Abstract

Background Lymphocytic choriomeningitis virus (LCMV) belongs to the Arenavirus family known for inducing strong cytotoxic T-cell responses in both mice and humans. LCMV has been engineered for the development of cancer immunotherapies, currently undergoing evaluation in phase I/II clinical trials. Initial findings have demonstrated safety and an exceptional ability to activate and expand tumor-specific T lymphocytes. Combination strategies to maximize the antitumor effectiveness of LCMV-based immunotherapies are being explored.

Methods We assessed the antitumor therapeutic effects of intratumoral administration of polyinosinic:polycytidylic acid (poly(I:C)) and systemic vaccination using an LCMV-vector expressing non-oncogenic versions of the E6 and E7 antigens of human papillomavirus 16 (artLCMV-E7E6) in a bilateral model engrafting TC-1/A9 cells. This cell line, derived from the parental TC-1, exhibits low MHC class I expression and is highly immune-resistant. The mechanisms underlying the combination’s efficacy were investigated through bulk RNA-seq, flow cytometry analyses of the tumor microenvironment, selective depletions using antibodies and clodronate liposomes, Batf3 deficient mice, and in vivo bioluminescence experiments. Finally, we assessed the antitumor effectiveness of the combination of artLCMV-E7E6 with BO-112, a GMP-grade poly(I:C) formulated in polyethyleneimine, currently under evaluation in clinical trials.

Results Intratumoral injection of poly(I:C) enhanced the antitumor efficacy of artLCMV-E7E6 in both injected and non-injected tumor lesions. The combined treatment resulted in a significant delay in tumor growth and often complete eradication of several tumor lesions, leading to significantly improved survival compared with monotherapies. While intratumoral administration of poly(I:C) did not impact LCMV vector biodistribution or transgene expression, it significantly modified leucocyte infiltrates within the tumor microenvironment and amplified systemic efficacy through proinflammatory cytokines/chemokines such as CCL3, CCL5, CXCL10, TNF, IFNα, and IL12p70. Upregulation of MHC on tumor cells and a reconfiguration of the gene expression programs related to tumor vasculature, leucocyte migration, and the activation profile of tumor-infiltrating CD8+ T lymphocytes were observed. Indeed, the antitumor effect relied on the functions of CD8+ T lymphocytes and macrophages. The synergistic efficacy of the combination was further confirmed when BO-112 was included.

Conclusion Intratumoral injection of poly(I:C) sensitizes MHClow tumors to the antitumor effects of artLCMV-E7E6, resulting in a potent therapeutic synergy.

  • Adjuvants, Immunologic
  • Drug Evaluation, Preclinical
  • Immunogenicity, Vaccine
  • Immunotherapy
  • Vaccination

Data availability statement

Data are available in a public, open access repository. Data are available on reasonable request. RNAseq datasets are available on the Gene Expression Omnibus (GEO) website (accession number: GSE245436 and GSE256478).

http://creativecommons.org/licenses/by-nc/4.0/

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

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

WHAT IS ALREADY KNOWN ON THIS TOPIC

  • Lymphocytic choriomeningitis virus (LCMV) has been engineered for the development of cancer immunotherapies, demonstrating both safety and potent activation of tumor-specific T lymphocytes in clinical trials. Combination strategies aimed at overcoming the immunosuppressive tumor microenvironment could enhance antitumor efficacy.

WHAT THIS STUDY ADDS

  • We demonstrate that intratumoral administration of polyinosinic:polycytidylic acid or a nanoplexed form (BO-112) enhances the antitumor efficacy of artLCMV-E7E6 by reconfiguring gene expression programs related to tumor vasculature and immune cell activation, as well as by upregulating MHC on tumor cells.

HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY

  • This study offers preclinical evidence supporting the need for clinical trials to assess the combination of BO-112 and artLCMV-E7E6, both compounds currently undergoing clinical evaluation.

Background

Lymphocytic choriomeningitis virus (LCMV) is an RNA virus that belongs to the Arenavirus family.1 The primary host of LCMV is the common house mouse, and infection in humans is generally asymptomatic or manifests as self-limiting influenza-like symptoms, which in 1%–2% of the cases may progress to a second phase of disease presenting as aseptic meningitis.2 The seroprevalence in humans is very low in most countries and geographic regions, including Europe, and does not normally exceed 5% of the population.3 4 The immunobiology of LCMV infection in mice has been extensively studied as a model of persistent viral infection-induced immunopathology, and to study cytotoxic T-cell immune responses.5–9 This virus induces a vigorous cellular immune response and has been exploited to develop vaccines targeting chronic infectious diseases and cancer. LCMV has been engineered to generate replication-competent but stably attenuated viral vectors that can deliver tumor-associated antigens to antigen-presenting cells.10 LCMV-based antitumor vaccines have demonstrated a remarkable ability to expand tumor-specific T lymphocytes, leading to tumor growth control in preclinical models.10–12 The mechanism for the extraordinary levels of expansion of tumor-specific T lymphocytes has been proposed to depend on the production of interleukin 33 by lymphoid tissue stroma cells on infection by the stably attenuated LCMV vectors.10 Another interesting feature of LCMV as a vaccine vector is that it is a weak inducer of virus-neutralizing antibodies due to the glycosylation of LCMV surface protein. Therefore, LCMV-based vectors can be readministered effectively.10 13 Based on these unique properties as vaccine vectors, several clinical trials are underway to evaluate the safety and efficacy of LCMV-based antitumor vaccines (NCT05553639, NCT04630353, and NCT04180215). Two of these clinical trials use LCMV vectors encoding human papillomavirus (HPV) 16 E6 and E7 antigens and focus on HPV 16-positive tumors.

HPVs are a group of DNA viruses that can cause warts on different parts of the body. Benign lesions can transform into malignant lesions and cause squamous carcinoma of the cervix and other cancers, such as head and neck, vulvar, and anogenital cancers.14 The presence of the E6 and E7 oncogenic proteins is required for the transformation process, and they are expressed in malignant lesions. Therefore, these foreign oncoviral antigens are considered ideal candidates for developing therapeutic antitumor vaccines. A wide variety of vaccine formats has been evaluated in preclinical models and clinical trials, including short and long peptides, fusion proteins, virus-like particles, viral vectors, peptide-pulsed dendritic cells, and mRNA.15–20 However, the immune escape mechanism deployed in the tumor microenvironment limits the antitumor efficacy of this therapeutic strategy against established tumors. Combination treatments with chemotherapy drugs able to remodel the tumor microenvironment can synergize with antitumor vaccines.16 Among different available cancer immunotherapy tools, intratumoral administration of pathogen-associated molecular patterns or damage-associated molecular patterns stands out in this regard.21–24 These microbial molecular patterns are recognized by innate receptors, such as toll-like receptors (TLRs), activating cancer immunity. Polyinosinic:polycytidylic acid (poly(I:C)) is a synthetic molecule that mimics double-stranded RNA. The main receptor for this viral-denoting substance is TLR-3, which is expressed on the endosomal membrane of B lymphocytes, macrophages, and dendritic cells. In addition, cytoplasmic mRNA receptors such as melanoma differentiation-associated protein 5 and protein kinase R can also interact with poly(I:C). Poly(I:C) derivatives with improved pharmacokinetic/pharmacodynamic properties are being studied in clinical trials for their potential therapeutic effects. Poly-ICLC is a stabilized formulation of poly(I:C) with carboxymethylcellulose and polylysine.25 BO-112 is a nanoplexed form of this moiety in polyethylenimine,26 which is being tested in combination with anti-PD-1 monoclonal antibodies to intratumorally treat patients with unresectable malignant melanoma.23 Evidence for clinical activity against checkpoint-refractory melanomas has been observed.27

To identify combination strategies that may enhance the antitumor efficacy of the therapeutic vaccination with LCMV vectors, we used a tumor model that constitutively expresses the E6 and E7 antigens and is resistant to vaccine alone due to low major histocompatibility complex (MHC) class I expression. In this experimental setting, intratumoral administration of poly(I:C) exhibits a synergistic effect with the vaccine, not only against the injected lesion but also against distant concomitant uninjected lesions.

Methods

Mice

C57BL/6 mice were purchased from Harlan Laboratories (Barcelona, Spain). C57BL/6 Batf3tm1Kmm/J (basic leucine zipper ATF-like transcription factor 3 (BATF3) knockout28 or wild-type counterparts were kindly provided by Dr. Kenneth M. Murphy (Washington University, St. Louis, Missouri, USA) and bred at the Cima Universidad de Navarra animal facility. Female mice aged 8–12 weeks were used and housed under specific pathogen-free conditions.

Cell lines and cell cultures

All experiments were carried out using the murine TC-1/A9 cell line (RRID: CVCL_ZW99). This cell line, derived from primary mouse lung epithelial cells, constitutively expresses the E6 and E7 proteins from HPV 16 and is characterized by low MHC class I expression.29 Cells were cultured in Roswell Park Memorial Institute 1640 media supplemented with GlutaMAX (Gibco), 10% heat-inactivated fetal bovine serum, 50 µM 2-mercaptoethanol, 0.4 mg/mL geneticin, 100 U/mL penicillin, and 100 µg/mL streptomycin at 37°C with 5% CO2. Cells were collected for tumor studies when they reached exponential growth and were tested for mycoplasma contamination (MycoAlert Mycoplasma Detection Kit, Lonza).

Mouse tumor models

1.5×105 TC-1/A9 tumor cells were subcutaneously injected into the right flank of C57BL/6 mice. To assess the effect on distant lesions, the left flank received a subcutaneous injection of 7.5×104 tumor cells. One week after tumor inoculation, mice were randomized into different treatment groups. Right-side tumors were injected intratumorally at days 7, 10, and 13 with 50 µg of poly(I:C) (Thermo Fisher, Massachusetts, USA) or BO-112 in 50 µL of phosphate-buffered saline (PBS) or the same volume of PBS. 1×105 replication competent virus focus forming units (RCV FFU) artLCMV-E7E6 was intravenously injected. The vector design, propagation, and characterization have been previously described.11 Treated and untreated tumors were measured twice a week with calipers, and the volume was calculated using the formula length×width2/2. Additionally, mice were monitored for survival and euthanized when any tumor size reached 15 mm or mice displayed signs of distress. In some experiments, C57BL/6 mice deficient for Batf3, or their wild-type counterparts were used.

For selective depletion studies of immune cell subsets, mice were intraperitoneally treated with 200 µg anti-NK1.1 (clone PK136, BioXcell), anti-CD8β (clone Lyt3.2, BioXcell) anti-CD4 (clone GK1.5, BioXcell), or 200 µL clodronate liposome (Liposoma, Amsterdam, The Netherlands) on days 6, 9, and 13 after tumor inoculation. Control mice received intraperitoneal injections of rat IgG (BioXcell).

RNA-seq analysis

Quality control of all RNA samples was performed with the FastQC tool (http://www.bioinformatics.bbsrc.ac.uk/projects/fastqc). Before alignment, low-quality reads and adapters were removed with Trimmomatic. A raw count matrix was obtained using the STAR30 aligner with an mm39 assembly and annotated with Gencode version M27. After that, quantification of reads was performed using featureCounts.31 The analysis of differentially expressed genes was carried out in R/Bioconductor32 33 following the workflow provided by edgeR.34 First, genes with less than five counts in at least 50% of samples for each group (non-expressed genes) were removed from the analysis prior to normalization. The datasets were normalized using trimmed mean of M-values, then the log2 counts per million reads values were calculated, and the expression matrix was used for the statistical analysis. We selected the set of differentially expressed genes for each comparison (false discovery rate (FDR)<0.05 and logFC <−1|>1). Gene Ontology enrichment analysis was performed with the differentially expressed genes between conditions using the clusterProfiler package35 with the biological process ontology as a reference. Gene set enrichment analysis was performed with all the expressed genes ordered by their LogFC using the fgsea package (V.1.20) against M5:GOBP collection of MSigDB as reference. Data are available on the Gene Expression Omnibus website (accession number: GSE24543636 and GSE25647837).

Serum cytokine analysis

The level of cytokines/chemokines in serum collected 24 hours after treatment was measured using a multiplex assay (Luminex MAGPIX Instrument System, Thermo Fisher Scientific, Waltham, Massachusetts, USA) with a custom designed Cytokine 13-Plex Mouse ProcartaPlex Panel (Thermo Fisher Scientific, Waltham, Massachusetts, USA) following the manufacturer’s instructions.

Bioluminescence analysis

For the bioluminescence assay, we employed a trisegmented, artLCMV vector that encodes the NanoLuc (Promega) open reading frame on both S genome segments downstream of the 5’UTR promotor. The reporter vector was generated and characterized as described previously.11 Seven days after TC1/A9 tumor inoculation, 1×105 RCV FFU artLCMV-NanoLuc was intravenously injected, and some mice received 50 µg poly(I:C) intratumorally. 24 hours after injection, in vivo and ex vivo bioluminescence, was detected to visualize virus biodistribution. To this end, 0.15 µmoL in 50 µL Nano-Glo In Vivo Substrate (Promega, Madison, Wisconsin, USA) was administered intraperitoneally. After 5 min, bioluminescence was detected using PhotonIMAGER TM (Biospace Lab, Paris, France). Then, different organs were extracted and imaged. Data were analyzed using M3 Vision software.

Flow cytometry analysis

For analysis of the immune cells in tumors, mice were sacrificed 14 days after tumor challenge. Tumors were mechanically disrupted, and single-cell suspensions were stained with fluorochrome-conjugated monoclonal antibodies. Stained cells were analyzed with Cytoflex XS (Beckmann Coulter, Indianapolis, Indiana, USA). Fluorescence minus one or biological comparison controls were used for cell analysis. Flow cytometry antibodies, tetramers, cell death staining and isotype control are listed in online supplemental table 1. FlowJo software (TreeStar) was used for data analysis. FCS files were uploaded in FlowJo Software, and the compensation matrix was adjusted. We selected the CD45+ cells for downstream analyses in the R/Bioconductor environment.32 33 Before starting the analysis, we randomly selected 50,000 cells from each FCS file to optimize time and computational resources. Quality control of cells and doublet removal was performed with the FlowCT package.38 Data integration was achieved with the fastMNN function from the batchelor package. After, unsupervised clustering was accomplished using the buildSNNGraph function from the scran package.39 The clusters generated were manually annotated from the mean expression of the different population markers. Some clusters were removed due to their ambiguous identification. After that, we generated the UMAP visualization using a subsampling of 1000 cells by each acquisition. Differential abundance analysis and differential expression marker analysis between conditions were performed using the edgeR package34 following the Orchestrating single-cell analysis (OSCA) pipeline.39

Supplemental material

For intracellular staining, mice were sacrificed, and single cell suspension was prepared from spleen. Splenocytes were restimulated in vitro using overlapping peptide sets for HPV16 E6 and E7 (from Hookipa). As a control, splenocytes were incubated with medium only. Cells were incubated first with the respective peptide set for 1 hour, and then Brefeldin A (Sigma Aldrich) was added for an additional 4 hours of incubation. Cells were then stained with antibodies against CD3, CD8, CD4, IFNγ, and CD107a (online supplemental table 1).

Statistical analysis

GraphPad Prism V.10.0.3 software (GraphPad Software, San Diego, California, USA) was used for statistical analysis. Data were analyzed by one-way ANOVA followed by Dunnett’s multiple comparisons test. Longitudinal data were fitted to a third-order polynomial equation and compared with an extra sum-of-squares F test. Survival analysis was performed using the log-rank test. Values of p<0.05 were considered to be statistically significant.

Results

Intratumoral injection of poly(I:C) improves the antitumor efficacy of artLCMV-E7E6, affecting both injected and non-injected tumor lesions

Previously, it has been demonstrated that LCMV-based vectors expressing HPV 16 E7 and E6 antigens can robustly expand tumor-specific lymphocytes with exceptional intensity. These lymphocytes exhibit the capacity to induce IFNγ, TNF, and IL2, resulting in a potent antitumor effect on transplanted TC1 tumors,11 which constitutively express the E6 and E7 proteins. To further characterize this antitumor vaccine using a more clinically relevant tumor model of HPV16+ cancers, we used TC1/A9 cells. This cell line, derived from tumors produced by subcutaneous injection of the TC1 parent cell line, is characterized by low levels of MHC class I on the cell surface. We evaluated the antitumor efficacy of the intravenous administration of artLCMV-E7E6 in mice bearing bilateral TC1 tumors (figure 1A). Seven days after tumor implantation, intravenous administration of the LCMV vector did not significantly impact mouse survival (figure 1B–D). Intratumoral administration of three doses of poly(I:C) on days 7, 10, and 13 also did not modify tumor growth. Thus, individual monotherapies failed to significantly improve survival. To assess whether the intratumoral administration of poly(I:C) could sensitize tumors to tumor-specific T lymphocytes induced by the artLCMV vaccine, we inoculated a single dose of the artLCMV vector intravenously on day 7, while poly(I:C) was injected into the tumor lesion on the right flank on days 7, 10, and 13. The non-injected tumor lesion on the left side was used to evaluate the antitumor responses in lesions that did not receive poly(I:C). The combined treatment resulted in a synergistic antitumor efficacy, significantly delaying tumor growth, and achieving complete eradication of several tumor lesions. The effect was not only limited to the injected lesion, as in four out of six mice, the treatment also eradicated the non-injected tumor lesion (figure 1B,C). Consequently, in contrast to monotherapies, the combination therapy significantly increased tumor control and survival (figure 1D).

Figure 1

Intratumoral poly(I:C) enhances the antitumor activity of an LCMV-based vaccine in both injected tumor lesions and distant non-injected engrafted tumors. Mice were subcutaneously injected with 1.5×105 TC-1/A9 cells in the right flank and 7.5×104 TC-1/A9 cells in the left flank. Seven days later, mice were intravenously treated with 1×105 RCV FFU artLCMV-E7E6 and received intratumoral injections of 50 µg of poly(I:C) in the right-sided tumor lesion on days 7, 10, and 13. (A) Schematic representation of the experimental setup. (B) Tumor growth (mm3) is depicted for individual mice in both injected (right flank) and non-injected (left flank) tumors. The numbers in each graph represent the fraction of mice achieving complete tumor regression. (C) The average of in vivo tumor growth is presented for injected (right flank) and non-injected (left flank) tumors. (D) Survival percentage over time is displayed for the experiments in (B). Data are representative of two independent experiments with six mice per group (mean±SEM). Extra sum-of-squares F test (C) or log-rank (D) tests were used to assess significance. Significant differences are indicated for comparisons of each group with the artLCMV-E7E6+poly(I:C) group (**p<0.01). LCMV, lymphocytic choriomeningitis virus; RCV FFU, replication competent virus focus forming units; poly(I:C), polyinosinic:polycytidylic acid.

Systemic artLCMV-E7E6 and intratumoral poly(I:C) jointly modulate the transcriptome in the tumor microenvironment and enhance MHC class I expression on tumor cells

Next, we assessed the changes in gene expression within the tumor microenvironment following intravenous administration of the LCMV vector combined with intratumoral injection of poly(I:C). We conducted a bulk RNA-seq analysis of tumors post-treatment to capture the effects on innate and adaptive immune responses. At 24 hours, poly(I:C) had a more pronounced impact on gene modulation in the tumor microenvironment compared with artLCMV-E7E6, with 1244 genes being upregulated or downregulated vs 751 genes modulated by artLCMV-E7E6 (online supplemental figure 1). The pathways modulated by poly(I:C) were related to immune responses, reflecting the activation of antigen-presenting cells and T cell-mediated cytotoxicity (online supplemental figure 2). In the case of artLCMV-E7E6, an early response to the virus was identified, with a prominent role for interferon-β induction (online supplemental figure 2). Interestingly, the combination therapy activated a distinct transcription program, different from the program triggered by each monotherapy. Indeed, the combination exhibited the highest number of upregulated or downregulated genes (2062 genes) (figure 2A and online supplemental figure 1). Compared with single-agent poly(I:C), the combination upregulated pathways related to the innate immune response to viruses, and downregulated neutrophil-related genes (figure 2B). In contrast, when comparing the transcriptional program induced by the combination and that initiated by artLCMV-E7E6, several pathways associated with upregulation of cytotoxic T lymphocyte responses and downregulation of pathways involved in lymphocyte proliferation, migration, and adhesion were observed (figure 2C). Notably, the activation of these innate immune pathways upregulated several MHC-related genes (figure 2D). This result was corroborated at the protein level on the cell surface of malignant cells using flow cytometry to analyze the expression of H2-Kb on tumor cells (figure 2E). To assess the divergent impact of the combined therapy on poly(I:C)-injected tumors and non-poly(I:C)-injected tumors at a distal site, an additional RNAseq analysis was conducted 24 hours post-therapy. A marked upregulation in the poly(I:C)-injected lesion of various pathways activated by type I interferon was noted, including the activation of NK cell-mediated immunity and MHC presentation. Furthermore, a more pronounced upregulation of pathways associated with immune effector functions was identified. Thus, the heightened exposure of poly(I:C) in the treated lesion is evident from the upregulation of these pathways (online supplemental figure 3).

Figure 2

The combination of systemic artLCMV-E7E6 and intratumoral poly(I:C) induces a distinct gene expression program compared with each monotherapy in the tumor microenvironment. TC1/A9 tumor-bearing mice received intravenous treatment with 1×105 RCV FFU artLCMV-E7E6 on day 7 and intratumoral injections of 50 µg of poly(I:C) in the right tumor lesion on days 7, 10, and 13. On day 8 and day 14, mice were euthanized, and RNA from tumors was subjected to RNA-seq analysis. (A) Venn diagram illustrating the comparison of differentially expressed genes (p<0.05% and FDR<0.05%) in the various experimental groups 24 hours after treatment. (B) Gene Set Enrichment Analysis (GSEA) showing the top upregulated and downregulated Gene Ontology Biological Process (GO:BP) terms comparing artLCMV-E7E6+poly(I:C) with poly(I:C) 24 hours after treatment (p. adj<0.05%). (C) GSEA illustrating the top upregulated and downregulated GO:BP terms comparing artLCMV-E7E6+poly(I:C) with artLCMV-E7E6 24 hours after treatment (p. adj<0.05%). (D) Heatmap displaying the level of gene expression related to the pathway of antigen processing and presentation via MHC. (E) Mean fluorescence intensity (MFI) of MHC class I on tumor cells 24 hour after treatment analyzed by using flow cytometry. Data are presented as mean±SEM, and statistical significance was determined with one-way ANOVA followed by Dunnet’s post-test comparing the combined treatment with the other experimental groups (**p<0.01). (F) Venn diagram demonstrating the comparison of differentially expressed genes (p<0.05% and FDR<0.05%) in the various experimental groups 7 days after treatment. (G) GSEA showing top upregulated and down-regulated GO:BP terms comparing artLCMV-E7E6+poly(I:C) with poly(I:C) 7 days after treatment. (p. adj<0.05%). (H) GSEA showing top upregulated and downregulated GO:BP terms comparing artLCMV-E7E6+poly(I:C) with artLCMV-E7E6 7 days after treatment (p. adj<0.05%). ANOVA, analysis of variance; LCMV, lymphocytic choriomeningitis virus; PBS, phosphate-buffered saline; poly(I:C), polyinosinic:polycytidylic acid; RCV FFU, replication competent virus focus forming units; FDR, false discovery rate; MHC, major histocompatibility complex.

Figure 3

Biodistribution of LCMV vectors remains unaffected by intratumoral administration of poly(I:C). C57BL/6 mice were subcutaneously injected in the right flank with 1.5×105 TC-1/A9 cells. Seven days later, mice received intravenous injections of 1×105 RCV FFU artLCMV-NanoLuc. (A) Representative image of in vivo bioluminescence measured 24 hours post-LCMV injection. (B) Quantification of the in vivo bioluminescence measured in all mice. (C) After in vivo imaging, mice were euthanized, and the bioluminescence in different organs was assessed. Representative images are shown. (D) Quantification of the ex vivo bioluminescence in different organs. Data are presented as mean±SEM, and statistical significance was determined with one-way ANOVA followed by Dunnet’s post-test for panels (B, D). ANOVA, analysis of variance; LCMV, lymphocytic choriomeningitis virus; PBS, phosphate-buffered saline; poly(I:C), polyinosinic:polycytidylic acid; RCV FFU, replication competent virus focus forming units.

The evaluation of the transcriptional programs 7 days after initiating treatments indicated that at this time point, the major driver of gene expression changes in the tumor microenvironment was the administration of artLCMV-E7E6, with 3565 genes upregulated or downregulated (online supplemental figure 4). Although the last administration of poly(I:C) was given 24 hours before the tumor RNA isolation, fewer genes were mobilized by poly(I:C), involving 1101 transcripts in this case (figure 2F and online supplemental figure 4). Importantly, the combined treatment mobilized the highest number of genes (4437), with 1321 genes exclusively affected by the combined treatment, suggesting a specific modulation of the adaptive immune response by the combined treatment (figure 2F,G and online supplemental figure 5). Comparing the Gene Ontology pathways modulated by the combination and artLCMV-E7E6 monotherapy provided insights into the mechanisms associated with the combination, impacting several pathways that indicate remodeling of tumor vasculature, leucocyte migration, and antigenic presentation processes (figure 2H).

Figure 4

The combination of systemic artLCMV-E7E6 and intratumoral poly(I:C) upregulates specific chemokines/cytokines. TC1/A9 tumor-bearing mice received intravenous treatment with 1×105 RCV FFU artLCMV-E7E6 and intratumoral injection of 50 µg of poly(I:C) in the right tumor lesion on day 7. After 24 hours, multiple cytokine/chemokines were assessed in serum samples. Data are presented as mean±SEM, and statistical significance was determined with one-way ANOVA followed by Dunnet’s post-test (*p<0.05; **p<0.01; ***p<0.001). ANOVA, analysis of variance; LCMV, lymphocytic choriomeningitis virus; poly(I:C), polyinosinic:polycytidylic acid; RCV FFU, replication competent virus focus forming unit.

Figure 5

The combination of systemic artLCMV-E7E6 and intratumoral poly(I:C) induces a significant infiltration of highly activated tumor-specific lymphocytes in the tumor microenvironment. TC1/A9 tumor-bearing mice were intravenously treated with 1×105 RCV FFU artLCMV-E7E6 at day 7 and received intratumoral injections of 50 µg of poly(I:C) in the right tumor lesion on days 7, 10, and 13. On day 15, mice were euthanized, and immune cells from tumors were analyzed using multiparametric flow cytometry. (A) Uniform Manifold Approximation and Projection (UMAP) plots illustrating the immune subsets in the different experimental conditions. (B) Quantification of the percentage of each immune subset and comparison between pairs of experimental conditions. Statistical differences were determined using the Wilcoxon test (ns, non-significant; *p<0.05). (C) Heatmap displaying differentially expressed proteins in CD8+ T lymphocytes in artLCMV-E7E6+poly(I:C) vs artLCMV-E7E6. LCMV, lymphocytic choriomeningitis virus; poly(I:C), polyinosinic:polycytidylic acid; RCV FFU, replication competent virus focus forming units.

Intratumoral administration of poly(I:C) does not alter artLCMV vector biodistribution

We explored whether poly(I:C) impacts vector biodistribution or viral gene expression by using an artLCMV encoding NanoLuc (Promega). The encoded reporter enzyme is an engineered luciferase that catalyzes its substrate, furimazine, to produce light with enhanced brightness, thermal stability, pH stability, and unbiased distribution of NanoLuc in cells.40 Twenty-four hours after intravenous administration of the artLCMV-NanoLuc vector, a strong bioluminescence signal was detected in the abdominal area in mice treated with the vector alone or in combination with intratumoral poly(I:C) (figure 3A). In both experimental conditions, the bioluminescence signal reached levels above 105 ph/s/cm2/sr, with no significant differences observed between groups (figure 3B). To further investigate the biodistribution of the vector, we sacrificed mice, and determined bioluminescence in the spleen, kidney, tumor, lymph nodes, omentum, and liver. The organs with the highest expression of the reporter were the liver and spleen, both with values above 105 ph/s/cm2/sr. In the other tissues, a significant expression was observed, though the bioluminescence values were one log lower. In any case, there were no significant difference when poly(I:C) was coadministered. Based on these results, we can conclude that the improved antitumor activity observed with the combination therapy was not due to differences in viral vector replication or gene expression.

Intratumoral administration of poly(I:C) induced changes in the innate and adaptive immune responses elicited by systemic administration of artLCMV-E7E6

To validate the results obtained by RNA-seq, we analyzed blood levels of several cytokines and chemokines 24 hours after initiating treatments to assess changes in the innate response. To evaluate the adaptive T-cell response, we conducted flow cytometry analysis 7 days after starting the treatments on both circulating tumor-specific T lymphocytes and the immune infiltrate in the tumor microenvironment.

Regarding the innate immune response, the combined treatment increased the levels of chemokines such as CCL3, CCL5, and CXCL10. Moreover, this combined treatment also elevated several proinflammatory cytokines that are essential for the development of an antitumor effector immune response. These included TNF, IFNα, and IL12p70 (figure 4). Intriguingly, the combined treatment led to lower levels of IFNα and IFNγ, at least at the chosen time point (figure 4).

Seven days after starting the treatments, a high percentage of circulating tumor-specific T lymphocytes was observed in both experimental groups that received the LCMV vector, with no significant difference between the combination and the monotherapy with artLCMV-E7E6 conditions (online supplemental figure 6A). In the tumor microenvironment of both groups that included the LCMV vaccine, the infiltration of CD8+ T lymphocytes was significantly increased, while CD4+ T lymphocytes were decreased (figure 5A,B). Significant differences were also identified for NK cells, B cells, and dendritic cells when comparing the combined treatment and vector monotherapy. Moreover, the higher expression levels of activation or proliferation markers such as tumor-specific TCRs, FoxP3, CD25, and Ki67 were indicative of a more active state of the infiltrating CD8+ T lymphocytes on combined treatment (figure 5C and online supplemental figure 6B,C).

Figure 6

The antitumor activity of the combination of systemic artLCMV-E7E6 and intratumoral poly(I:C) is preserved in Batf3-deficient mice and requires the presence of macrophages and CD8+ T lymphocytes but not of CD4+ T lymphocytes or NK cells. (A) C57BL/6 mice or Batf3−/− mice were subcutaneously injected with 1.5×105 TC-1/A9 cells in the right flank. Seven days later, mice were intravenously treated with 1×105 RCV FFU artLCMV-E7E6 and received intratumoral injections of 50 µg of poly(I:C) on days 7, 10 and 13. Individual tumor growth (mm3) is shown for each mouse, with numbers in each graph representing the fraction of mice achieving complete tumor regression. (B) The average in vivo tumor growth is shown. (C) The percentage of survival over time. Data are represented as mean±SEM. Extra sum-of-squares F test (A) or log-rank (C) tests were used to assess the significance between the treated mice and control mice in each mouse strain. No significant differences were observed when comparing artLCMV-E7E6+poly(I:C) treatment in wild type versus Batf3−/− mice. (D) Mice were subcutaneously injected with 1.5×105 TC-1/A9 cells in the right flank. Seven days later, mice were intravenously treated with 1×105 RCV FFU artLCMV-E7E6 and received intratumoral injections of 50 µg of poly(I:C) in the right tumor lesion on days 7, 10, and 13. Depleting monoclonal antibodies or clodronate liposomes were administered on days 6, 9, and 13. Individual tumor growth (mm3) is shown for each mouse. (E) The average in vivo tumor growth is presented. (F) The percentage of survival over time. Data are represented as mean±SEM Extra sum-of-squares F test (A, D) or log-rank (C, F) tests were used to assess significance (**p<0.01, ***p<0.001). LCMV, lymphocytic choriomeningitis virus; PBS, phosphate-buffered saline; poly(I:C), polyinosinic:polycytidylic acid; RCV FFU, replication competent virus focus forming units.

The antitumor immune response elicited by artLCMV-E7E6 and poly(I:C) relies on CD8+ T lymphocytes and macrophages but is independent of cDC1, CD4+ T lymphocytes or NK cells

To identify the cellular requirements for the antitumor activity of the combined treatment, we used knockout mice, depleting monoclonal antibodies, and macrophage-depleting clodronate liposomes.

First, we compared the antitumor effect in wild-type and Batf3-deficient mice. These mice lack functional cDC1, a dendritic cell subset specialized in cross-presentation and IL-12 release. Surprisingly, the combined treatment exerted a potent antitumor effect in both mouse strains, with no differences observed in tumor growth or survival between wild-type or Batf3−/− mice treated with the combination therapy (figure 6A–C). Next, we depleted macrophages using clodronate liposomes to identify critical antigen-presenting cells in this system. This treatment abrogated the antitumor activity of the combination, indicating an important role for macrophages in the therapeutic effects of the combined treatment (figure 6D–F). In this setting, depletion of CD4+ T lymphocytes or NK cells had no impact, while depletion of CD8+ T lymphocytes significantly reduced the antitumor effect (figure 6D–F).

Intratumoral injection of BO-112 improves the antitumor efficacy of artLCMV-E7E6 in both injected and non-injected tumor lesions

To assess whether the antitumor efficacy was maintained with a GMP-grade formulated form of poly(I:C), we used intratumoral BO-112 (a nanoplexed poly(I:C)-based compound) in combination with intravenous administration of artLCMV-E7E6 in the bilateral tumor model (figure 7A). Each treatment alone achieved only a slight delay in tumor growth, while the combination resulted in a remarkable synergy, eradicating the bilateral tumors in 80% of treated mice (figure 7B–D).

Figure 7

Intratumoral BO-112 enhances the antitumor activity of artLCMV-E7E6 in both injected tumor lesions and distant non-injected engrafted tumors. Mice were subcutaneously injected with 1.5×105 TC-1/A9 cells in the right flank and with 7.5×104 TC-1/A9 cells in the left flank cells. Seven days later, mice received intravenous treatment with 1×105 RCV FFU artLCMV-E7E6 and intratumoral injections of 50 µg of BO-112 in the right tumor lesion on days 7, 10, and 13. (A) Schematic representation of the experimental setup. (B) The tumor growth (mm3) is depicted for each individual mouse in both injected (right flank) and non-injected (left flank) tumors. The numbers in each graph represent the fraction of mice achieving complete tumor regression. (C) The average in vivo tumor growth is presented for injected (right flank) and non-injected (left flank) tumors. (D) The percentage of survival over time is shown for experiments in B. n=9–10 (mean±SEM). Extra sum-of-squares F test (C) or log-rank (D) tests were used to assess significance. Significant differences are indicated for comparisons of each group with the artLCMV-E7E6+BO-112 group (**p<0.01). LCMV, lymphocytic choriomeningitis virus; poly(I:C), polyinosinic:polycytidylic acid; RCV FFU, replication competent virus focus forming units.

A side-by-side comparison of BO-112 and poly(I:C) in combination with art-LCMV-E7E6 was performed in an experimental setting where the combination encompassing poly(I:C) failed to eliminate both tumors. Thus, treatment was initiated when tumors on the right surpassed a diameter of 6 mm. In this scenario, the combination of BO-112 demonstrated significantly higher antitumor effects in both the right tumor lesion, injected with the TLR3 ligand, and the left tumor lesion, which did not receive the TLR3 ligand. Notably, 33% of the mice completely eradicated the tumor and survived until the end of the experiment (online supplemental figure 7).

Analysis of circulating cytokines and chemokines 24 hours after administering the combination treatment revealed a similar activation pattern, although with significantly higher levels of CCL3 and TNF induced by the BO-112 combination with the viral vector compared with poly(I:C) and artLCMV-E7E6 (online supplemental figure 8).

Flow cytometry analysis of both tumor lesions 7 days after initiating the combined treatments demonstrated that both combinations upregulated CD8+ T lymphocytes and decreased T regulatory cells in the right tumor lesion (injected with BO-112 or poly(I:C)) and in the uninjected left tumor lesion. The higher variability observed in CD8+ T lymphocytes in the uninjected tumor lesions in mice treated with poly(I:C)+artLCMV-E7E6 may account for the reduced efficacy (online supplemental figure 9). Both combinations induced polyfunctional CD8+ and CD4+ T lymphocytes capable of releasing IFNγ and degranulating on restimulation with a panel of E7 and E6 peptides (online supplemental figure 10).

The most pronounced impact of BO-112 was further validated at the RNA level. BO-112-treated lesions demonstrated a more substantial modulation of gene expression compared with poly(I:C)-treated lesions. Interestingly, the analysis of the pathways activated by BO-112 highlighted pathways that are not influenced by poly(I:C), such as the activation of the cellular response to interleukin 1. This suggests that cytokines of the interleukin 1 family may contribute to the superior antitumor effect of BO-112 compared with poly(I:C) (online supplemental figure 11).

Therefore, the combination with clinical-grade BO-112 clearly outperformed the antitumor efficacy achieved with the combination involving unformulated poly(I:C).

Discussion

Therapeutic antitumor vaccination has been a long-standing focus of numerous preclinical and clinical studies.41 42 However, as of now, the only FDA-approved antitumor vaccine is sipuleucel T, which is an autologous cellular immunotherapy generated by activating peripheral blood mononuclear cells in cell culture using a recombinant human fusion protein of prostatic acid phosphatase and granulocyte–macrophage colony-stimulating factor.43 The discovery of immune checkpoint inhibitors, neoantigens, and innovative vaccine vectors has rekindled interest in therapeutic cancer vaccines.44 45 Personalized neoantigen vaccines based on mRNA have made significant progress in clinical trials, yielding promising results.46–50 Additionally, intense clinical research is ongoing with conventional tumor-associated antigens using novel vaccine platforms.41 One such platform is replication-competent, stably-attenuated LCMV vectors, which have demonstrated a remarkable ability to expand tumor-specific T lymphocytes.10–12 In our study, we confirm the induction of impressive levels of circulating tetramer-positive CD8+ T cells just 1 week after intravenous LCMV administration. Given that tumors can evade CD8+ T cell-mediated antitumor immunity by downregulating MHC class I expression,51–53 we sought to investigate this vector in a clinically relevant tumor model with low expression levels of MHC class I, mirroring observations in human cervical carcinoma.54 Low MHC-I expression has been established to render tumors resistant to therapeutic vaccination.55 Interestingly, we observed massive infiltration of tumor-specific T lymphocytes on treatment with artLCMV-E7E6 in this low MHC-expressing tumor model, ruling out tumor infiltration failure as the primary limitation for this monotherapy. To sensitize the tumors to the effector CTL immune response, we explored the administration of intratumoral poly(I:C) in a single tumor lesion. Prior studies had reported synergistic effects of intratumoral administration of interferon-alpha with other immunotherapies in various preclinical settings.16–18 56 However, this approach had not been used in combination with a virus because type I interferon is a potent inhibitor of viral replication and expression.57 Intriguingly, the local administration of poly(I:C), a potent type I interferon inducer, did not affect LCMV gene expression in the injected tumor or elsewhere. In contrast, we detected activation of immune gene expression programs in the tumor microenvironment just 1 day after poly(I:C) injection, promoting the migration, proliferation, and activity of leucocytes. This activation of the innate immune response was also reflected in increased levels of several cytokines/chemokines measured in circulation. Importantly, this potent proinflammatory response rescued the expression of MHC class I molecules on tumor cells, rendering them more susceptible to attack by CD8+ T lymphocytes. Of note, direct activation of type I interferon was not detected. Hence, the synergy between heightened susceptibility in tumor cells and enhanced activation of infiltrating immune cells may underscore the robust antitumor effect resulting from the combined treatment.

One element of the combination involves administering poly(I:C) intratumorally, while the LCMV vector is delivered intravenously. However, a significant drawback of this approach is the potential restriction of the antitumor effect to the poly(I:C)-treated lesion, thereby limiting its overall impact on the disease course.22 58 We implemented a bilateral tumor model to evaluate whether our combined immunotherapy could induce therapeutic effects in both poly(I:C)-treated and untreated lesions. Encouragingly, we observed tumor eradication in both treated and untreated tumor sites, suggesting the potential for translational investigations into the efficacy of this combination.

Several differences between the combination and each monotherapy were identified at early and late time points, explaining the synergistic effects on the antitumor immune response. Twenty-four hours after administration, higher levels of circulating CCL3, CCL5, CXCL10, TNF, IFNα, and IL12p70 were observed, along with an upregulation of pathways in the tumor microenvironment related to the innate immune response to viruses. Therefore, poly(I:C) not only did not interfere with LCMV replication but actually produced synergistic effects by signaling danger in the tumor microenvironment, reshaping the effector immune response. This early innate immune response likely contributed to differences in the adaptive immune response observed 5 days later. Significantly, the combination treatment not only increased the proliferation of CD8+ T lymphocytes but also upregulated activation markers in this crucial immune cell population. Nevertheless, no differences were observed in the total infiltration of CD8+ T cells into the tumor when comparing the combined therapy to each monotherapy. Hence, it is the qualitative alterations rather than quantitative changes that underlie the efficacy of the combination. Regardless, CD8+ T lymphocytes were essential, as the depletion of CD8+ T lymphocytes abolished the antitumor effect.

Surprisingly, the crucial antigen-presenting cells were not Batf3-expressing cDC1 cells but rather macrophages or other dendritic cell types. Prior reports have established the significance of these phagocytic cells in the immune response against LCMV.59 60 Given that most developed immunotherapies rely on cDC1 cells,61 the capacity of LCMV to be effective in patients with low levels of these antigen-presenting cells is a potential advantage that merits further evaluation. In this context, direct presentation rather than cross-presentation appears to be important.62

Lastly, we evaluated the antitumor efficacy of the combination of LCMV vectors and BO-112, a poly(I:C) formulation currently under evaluation in clinical trials.23 24 Intratumoral administration of BO-112 significantly potentiated the antitumor efficacy of systemic LCMV vaccine administration in both BO-112-treated and untreated tumor lesions. Importantly, the antitumor effect of BO-112 was markedly superior to that of unformulated poly(I:C) in the same setting. The observed antitumor effect on well-established distant non-injected lesions underscores the clinical relevance of this combination and justifies its development in patients with injectable HPV-associated tumors. It is essential to note that all the data presented in this study were obtained from experiments conducted in mice, and several variables may limit the direct application of this combination in humans. For example, the proximity of bilateral tumors in mice may enhance the leakage of the TLR3 ligand from the injected lesion to the non-injected lesion. Therefore, further studies will be required to assess the therapeutic potential of this combination in humans. Both BO-112 and artLCMV-E7E6 are being evaluated as monotherapies, and in combination with checkpoint inhibitors (NCT04180215 for HB-200 and NCT04570332, NCT04508140, NCT04777708, NCT04420975, NCT05265650 for BO-112) making clinical trials of this strategy feasible.

Data availability statement

Data are available in a public, open access repository. Data are available on reasonable request. RNAseq datasets are available on the Gene Expression Omnibus (GEO) website (accession number: GSE245436 and GSE256478).

Ethics statements

Patient consent for publication

Ethics approval

Experiments involving mice were approved by the Ethics Committee of the University of Navarra (055-21).

Acknowledgments

We are grateful to Paul Miller for English editing.

References

Supplementary materials

  • Supplementary Data

    This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.

Footnotes

  • Contributors HL and PB designed experiments. CG, CADT, AB, LA, MF-S, JSR-C, NA and FA performed the experiments and processed samples. TS designed and generated the artLCMV-NanoLuc vectors. MQ provided BO-112 and provided details for experimental usage thereof. IM and KKO contributed to the main conceptional idea of the study. JG-G, MA and PB performed all statistical analyses and analyzed the data. PB wrote the manuscript. All authors critically reviewed the manuscript for important intellectual content and gave final approval of the manuscript. PB is responsible of the overall content as the guarantor.

  • Funding This study was supported by Hookipa Pharma, Instituto de Salud Carlos III (PI20/00002 and PI22/00147), cofinanced by Fondos Feder and Gobierno de Navarra Proyecto ARNMUNE Ref.: 0011-1411-2023. Work produced with the support of a 2022 Leonardo Grant for Researchers and Cultural Creators (BBVA Foundation). MA was supported by the Spanish Association against Cancer Research (AECC-2019 Investigator) and is recipient of the “Ayudas Ramon y Cajal” (RYC2021-033381) from the MCIN/ AEI/10.13039/501100011033. FA receives a Miguel Servet I (CP19/00114) contract from ISCIII (Instituto de Salud Carlos III) cofinanced by FSE (Fondo Social Europeo). AB is recipient of PFIS fellowship from ISCIII (FI20/00058), and LA is the recipient of an FPU grant from The Spanish Ministry of Education and Professional training (FPU21/00042).

  • Competing interests PB received research funding from Hookipa Pharma. IM reports receiving commercial research grants from AstraZeneca, BMS, Highlight Therapeutics, Alligator, Pfizer Genmab and Roche; has received speakers bureau honoraria from MSD; and is a consultant or advisory board member for BMS, Roche, AstraZeneca, Genmab, Pharmamar, F-Star, Bioncotech, Bayer, Numab, Pieris, Gossamer, Alligator and Merck Serono. MA declares receiving a commercial research grant from Highlight Therapeutics. TS, KKO, and HL are employees of Hookipa Pharma. MQ is an employee of Highlight Therapeutics. The rest of the authors have no conflict of interest to declare.

  • Provenance and peer review Not commissioned; externally peer reviewed.

  • Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.