You are here

  • Home
  • Archive
  • Volume 12, Issue 7
  • Second-generation checkpoint inhibitors and Treg depletion synergize with a mouse cancer vaccine in accordance with tumor microenvironment characterization

Article Text

Article menu
Clinical/translational cancer immunotherapy
Original research
Second-generation checkpoint inhibitors and Treg depletion synergize with a mouse cancer vaccine in accordance with tumor microenvironment characterization
  1. William Becker1,
  2. Purevdorj B Olkhanud1,
  3. Noriko Seishima1,
  4. Paloma A Moreno1,
  5. Katherine C Goldfarbmuren1,2,
  6. Hoyoung M Maeng1 and
  7. Jay A Berzofsky1
  1. 1Vaccine Branch, CCR, National Cancer Institute, National Institutes of Health, Bethesda, Maryland, USA
  2. 2Advanced Biomedical Computational Science, Leidos Biomedical Research Inc, Frederick, Maryland, USA
  1. Correspondence to Dr William Becker; beckerwj{at}nih.gov

Abstract

Background Despite advances in checkpoint inhibitor (CPI) therapy for cancer treatment, many cancers remain resistant. Tumors deemed “cold” based on lack of T cell infiltration show reduced potential for CPI therapy. Cancer vaccines may overcome the inadequacy of existing T cells by inducing the needed antitumor T cell response to synergize with CPIs and overcome resistance.

Methods CT26 and TC1 tumor cells were injected subcutaneously into mice. Mice were treated with combinations of CPIs alone or a cancer vaccine specific to the tumor antigen E7 present in TC1 cells. CPIs for the TC1 model were selected because of immunophenotyping TC1 tumors. Antitumor and protumor immunity, tumor size and survival, sequence and timing of vaccine and CPI administration, and efficacy of treatment in young and aged mice were probed.

Results While “hot” CT26 tumors are treatable with combinations of second-generation CPIs alone or with anti-TGFβ, “cold” TC1 tumor reduction requires the synergy of a tumor-antigen-specific vaccine in combination with two CPIs, anti-TIGIT and anti-PD-L1, predicted by tumor microenvironment (TME) characterization. The synergistic triple combination delays tumor growth better than any pairwise combination and improves survival in a CD8+T cell-dependent manner. Depletion of CD4+T cells improved the treatment response, and depleting regulatory T cells (Treg) revealed Tregs to be inhibiting the response as also predicted from TME analysis. We found the sequence of CPI and vaccine administration dictates the success of the treatment, and the triple combination administered concurrently induces the highest E7-specific T cell response. Contrary to young mice, in aged mice, the cancer vaccine alone is ineffective, requiring the CPIs to delay tumor growth.

Conclusions These findings show how pre-existing or vaccine-mediated de novo T cell responses can both be amplified by and facilitate synergistic CPIs and Treg depletion that together lead to greater survival, and how analysis of the TME can help rationally design combination therapies and precision medicine to enhance clinical response to CPI and cancer vaccine therapy.

  • Combination therapy
  • Immune Checkpoint Inhibitor
  • T regulatory cell - Treg
  • Tumor infiltrating lymphocyte - TIL
  • Vaccine

Data availability statement

All data relevant to the study are included in the article or uploaded as online supplemental information. Values for all data points in graphs are reported in online supplemental file 2.

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/.

https://doi.org/10.1136/jitc-2024-008970

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

    • Checkpoint inhbitiors (CPIs) are effective at treating certain cancers, typically those with high CD8+T cell infiltration. Cancer vaccines can stimulate tumor antigen-specific CD8+T cells.

    WHAT THIS STUDY ADDS

    • Cancer vaccines can synergize with rationally selected CPIs to succeed at delaying tumor growth where either treatment singly is insufficient. Giving CPIs concurrently or after cancer vaccination, but not before, improves the response, and depletion of T regulatory cells (Tregs) leads to complete remission.

    HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY

    • The rational selection of CPIs based on tumor immunophenotyping combined with the resurgence of cancer vaccines may lead to improved progression-free and overall survival in patients with cancer.

    Introduction

    As a heterogeneous and multifactorial disease, cancer requires treatment modalities personalized to each patient. Immunotherapy as a tool to harness the host anticancer immune response has emerged as a breakthrough in oncology, poised to meet the demands of precision medicine. The primary approved immunotherapy agents include antibodies that target the PD-1/PD-L1 axis or CTLA-4; however, more than 55% of patients with cancer are ineligible for existing immune checkpoint blockade,1 and less than 25% of eligible patients achieve an objective response.2 Thus, finding new immunotherapies and developing strategies to enhance pre-existing treatments will be key to improving this nascent technology.

    After the initial wave of success with antibodies targeting the immune checkpoints PD-1/PD-L1 or CTLA-4, progress slowed, and combining the two first-generation checkpoint inhibitors (CPIs) often resulted in toxicity at the price of hope for improved efficacy.3 In the wake of these first-generation CPIs, “second/next generation” CPIs emerged as promising clinical targets for tumor suppression. Among the new wave of CPIs are antibodies targeting LAG-3, TIGIT, Tim3, KLRG1,4 and VISTA.5 While these second-generation CPIs may provide higher safety profiles than PD-1 or CTLA-4, their antitumor efficacy is primarily limited to preclinical studies. Drugs targeting TGFβ are still under investigation,6 and anti-TIGIT and anti-LAG-3 have only been approved to treat non-small cell lung cancer (NSCLC) and advanced melanoma in combination strategies, respectively. Further combinations are warranted, but predicting which CPI a patient will respond to is a daunting process.

    Myriad predictors of patient response to CPI therapy exist, with tumor mutational burden (TMB), PD-L1 expression, T-cell immune infiltrate, genetic factors, and the microbiome being implicated. TMB, when combined with other factors, is the most consistent pan-cancer predictor of success,7 and TMB correlation to CPI success depends on enough antitumor lymphocyte infiltration to react to non-self antigenic drift. Tumor-reactive T cells are enriched among CD8+T cells expressing activation or exhaustion markers PD-1, Tim3, and LAG-3,8 and T cell responses, more than B cell responses, are the predominant component of lymphocyte antitumor activity activated by CPIs.9 Thus, stimulating a potent antitumor CD8+T cell response is paramount to improving antitumor CPI responsiveness. Cancer vaccines are a recently resurrected strategy to marshall the much-needed antitumor CD8+T cell response to pair with CPIs.10 11 The decades of work improving vaccines and unearthing tumor-specific antigens have primed cancer vaccines for a resurgence in oncologic use, and combining cancer vaccines with CPIs promises to be a salutary combination.12 13

    Here, we report combinations of second-generation CPIs with anti-TGFβ to be effective at delaying tumor growth in the “hot” CT26 tumor model. However, in the “cold” TC1 tumor, CPIs alone fall short, and the CPI combination of anti-TIGIT and anti-PD-L1 are effective only when promoting a tumor-antigen specific CD8+T cell response elicited via a vaccine, synergizing with the vaccine to delay tumor growth better than vaccine or CPIs alone. Tumor-intrinsic factors were probed, revealing that TC1 tumors exhibit greater CD155 expression on myeloid cells, and higher infiltration of Treg cells compared with CT26 tumors, such that Treg cell depletion resulted in complete tumor remission when combined with the triple combination of cancer vaccine plus CPIs (anti-TIGIT and anti-PD-L1). We found concurrent administration of vaccine plus CPIs to be the ideal paradigm that was then used to delay tumor growth in experiments using aged mice. We show how to effectively transition a CPI-resistant tumor into a CPI-susceptible tumor through the use of cancer vaccines. More broadly, we see that such immunological characterization of a tumor microenvironment (TME) allows for the selection of the most effective combination immunotherapy and points the way to a more general stratagem for personalized “precision medicine” cancer immunotherapy.

    Materials and methods

    Mice

    Female BALB/c (strain code: 028) or C57BL/6 (strain code: 027) mice aged 8–12 weeks old were purchased from Charles River and FoxP3-GFPDTR (B6.129(Cg)-Foxp3tm3(Hbegf/GFP)Ayr/J, RRID:IMSR_JAX:016958) transgenic mice and WT controls were purchased from Jackson Laboratories. Aged C57BL/6 mice were purchased from Charles River at a young age (8–12 weeks old) and maintained in-house at the NIH animal facility until old enough to be used for aged experiments (>18 months). All animals were bred and housed at the NIH Animal facility.

    Mouse tumor studies

    CT26 (RRID:CVCL_7256) and TC1 (RRID:CVCL_4699) tumor cells were cultured in a complete RPMI medium (containing 10% FBS and 1X penicillin/streptomycin) at 37°C and 5% CO2. TC1, a C57BL/6-derived lung epithelial cell line transfected with human papilloma virus (HPV) 16 E6 and E7 genes, was cultured with G418 (Geneticin, GIBCO), to select for transfectants. Cells were harvested by trypsinization (GIBCO) and washed with PBS. Cells were resuspended in PBS and 2.5×105 CT26 or 1×105 TC1 cells were injected subcutaneously into the flanks of mice. Health, weight, and tumor size were tracked twice per week until the endpoint of the study.

    CPI and cancer vaccine treatments

    Around 7–8 days post-tumor inoculation, or when average tumor volume reached ~50 mm3, mice received the first of four intraperitoneal injections over 2 weeks (200 µg/dose) of CPIs (anti-TIGIT (clone: 1G9, RRID:AB_2687797), anti-LAG-3 (clone: C9B7W, RRID:AB_10949602), anti-KLRG1 (clone: 2F1, RRID:AB_10949054), anti-Tim3 (clone: RMT3-23, RRID:AB_10949464), or anti-PD-L1 (clone:10F.9G2, RRID:AB_10949073), all BioXCell), anti-TGFβ (clone: 1D11, Genzyme) or the relevant control IgG. For cancer vaccine studies in the TC1 model, mice were given a single subcutaneous vaccine concurrently (unless otherwise stated) with the first administration of CPIs consisting of 1 µg of E7 (43–77) synthetic long peptide from HPV16 E7 oncogene (United Biosystems), 250 nmol of alpha-galactosylceramide (aGalCer), 1 µg of granulocyte-macrophage colony-stimulating factor (GM-CSF) (Biolegend), and 20 µg of DOTAP liposomal transfection reagent (Roche) all of which were mixed together in PBS before administration.

    In vivo depletion

    In vivo depletion of CD4+T cells, CD8+T cells or the combination of the two was achieved by intraperitoneal injection of anti-CD4 (clone GK1.5, RRID:AB_1107636), anti-CD8 (clone 53–6.7, RRID:AB_1107671) or isotype control IgG (RRID:AB_1107769) (all from BioXCell) with a 200 µg dose of antibody −2 days before CPI administration and a 100 µg dose every 2–3 days after that, continuing until termination of the study. For FoxP3+Treg cell depletion, FoxP3-GFPDTR (B6.129(Cg)-Foxp3tm3(Hbegf/GFP)Ayr/J, RRID:IMSR_JAX:016958) mice were injected with 10 µg/kg of diptheria toxin every 2–3 days starting −2 days before CPI administration and continuing until termination of the study or humane endpoints were reached. Health, weight, and tumor size were tracked three times per week. Treg cell depletion was confirmed by flow cytometry of FoxP3-GFP+in blood and tumors.

    Isolation of mononuclear cells

    Spleens were minced over a 70 µm filter, spun down and red blood cells were removed using ACK lysing buffer. Cells were then passed through a 40 µm filter to obtain a single-cell suspension. For tumor dissociation, tumors were harvested, fat and necrotic tissue was removed and subjected to dissociation using the gentleMACS tissue dissociator with the Tumor Dissociation Kit (Miltenyi Biotec) according to manufacturer’s directions. Once a single-cell suspension was obtained, cells were passed through an 80%/40% percoll gradient and the interface was collected and washed with PBS+2% FBS twice to obtain mononuclear cells (MNCs). MNCs from each mouse were cultured in culture media (RPMI1640 supplemented with 10% heat-inactivated FBS, penicillin-streptomycin, L-glutamine, HEPES, NEAA, sodium pyruvate, and 2-ME) in 96-well plates at 37°C for 6 hours in the presence of E7 (43-77) long peptide and brefeldin A (BD Biosciences).

    Flow cytometry

    Once a single cell suspension was obtained, MNCs were washed with PBS and incubated on ice with Fc block and Live/Dead fixable blue dead cell stain (Invitrogen) for 20 min at RT. Cells were then labeled for surface markers on ice for 30 min, fixed/permeabilized using BD Cytofix/Cytoperm solution kit (BD Biosciences) for detection of intracellular proteins, or the FoxP3/Transcription Factor Staining Buffer Set (Invitrogen) for detection of intranuclear proteins according to the manufacturer’s instructions. Cells were then washed and labeled with an intracellular antibody cocktail. All flow cytometry data were acquired on a BD FACS Symphony A5 and were analyzed using FlowJo Software (RRID:SCR_008520). Antibodies used are detailed in table 1.

    View this table:
    Table 1

    Flow cytometry antibodies

    Immunohistochemistry

    Tumors were excised and fixed in 10% formalin overnight then transferred to 70% ethanol. FFPE blocks were sectioned on a microtome and placed onto charged slides. Slides were deparaffinized in xylenes and hydrated using graded alcohols. Slides were then retrieved in a sodium citrate buffer and blocked with a peroxidase block. Next, slides were incubated with anti-CD8 (CST #98941, 1:250, 2.56 µg/mL) antibody and detected with a cross-adsorbed, HRP-conjugated anti-rabbit secondary, followed by DAB staining. Slides were counterstained with hematoxylin, dehydrated in graded alcohols, cleared in xylene, and mounted with a permanent mounting medium. Slides were subsequently scanned on an Aperio AT2 system at 0.2531 µm/px and distributed via the Histoserv Digital online platform. The analysis of immunohistochemically stained images was performed using QuPath V.0.5.1 (Belfast, UK). The cell detection parameters were optimized by applying the cell detection across the entire tumor section for each sample analyzed, resulting in the following QuPath parameters: optical density sum detection image, pixel size: 1 µm, intensity threshold 0.1, nucleus parameters with background radius 8 µm, minimum area 10 µm2, maximum area 400 µm2, sigma 1.5.

    Statistical analysis and illustration

    Comparison of data between two groups was performed using Mann-Whitney tests, comparisons involving greater than two groups were performed using a one or two-way analysis of variance test with post hoc Tukey’s multiple comparison correction. Survival differences of tumor-bearing mice were assessed using Kaplan-Meier curves and analyzed by log-rank tests. Results are expressed as mean and SEM. All statistical tests were calculated using GraphPad Prism (GraphPad software, RRID:SCR_002798). P values of <0.05 were considered statistically significant (*p<0.05; **p<0.005; ***p<0.0005; ****p<0.00001). The row-scaled heatmap was generated using the pheatmap package in R, with columns hierarchically clustered (method “ward.d2”). The illustration was created by web-based BioRender tools (BioRender).

    Results

    Second-generation CPIs synergize with TGFβ blockade to delay tumor growth through CD8+ T cells

    To identify what second-generation CPIs can delay tumor growth, we used the CT26 colon carcinoma (“hot” tumor) model on a BALB/c background. The second-generation CPIs, anti-Tim3, anti-LAG-3, anti-TIGIT, and anti-KLRG1, were tested. None of the CPIs tested alone delayed tumor growth (figure 1A). Previous work by our lab and others6 14 has shown the blockade of all isoforms of TGFβ to synergize with cancer vaccines and other CPIs. We combined each second-generation CPI with TGFβ blockade and found that anti-TGFβ synergized with both anti-LAG-3 and anti-TIGIT to delay tumor growth (figure 1B). To see if there would be further combinatorial effects, we combined anti-LAG-3, anti-TIGIT, and anti-TGFβ; however, we saw no increased protection with the triple combination compared with either double combination (figure 1C,F). To determine the mechanism of action of the efficacious CPI combinations, CD4+ and CD8+ T cells were depleted by antibody administration starting 1 day before CPI administration and continuing until the termination of the study. Depletion of each T cell compartment revealed that the tumor suppressive effect of anti-TGFβ combined with anti-LAG-3 or anti-TIGIT was CD8+T cell-dependent and CD4+T cell-independent (figure 1E,F).

    Figure 1
    Figure 1

    Second-generation CPIs synergize with TGFβ blockade to delay tumor growth through CD8+T cells. (A) Female BALB/c mice were injected subcutaneously (s.c.) with 2.5×105 syngeneic CT26 tumor cells and treated with isotype-matched control IgG, anti-KLRG1, anti-Tim3, anti-TIGIT or anti-LAG-3 antibodies (all 200 µg/dose) or the combination of those with (B, C) anti-TGFβ (100 µg/dose) starting 8 days post-tumor inoculation. Tumor growth was tracked and grouped analysis (n=10 per group) are shown. (D) Survival analysis of mice injected with CT26 tumor cells on day 0 and treated with the indicated CPIs. (E, F) Mice injected with CT26 were administered antibodies against CD4, CD8, or both starting −1 day before the start of the indicated CPI regimen and T cell depletions were continued twice a week until termination of the study. The data are represented as the mean±SEM and represent two independent experiments. P values are shown for entirety of study using two-way ANOVA test with post hoc Tukey’s multiple comparisons; **p<0.01; ****p<0.0001. Survival differences of tumor-bearing mice were assessed using Kaplan-Meier curves and analyzed by log-rank testing. ANOVA, analysis of variance; CPI, checkpoint inhibitor; PBS, phosphate-buffered saline.

    Second-generation CPIs do not reduce TC1 tumor growth, but a vaccine can

    We next asked whether either of these beneficial CPI combinations would be effective in the TC1 tumor (“cold” tumor) model on the C57BL/6 background. Contrary to the case in CT26 tumor-bearing mice, neither the combination of anti-TGFβ with anti-LAG-3 nor the combination with anti-TIGIT delayed TC1 tumor growth (figure 2A). The TC1 tumor expresses HPV16 E6 and E7 oncogenes and allows for the single administration of a vaccine (E7 vaccine) that consists of the E7 (43-77) synthetic long peptide15 combined with alpha-galactosylceramide (αGalCer, a potent NKT cell agonist) and GM-CSF as adjuvants,14 all mixed together with DOTAP in PBS. The E7 vaccine alone reduced tumor growth compared with PBS control and anti-TGFβ with CPI combinations, but adding anti-TGFβ with anti-LAG-3 or anti-TIGIT to the vaccine did not synergize to further delay tumor growth compared with vaccine plus Ctrl IgG (figure 2B,C) nor did it increase the amount of E7-specific (MHC I-loaded E7 tetramer+) CD8+T cells in the circulation or in the TME compared with the vaccine alone control (figure 2E, online supplemental figure 1C, D). The E7 vaccine also increased the prevalence of infiltrating CD8+T cell into TME compared with PBS or anti-TGFβ with anti-LAG-3 (CPIs) as assessed by immunohistochemistry (IHC) (online supplemental figure 1E, F). Tumor-infiltrating CD8+E7 tetramer+T cells were plotted against endpoint tumor volume (figure 2F), revealing a significant negative correlation (r2=0.207, p=0.0006) between prevalence of CD8+E7 tetramer+T cells and increased tumor volume.

    Supplemental material

    Figure 2
    Figure 2

    Second-generation CPIs do not reduce TC1 tumor growth but a vaccine can. C57BL/6 mice were injected s.c. with 1×105 syngeneic TC1 tumor cells on day 0, and 7 days post-tumor inoculation mice were administered a combination of anti-TGFβ and CPIs (A) or the CPIs and anti-TGFβ with a single dose of vaccine (B, C) containing (1 µg of E7, 1 µg of GM-CSF, 250 nmol of α-galactosylceramide, formulated in PBS with 20 µg of DOTAP; vaccine) simultaneously. 25 days after tumor inoculation, mice were euthanized, and tumors were collected and subjected to flow cytometric analysis of infiltrating CD8+and E7 tetramer+T cells. (D) Representative flow cytometry contour plots and (E) individual mouse values. (F) Correlation between tumor infiltrating CD8+E7 tetramer+T cells (y-axis) and endpoint tumor volume (x-axis) between indicated groups. Data are represented as the mean±SEM and represent two independent experiments. *p<0.05; **p<0.01; ***p<0.001; ****p<0.0001 determined using two-way ANOVA over entirety of the study (A–C), one-way ANOVA (E) test with post hoc Tukey’s multiple comparison corrections, or linear regression analysis (F). ANOVA, analysis of variance; CPI, checkpoint inhibitor; GM-CSF, granulocyte-macrophage colony-stimulating factor; PBS, phosphate-buffered saline; s.c., subcutaneously.

    Intrinsic differences in immune cell composition between CT26 and TC1 tumors

    To determine what parts of the baseline tumor landscape could be contributing to the differences in susceptibility to CPIs, CT26 and TC1 tumors were left untreated, and once tumors were of equivalent size, removed and subjected to flow cytometric analysis of immune cell subsets of interest (Gating strategies in online supplemental figures 1B and 2C). The primary immune cell of interest was the CD8+T cell which constituted a significantly higher proportion of the tumor infiltrating cells in the CT26 tumors compared with the TC1 tumors as identified via flow cytometry (figure 3A) and IHC of CD8+cells (online supplemental figure 2A, B), identifying CT26 as a “hot” tumor. The expression of the checkpoint molecules KLRG1, PD-1, TIGIT, Tim-3, CD38, and 4-1BB were all higher on CD8+T cells infiltrating CT26 tumors compared with those of TC1 tumors (figure 3B). An examination of the immunosuppressive Treg cell compartment revealed that TC1 tumors had greater Treg cell infiltration than CT26 tumors, and these Treg cells expressed higher levels of KLRG1 and PD-1, but lower levels of TIGIT, Tim3, and CD38 than Tregs in CT26 tumors (figure 3C,D).

    Figure 3
    Figure 3

    Intrinsic differences in immune cell composition between CT26 and TC1. Female C57BL/6 and BALB/c mice were injected s.c. with 1×105 TC1 cells or 2.5×105 CT26 tumor cells, respectively. Once average tumor volume reached ~250 mm3, tumors were harvested, weighed and subjected to flow cytometry for immune cell markers of interest. Individual numbers of tumor infiltrating CD8+T cells (A); levels of T-cell expression of checkpoint molecules (MFI) (B); CD4+FoxP3+ Treg cells (C); MDSC subsets: M-MDSC (Ly-6CHILy-6G-) and Gr-MDSC (Ly-6C+Ly-6G+) (E); dendritic cell subsets: cDC (B220-CD11c+) and pDC (B220+CD11c+) (G); cDC subsets: cDC1 (CD11b-CD8+) and cDC2 (CD11b+CD8−) (H); and macrophages (CD11b+F4/80+) (J). MFI of the indicated markers, on the indicated cell type (B, D, F, I, and K). Data from (A–K) are represented as the mean±SEM and represent 1–3 independent experiments with each symbol representing an individual mouse, n=6–22. *p<0.05; **p<0.01; ***p<0.001; ****p<0.0001 determined using a Mann-Whitney U test. (L) Heatmap was generated summarizing flow cytometry data from one representative experiment (n=6–7). Each column corresponds to an individual mouse, and relative cell population frequency is row-scaled for ease of comparison between TC1 and CT26 tumors. P value was determined using Mann-Whitney test (red/dark gray/light gray/white color bar at left). MFI, median fluorescence intensity; MDSC, myeloid-derived suppressor cell; pDC, plasmacytoid DC; s.c., subcutaneously.

    We then evaluated the myeloid cells within these tumors and found that TC1 tumors are infiltrated by a higher proportion of Ly-6CHI Ly-6GNEG monocytic myeloid-derived suppressor cells (MDSCs) (M-MDSCs) with higher CD155 expression, a ligand for TIGIT, than M-MDSCs in CT26 tumors, but a lower proportion of Ly-6C+Ly-6G+granulocytic MDSCs (Gr-MDSCs) that also express lower comparative levels of CD155 and PD-L1 compared with Gr-MDSCs in CT26 tumors (figure 3E,F). Moreover, TC1 tumors exhibited higher levels of B220–CD11c+classical dendritic cells (cDCs) and a lower proportion of B220+CD11c+plasmacytoid DCs (pDCs) compared with CT26 tumors (figure 3G). Among the cDC compartment, TC1 tumors showed a higher proportion of cDC2 (CD11b+CD8–) and lower cDC1 (CD11b–CD8+) cells compared with CT26 tumors (figure 3H). cDCs in TC1 tumor-bearing mice show higher expression of CD155 than their counterparts in CT26 tumor-bearing mice and the pDCs in TC1 tumors have higher PD-L1 expression (figure 3I). CT26 tumors have a greater proportion of F4/80+macrophages than TC1 tumors, but they express less CD155 and more PD-L1, than TC1 tumors (figure 3J,K). Flow cytometric data from a representative experiment were conglomerated to summarize the differences between the TC1 and CT26 TME (figure 3L), supporting an increase in CD155 expressing macrophages, M-MDSCs and cDCs, and an increase in Tregs in TC1 tumor-bearing mice compared with CT26 tumor-bearing mice. Conversely, in CT26 tumor-bearing mice there is higher MHC II, PD-L1, and XCR1 on DCs which is associated with higher infiltration of CD8+T cells expressing higher levels of checkpoint molecules—Tim-3, KLRG-1, TIGIT, and CD38.

    Anti-TIGIT and anti-PD-L1 synergize with the E7 vaccine through CD8+ T cells with a role for CD4+ T cells

    Due to the increased CD155 expressed on cDCs and M-MDSCs and tumor infiltrating M-MDSCs in TC1 tumors, we continued using anti-TIGIT and combined it with anti-PD-L1, as has been shown to be an effective CD8+T cell-dependent combination.16 The combination of anti-TIGIT and anti-PD-L1 synergized with the E7 vaccine to delay tumor growth and increase survival to a better degree than any two of the three alone (figure 4A,B). Depleting CD4+and CD8+ T cells revealed that the beneficial effect of the triple combination is dependent on CD8+T cells; however, CD4+T cell depletion unexpectedly further enhanced, rather than diminishing, the effect of the triple combination, with the majority of the mice completely clearing the tumor (figure 4C). This observation suggested that Treg cells, which we had found to be at high levels in TC1 tumors, may be another critical regulatory component in the TC1 tumor. 30 days after tumor inoculation, spleens were harvested from the mice receiving the triple combination with isotype-matched control IgG or PBS-treated mice, and spleen cells were restimulated ex vivo with E7 (43–77) long peptide. Mice treated with E7 vaccine plus anti-TIGIT and anti-PD-L1 (Vax+CPIs) showed a durable polyfunctional IFNγ+TNFα+ CD8+ T cell response when stimulated with E7 (43–77) long peptide that was not present in PBS-treated or vehicle controls (figure 4D,E).

    Figure 4
    Figure 4

    Anti-TIGIT and anti-PD-L1 synergize with the E7 vaccine through CD8+T cells with an inhibitory role for CD4+T cells. C57BL/6 mice were injected s.c. with 1×105 syngeneic TC1 tumor cells on day 0, and 7 days post-tumor inoculation mice were administered either a vaccine, CPIs or a combination. Tumor volume (A) and percent survival (B) of the indicated groups. Mice were injected with TC1 and 1 day before start of the Vaccine plus CPI regimen, CD4+, CD8+ or both T cell subsets were depleted by intraperitoneal injection of anti-CD4 and/or anti-CD8 that continued every 2–3 days until termination of the study (C). Day 21 post-tumor inoculation, spleens were harvested, brought to a single cell suspension and stimulated ex vivo with E7 peptide (10 µg/mL) or vehicle control (PBS) in the presence of brefeldin A for 6 hours then cells were collected and subjected to flow cytometry for expression of IFNγ+TNFα+ by polyfunctional CD8+T cells. Representative flow cytometry plot (D) and graphed individual values (E) of IFNγ+TNFα+ expressing CD8+ T cells. The data are represented as the mean±SEM and represent at least two independent experiments, n=4–10. *p<0.05; **p<0.01; ***p<0.001; ****p<0.0001 determined using two-way ANOVA test with post hoc Tukey’s multiple comparison correction. Survival differences of tumor-bearing mice were assessed using Kaplan-Meier curves and analyzed by log-rank testing. ANOVA, analysis of variance; CPI, checkpoint inhibitor; s.c., subcutaneously; PBS, phosphate-buffered saline.

    Triple combination therapy enhanced by Treg cell depletion

    To determine if FoxP3+Treg cells were the CD4+T cell type whose depletion is responsible for the enhanced antitumor activity observed with the triple combination in CD4+T cell-depleted mice (figure 4C), we selectively depleted FoxP3+cells. We used FoxP3-GFPDTR (B6.129(Cg)-Foxp3tm3(Hbegf/GFP)Ayr/J) mice expressing the diphtheria-toxin receptor (DTR) exclusively on FoxP3+cells so that FoxP3-expressing cells were selectively depleted upon administration of diphtheria toxin (DT). Treg cells were transiently depleted by injecting DT (10 µg/kg) every 2–3 days, starting 3 days before the triple combination therapy began (figure 5A; online supplemental figure 2D, E). The mice receiving E7 vaccine plus anti-TIGIT and anti-PD-L1 (CPIs) and DT had tumor volume comparable to vaccine plus CPIs with anti-CD4, both of which were better than vaccine plus CPIs alone (figure 5B,C). Note that this marked improvement in response after Treg cell depletion with either anti-CD4 or DT in the TC1 model is in sharp contrast to the lack of any effect of anti-CD4 in the CT26 model (figure 1E,F) in which Treg cells were not prominent in the TME, another indication that characterization of the TME within each tumor can be predictive of the best combination immunotherapy for that tumor.

    Figure 5
    Figure 5

    Triple combination therapy enhanced by Treg cell depletion. (A) Experimental scheme: (B, C) FoxP3-GFPDTR mice were injected s.c. with 1×105 TC1 tumor cells on day 0 and then DT administration (10 µg/kg, orange and red) or anti-CD4 (200 µg first dose, 100 µg subsequent doses, green and blue) was given to mice 3 days before treatment with vaccine plus CPIs began (purple, red, and blue). Average tumor growth (B) and individual tumor growth (C) of each tumor-bearing mouse. The data are represented as the mean±SEM and represent two independent experiments, n=7–8. *p<0.05; ****p<0.0001 determined using two-way ANOVA test with post hoc Tukey’s multiple comparison correction. ANOVA, analysis of variance; CPI, checkpoint inhibitor; s.c., subcutaneously; PBS, phosphate-buffered saline.

    CPIs improve the intratumoral E7-specific CD8+ T cell response

    Since the effect of E7 vaccine plus CPIs on tumor growth was CD8+T cell-dependent, we next asked what changes occurred on those cells between the mice receiving vaccine alone or vaccine plus CPIs. The vaccine alone increases the percent of intratumoral E7-specific CD8+T cells, and those cells have increased PD-1 and TIGIT expression (online supplemental figure 3A, B). The vaccine also increases intratumoral expression of PD-L1 on DCs, but CD155 expression is reduced compared with PBS-treated controls, with no change in the percentage of Treg cells or PD-1 or TIGIT expression on the Treg cells (online supplemental figure 3C, D). At the time point when tumor growth begins to diverge between vaccine alone and vaccine plus CPIs (figure 6A), tumors, spleens, and tumor draining lymph nodes (tdLNs) were harvested and analyzed. The tumors of vaccine alone and vaccine plus CPIs-treated mice had an equivalent number of E7-specific CD8+T cells (figure 6B,C); however, the mice receiving vaccine plus CPIs had lower coexpression of exhaustion markers, PD-1 and CD38, on their E7-specific CD8+T cells than those receiving vaccine alone (figure 6D,E). More E7-specific CD8+T cells from vaccine-treated mice expressed IFNγ and TNFα in the spleen and tdLN than those of control mice, whereas, in the tumor, only the vaccine plus CPIs-treated mice had significantly more IFNγ+TNFα+ E7-specific CD8+T cells than those in the PBS control (figure 6F,G).

    Figure 6
    Figure 6

    CPIs improve the intratumoral E7-specific CD8 T cell response. Female C57BL/6 mice were injected s.c. with 1×105 TC1 tumor cells on day 0, and on day 9, once average tumor volume reached ~50 mm3, mice were given either vaccine alone or vaccine plus CPIs. Average tumor volume for each group (A) is shown. On day 21, spleens, tumor draining lymph nodes (tdLNs), and tumors were harvested and subjected to flow cytometry for cells of interest. Representative flow cytometry plots (B) and graphed individual values show CD8+E7 tetramer+T cells (C), and the proportion of those cells that are PD-1+CD38+ (D, E) and IFNγ+TNFα+ (F, G). The data are represented as the mean±SEM and represent one independent experiment with each symbol representing an individual mouse, n=7–10. *p<0.05; **p<0.01; ***p<0.001; ****p<0.0001 determined using two-way ANOVA test with post hoc Tukey’s multiple comparison corrections. ANOVA, analysis of variance; CPI, checkpoint inhibitor; ns, not significant; s.c., subcutaneously; PBS, phosphate-buffered saline.

    Optimization of vaccine plus CPIs sequence in young and aged mice

    There have been reports that the timing and sequence of CPIs relative to vaccines affect the success of cancer vaccines.17 18 To test the importance of timing of the CPI administration on the success of our vaccine plus CPIs treatment model, we began CPI administration either concurrently with the vaccine (CPI-con) or at one CPI interval before (CPI-pre) or after (CPI-post) vaccination. Only vaccine plus CPIs given concurrently was able to significantly reduce tumor growth compared with controls (figure 7A). Both vaccine plus CPI-con and vaccine plus CPI-post were able to significantly improve the survival of TC1 tumor-bearing mice, while the vaccine plus CPI-pre group lost all protective effects (figure 7A,B). CPI-con and CPI-post resulted in greater induction of E7-specific CD8+T cells in the circulation compared with the PBS and CPI-pre groups (figure 7C,D). It is known that aged animals are less responsive to vaccines so we tested our triple combination in aged mice (>18 months), revealing that, contrary to in young mice, the vaccine alone was unable to reduce tumor volume or tumor weight compared with PBS controls, but combining the vaccine with CPIs was able to delay tumor growth and reduce tumor weight (figure 7E,F). The group of aged animals receiving vaccine plus CPIs had a non-statistically significant increase in E7-specific CD8+T cells compared with PBS controls (figure 7G) and significantly higher levels of IFNγ+ E7-specific CD8+T cells than either PBS controls or the group receiving vaccine alone (figure 7H and online supplemental figure 4A). Since the E7 vaccine alone was ineffective at delaying tumor growth in aged mice, we tested whether the combinations of anti-TGFβ with anti-LAG-3 or anti-TIGIT, that delayed CT26 tumor growth in a CD8+T cell dependent manner in BALB/c mice, would continue to work in aged mice. Neither combination of anti-TGFβ with anti-LAG-3 or anti-TIGIT was able to reduce tumor growth or improve survival in aged BALB/c mice (online supplemental figure 4B, C). We then chose to test the effect of CD4+T cell depletion on the vaccine plus CPIs regimen in aged mice. Comparing young and aged mice inoculated with TC1 tumors revealed that TC1 tumors grow at the same rate in young and aged mice; and young and old PBS groups did not have signficicantly different tumor weights at the end of the experiment. Young mice administered vaccine plus CPIs with anti-CD4 exhibited the previously observed reduction in tumor volume post-treatment that eventually leads to regression; aged mice that received this regimen were able to delay tumor growth to an extent similar to vaccine plus CPIs without anti-CD4, but neither treatment significantly reduced tumor weight (figure 7I,J). Young mice given the vaccine plus CPIs with anti-CD4 have significantly reduced PD-1 and CD38 coexpression on tumor-infiltrating CD8+E7 tetramer+T cells compared with vaccine plus CPIs and PBS groups, but the aged mice receiving vaccine plus CPIs with anti-CD4 do not exhibit this effect (figure 7K and online supplemental figure 4D).

    Figure 7
    Figure 7

    Optimization of vaccine plus CPI sequence in young and aged mice. C57BL/6 mice were injected s.c. with 1×105 TC1 tumor cells on day 0 and then CPI administration (anti-TIGIT+anti-PD-L1, 200 µg/dose) began 3 days before vaccination (Pre, orange), concurrently with vaccination (Concurrent (Con), purple), or 4 days postvaccination (Post, green). Average tumor growth (A) and survival (B) of each tumor-bearing mouse. Blood was collected 21 days post- tumor inoculation and subjected to flow cytometry for CD8+CD44+E7 tetramer+T cells in circulation. Representative flow cytometry plots (C) and individual mice (D) are shown. Aged (>18 months) female C57BL/6 mice were injected s.c. with 1×105 TC1 tumor cells on day 0 and received vaccine±CPIs starting on day 7. (E) Average tumor volumes and (F) tumor weights are shown. On day 20, tumors were harvested and subjected to flow cytometry for cells of interest. Graphed individual values of CD8+E7 tetramer+T cells (G), and the proportion of those cells that are IFNγ+ (H). Young (2–3 months) and aged (>18 months) female C57BL/6 mice were injected s.c. with 1×105 TC1 tumor cells on day 0 and received vaccine+CPIs (purple) starting on day 7 or with anti-CD4 (200 µg first dose, 100 µg subsequent doses, blue) starting on day 4. Average tumor growth (I) and tumor weight (J) of each tumor-bearing mouse. Tumors were harvested and subjected to flow cytometry for cells of interest. Graphed individual values of percentage of PD-1+CD38+ cells of all CD8+E7 tetramer+cells (K). The data are represented as the mean±SEM and represent two independent experiments with each symbol representing an individual mouse, n=7–10 (A–D), n=5 (E–H), n=4–10 (I, J). *p<0.05; **p<0.01; ***p<0.001; ****p<0.0001 determined using two-way ANOVA test with post hoc Tukey’s multiple comparison corrections. ANOVA, analysis of variance; CPI, checkpoint inhibitor; PBS, phosphate-buffered saline; s.c., subcutaneously.

    Discussion

    Checkpoint molecule blockade revolutionized the cancer field with the introduction of antibodies targeting PD-1, PD-L1, and CTLA-4 that can singly elicit an antitumor response.19 20 The array of second-generation CPIs, while more tolerable,21 22 are less efficacious, but can be combined with first-generation CPIs to greater effect.23 24 The infancy of this technology mandates further combinatorial experimentation with second generation CPIs to accomplish the next breakthrough in immunotherapy. We show that while the second-generation CPIs tested (anti-Tim3, anti-LAG-3, anti-TIGIT, and anti-KLRG1) are ineffective at delaying the growth of the immunogenic CT26 tumor as single agents, combining anti-LAG-3 or anti-TIGIT with pan blockade of all isoforms of TGFβ is protective in a CD8+T cell-dependent manner. Early indications of anti-LAG-3 and anti-TIGIT efficacy were relayed expectedly to the clinic; and anti-TGFβ preparations are under evaluation.25–27 In contrast, neither combination of anti-LAG-3 or anti-TIGIT with anti-TGFβ could dent the growth of TC1 tumors.

    TC1 tumors are significantly less immunogenic than CT26 tumors as judged by CD8+T cell infiltration (ie, “colder”), and are duly less responsive to CPIs, requiring de novo antitumor CD8+T cell activity. Although studies have determined that a fraction of TCF1+tumor antigen-specific CD8+T cells can respond to CPI therapy,28 most cells that respond to CPIs are replacing exhausted cells present in the tumor,29 concordant with data suggesting the epigenetic fate of PD-1HI intratumoral T cells has rendered them inflexible.30 For the TC1 tumor, the existing CD8+T cells are inadequate, requiring measures such as cancer vaccines to elicit an antitumor response. We show that a tumor-antigen-specific vaccine synergizes with anti-TIGIT and anti-PD-L1, but not anti-TGFβ with either anti-TIGIT or anti-LAG-3. This could be due to increased CD155 expression on selected members of the myeloid compartment of cells (macrophages, cDCs, and M-MDSCs) infiltrating TC1 tumor compared with CT26 tumor; however, TIGIT coactivation with PD-L1 has been shown to inhibit T cell activation through either direct inhibition of T cells, inducing CD155 on APCs and tumor cells, or by inhibiting the proactivation of CD226 receptor also found on CD8+T cells and NK cells with lower binding affinity for CD155 than TIGIT.16 Indeed, the combination of TIGIT and PD-L1 blockade has shown promise in a recent Phase II trial for NSCLC,23 leading to over 130 clinical trials investigating some aspect of TIGIT inhibition.31 Our data also revealed Treg cells to be a significant component of the protumor response, such that Treg cell depletion in our TC1 model with the triple combination provided complete protection. Future studies will need to be done to further investigate how to exploit this dynamic.

    While the field of CPIs is bourgeoning, its dependence on a pre-existing immune response in the case of some CPIs has renewed interest in cancer vaccines. Long dormant, there is currently only one approved cancer vaccine, Sipuleucel-T, for prostate cancer. But the advent of neoantigen vaccines,32 and the improvement in vaccine platforms heralds a reinvigoration in this field, especially when combining cancer vaccines with CPIs.33 Cancer vaccines have been shown in many studies to stimulate a potent antitumor T cell response34; however, further work needs to be done to successfully translate the findings into a clinical success and to figure out the best strategy for combining CPIs with cancer vaccines. For example, it has been shown that using CPI therapy before a stable population of optimally primed T cells is present quenches the antitumor response,17 and in a small Phase I/II trial, melanoma patients who were anti-PD-1 Ab naïve had better outcomes than patients experienced with anti-PD-1 Abs.35 Similarly, post-vaccine administration of CPIs had reduced effects compared with concurrent administration in a clinical trial of a DNA vaccine in patients with metastatic, castration-resistant prostate cancer18; however, this study administered 12 vaccinations weekly before CPI therapy began, so anergy may have been a contributing factor inhibiting the response to CPIs. Our work, testing all possible combinations of prevaccination, concurrent vaccination, or postvaccination timings of CPI, revealed concurrent administration of vaccine plus CPIs to provide the best survival benefit and highest number of effector tumor-specific T cells, although postvaccination administration of CPIs proved to be an effective strategy as well. Prevaccination treatment with CPIs, on the other hand, abolished the effect of the triple combination, further suggesting that for future trials investigating the effects of cancer vaccines with CPIs, patients may be more responsive to cancer vaccines if they are CPI therapy-naïve.

    Our treatment paradigm was then tested in aged mice, who have exhibited a lower ability to respond to vaccinations.36 Accordingly, we showed that in our hands the E7 vaccine alone cannot delay tumor growth in aged mice, as it did in young mice; however, addition of CPIs to the vaccine was able to augment the vaccine efficacy to delay tumor growth, decrease tumor weight, and increase the prevalence of tumor-infiltrating E7-specific IFNγ+ CD8+ T cells compared with PBS and vaccine alone controls. The response of aged animals to CPIs is yet to be determined conclusively,37 but our work suggests that CPIs can rescue vaccine efficacy in aged mice, although they may be less effective when used alone as was shown when anti-TGFβ with either anti-LAG-3 or anti-TIGIT could not delay CT26 growth in aged mice, as they did in young mice. Since both our vaccine results in TC1, and CPI combination results in CT26, were CD8+T cell-dependent, it could be the well-documented increase in CD8+T cell exhaustion and decrease in IFNγ production,38 coupled with loss of naïve cell priming efficacy in aged adults39 that leads to these changes. Our results that CD4+T cell depletion was unable to increase the potency of our vaccine plus CPIs combination in aged mice, as it did in young mice, is likely due to some aspect of Treg cell dysfunction that occurs with age. In a mouse model of melanoma, Treg cell depletion induced antitumor immunity in young mice but was ineffective in aged animals.40 Disproportionately, preclinical and clinical studies concern young mice (~2 months) or young patients where their aged counterparts are ineligible, and yet the majority of cancers arise in individuals over the age of 60, at which point cancer becomes the leading cause of death.41 Thus the complexitites and differences of the aged TME necessitates further study.

    Our findings have multiple clinical implications. We found two combinations of anti-TGFβ with either anti-LAG-3 or anti-TIGIT to be effective CD8+T cell-dependent therapies that can be translated to cases of immunogenic tumors, although these treatments become ineffective in aged mice. We showed an E7 tumor-antigen-specific vaccine in the TC1 model can elicit a CD8+T cell response that can be further amplified by CPIs, anti-TIGIT, and anti-PD-L1, that are alone ineffective to delay tumor growth, and that this combination can be further potentiated to completely clear tumors with the depletion of Treg cells, an advance ripe for exploitation. Further, our characterization of each tumor type and TME allowed us to successfully predict the most effective combination of immunotherapy. This strategy, which we are continuing to develop by comparing multiple tumors, can be utilized to select the most effective immunotherapy for each patient in a personalized, precision medicine approach. Finally, we showed that the optimal timing of CPI administration can either enhance the synergy with a cancer vaccine or abolish it altogether. Using this knowledge to delay tumor growth in aged animals resistant to vaccination will pave the road for translating these therapies to those most in need.

    Supplemental material

    Data availability statement

    All data relevant to the study are included in the article or uploaded as online supplemental information. Values for all data points in graphs are reported in online supplemental file 2.

    Ethics statements

    Patient consent for publication

    Ethics approval

    All animal procedures reported in this study that were performed by NCI-CCR affiliated staff were approved by the NCI Animal Care and Use Committee (ACUC) and in accordance with federal regulatory requirements and standards (reference number: METB-033). All parts of the intramural NIH ACU program are accredited by AAALAC International.

    Acknowledgments

    We kindly thank the NCI CCR Vaccine Branch Flow Cytometry Core staff for their expertise in operating the flow cytometer and the NCI Animal Facility staff for their assistance with the breeding and maintaining of our mice used in this study.

    References

    1. ↵
      1. Haslam A,
      2. Prasad V
      . Estimation of the percentage of US patients with cancer who are eligible for and respond to checkpoint inhibitor immunotherapy drugs. JAMA Netw Open 2019;2:e192535. doi:10.1001/jamanetworkopen.2019.2535
    2. ↵
      1. Zhao B,
      2. Zhao H,
      3. Zhao J
      . Efficacy of PD-1/PD-L1 blockade monotherapy in clinical trials. Ther Adv Med Oncol 2020;12. doi:10.1177/1758835920937612
    3. ↵
      1. Boutros C,
      2. Tarhini A,
      3. Routier E, et al
      . Safety profiles of anti-CTLA-4 and anti-PD-1 antibodies alone and in combination. Nat Rev Clin Oncol 2016;13:473–86. doi:10.1038/nrclinonc.2016.58
      OpenUrlCrossRefPubMed
    4. ↵
      1. Anderson AC,
      2. Joller N,
      3. Kuchroo VK
      . Tim-3, and TIGIT: co-inhibitory receptors with specialized functions in immune regulation. Immunity 2016;44:989–1004. doi:10.1016/j.immuni.2016.05.001
      OpenUrlCrossRefPubMed
    5. ↵
      1. Xu W,
      2. Dong J,
      3. Zheng Y, et al
      . Immune-checkpoint protein VISTA regulates antitumor immunity by controlling myeloid cell-mediated inflammation and immunosuppression. Cancer Immunol Res 2019;7:1497–510. doi:10.1158/2326-6066.CIR-18-0489
      OpenUrlAbstract/FREE Full Text
    6. ↵
      1. Ali S,
      2. Rehman MU,
      3. Yatoo AM, et al
      . TGF-Β signaling pathway: therapeutic targeting and potential for anti-cancer immunity. Eur J Pharmacol 2023;947:175678. doi:10.1016/j.ejphar.2023.175678
      OpenUrl
    7. ↵
      1. Chowell D,
      2. Yoo SK,
      3. Valero C, et al
      . Improved prediction of immune checkpoint blockade efficacy across multiple cancer types. Nat Biotechnol 2022;40:499–506. doi:10.1038/s41587-021-01070-8
      OpenUrlPubMed
    8. ↵
      1. Gros A,
      2. Robbins PF,
      3. Yao X, et al
      . PD-1 identifies the patient-specific CD8+ tumor-reactive repertoire infiltrating human tumors. J Clin Invest 2014;124:2246–59. doi:10.1172/JCI73639
      OpenUrlCrossRefPubMedWeb of Science
    9. ↵
      1. Salmon H,
      2. Idoyaga J,
      3. Rahman A, et al
      . Expansion and activation of CD103(+) dendritic cell progenitors at the tumor site enhances tumor responses to therapeutic PD-L1 and BRAF inhibition. Immunity 2016;44:924–38. doi:10.1016/j.immuni.2016.03.012
      OpenUrlCrossRefPubMed
    10. ↵
      1. Melief CJM,
      2. van Hall T,
      3. Arens R, et al
      . Therapeutic cancer vaccines. J Clin Invest 2015;125:3401–12. doi:10.1172/JCI80009
      OpenUrlCrossRefPubMed
    11. ↵
      1. Maeng H,
      2. Terabe M,
      3. Berzofsky JA
      . Cancer vaccines: translation from mice to human clinical trials. Curr Opin Immunol 2018;51:111–22. doi:10.1016/j.coi.2018.03.001
      OpenUrl
    12. ↵
      1. Maeng HM,
      2. Moore BN,
      3. Bagheri H, et al
      . Phase I clinical trial of an autologous dendritic cell vaccine against HER2 shows safety and preliminary clinical efficacy. Front Oncol 2021;11:789078. doi:10.3389/fonc.2021.789078
    13. ↵
      1. Lin MJ,
      2. Svensson-Arvelund J,
      3. Lubitz GS, et al
      . Cancer vaccines: the next immunotherapy frontier. Nat Cancer 2022;3:911–26. doi:10.1038/s43018-022-00418-6
      OpenUrl
    14. ↵
      1. Terabe M,
      2. Robertson FC,
      3. Clark K, et al
      . Blockade of only TGF-Β 1 and 2 is sufficient to enhance the efficacy of vaccine and PD-1 checkpoint blockade immunotherapy. Oncoimmunology 2017;6:e1308616. doi:10.1080/2162402X.2017.1308616
    15. ↵
      1. van Poelgeest MIE,
      2. Welters MJP,
      3. van Esch EMG, et al
      . HPV16 synthetic long peptide (HPV16-SLP) vaccination therapy of patients with advanced or recurrent HPV16-induced gynecological carcinoma, a phase II trial. J Transl Med 2013;11:88. doi:10.1186/1479-5876-11-88
    16. ↵
      1. Banta KL,
      2. Xu X,
      3. Chitre AS, et al
      . Mechanistic convergence of the TIGIT and PD-1 inhibitory pathways necessitates co-blockade to optimize anti-tumor CD8+ T cell responses. Immunity 2022;55:512–26. doi:10.1016/j.immuni.2022.02.005
      OpenUrlCrossRefPubMed
    17. ↵
      1. Verma V,
      2. Shrimali RK,
      3. Ahmad S, et al
      . PD-1 blockade in subprimed CD8 cells induces dysfunctional PD-1+CD38Hi cells and anti-PD-1 resistance. Nat Immunol 2019;20:1231–43. doi:10.1038/s41590-019-0441-y
      OpenUrlCrossRefPubMed
    18. ↵
      1. McNeel DG,
      2. Eickhoff JC,
      3. Wargowski E, et al
      . Concurrent, but not sequential, PD-1 blockade with a DNA vaccine elicits anti-tumor responses in patients with metastatic, castration-resistant prostate cancer. Oncotarget 2018;9:25586–96. doi:10.18632/oncotarget.25387
      OpenUrlPubMed
    19. ↵
      1. Sharma P,
      2. Allison JP
      . Immune checkpoint targeting in cancer therapy: toward combination strategies with curative potential. Cell 2015;161:205–14. doi:10.1016/j.cell.2015.03.030
      OpenUrlCrossRefPubMed
    20. ↵
      1. Ribas A,
      2. Wolchok JD
      . Cancer immunotherapy using checkpoint blockade. Science 2018;359:1350–5. doi:10.1126/science.aar4060
      OpenUrlAbstract/FREE Full Text
    21. ↵
      1. Maruhashi T,
      2. Sugiura D,
      3. Okazaki IM, et al
      . LAG-3: from molecular functions to clinical applications. J Immunother Cancer 2020;8:e001014. doi:10.1136/jitc-2020-001014
    22. ↵
      1. Niu J,
      2. Maurice-Dror C,
      3. Lee DH, et al
      . First-in-human phase 1 study of the anti-TIGIT antibody vibostolimab as monotherapy or with pembrolizumab for advanced solid tumors, including non-small-cell lung cancer☆. Ann Oncol 2022;33:169–80. doi:10.1016/j.annonc.2021.11.002
      OpenUrlCrossRef
    23. ↵
      1. Cho BC,
      2. Abreu DR,
      3. Hussein M, et al
      . Tiragolumab plus atezolizumab versus placebo plus atezolizumab as a first-line treatment for PD-L1-selected non-small-cell lung cancer (CITYSCAPE): primary and follow-up analyses of a randomised, double-blind, phase 2 study. Lancet Oncol 2022;23:781–92. doi:10.1016/S1470-2045(22)00226-1
      OpenUrlCrossRefPubMed
    24. ↵
      1. Kraman M,
      2. Faroudi M,
      3. Allen NL, et al
      . FS118, a bispecific antibody targeting LAG-3 and PD-L1, enhances T-cell activation resulting in potent antitumor activity. Clin Cancer Res 2020;26:3333–44. doi:10.1158/1078-0432.CCR-19-3548
      OpenUrlAbstract/FREE Full Text
    25. ↵
      1. Morris JC,
      2. Tan AR,
      3. Olencki TE, et al
      . Phase I study of GC1008 (Fresolimumab): a human anti-transforming growth factor-beta (TGFβ) monoclonal antibody in patients with advanced malignant melanoma or renal cell carcinoma. PLoS One 2014;9:e90353. doi:10.1371/journal.pone.0090353
    26. ↵
      1. de Gramont A,
      2. Faivre S,
      3. Raymond E
      . Novel TGF-Β inhibitors ready for prime time in onco-immunology. Oncoimmunology 2017;6:e1257453. doi:10.1080/2162402X.2016.1257453
    27. ↵
      1. Huang C-Y,
      2. Chung C-L,
      3. Hu T-H, et al
      . Recent progress in TGF-Β inhibitors for cancer therapy. Biomed Pharmacother 2021;134:111046. doi:10.1016/j.biopha.2020.111046
      OpenUrlCrossRefPubMed
    28. ↵
      1. Miller BC,
      2. Sen DR,
      3. Al Abosy R, et al
      . Subsets of exhausted CD8+ T cells differentially mediate tumor control and respond to checkpoint blockade. Nat Immunol 2019;20:326–36. doi:10.1038/s41590-019-0312-6
      OpenUrlCrossRefPubMed
    29. ↵
      1. Yost KE,
      2. Satpathy AT,
      3. Wells DK, et al
      . Clonal replacement of tumor-specific T cells following PD-1 blockade. Nat Med 2019;25:1251–9. doi:10.1038/s41591-019-0522-3
      OpenUrlCrossRefPubMed
    30. ↵
      1. Philip M,
      2. Fairchild L,
      3. Sun L, et al
      . Chromatin states define tumour-specific T cell dysfunction and reprogramming. Nature 2017;545:452–6. doi:10.1038/nature22367
      OpenUrlCrossRefPubMed
    31. ↵
      1. Chu X,
      2. Tian W,
      3. Wang Z, et al
      . Co-inhibition of TIGIT and PD-1/PD-L1 in cancer Immunotherapy: mechanisms and clinical trials. Mol Cancer 2023;22:93. doi:10.1186/s12943-023-01800-3
    32. ↵
      1. Ott PA,
      2. Hu Z,
      3. Keskin DB, et al
      . An immunogenic personal neoantigen vaccine for patients with melanoma. Nature 2017;547:217–21. doi:10.1038/nature22991
      OpenUrlCrossRefPubMed
    33. ↵
      1. Sahin U,
      2. Oehm P,
      3. Derhovanessian E, et al
      . An RNA vaccine drives immunity in checkpoint-inhibitor-treated Melanoma. Nature 2020;585:107–12. doi:10.1038/s41586-020-2537-9
      OpenUrlCrossRefPubMed
    34. ↵
      1. Singh J,
      2. Bowne WB,
      3. Snook AE
      . Cancer vaccines and immunotherapy for tumor prevention and treatment. Vaccines (Basel) 2021;9:1298. doi:10.3390/vaccines9111298
    35. ↵
      1. Vavolizza RD,
      2. Petroni GR,
      3. Mauldin IS, et al
      . Phase I/II clinical trial of a helper peptide vaccine plus PD-1 blockade in PD-1 antibody-naïve and PD-1 antibody-experienced patients with Melanoma (MEL64). J Immunother Cancer 2022;10:e005424. doi:10.1136/jitc-2022-005424
    36. ↵
      1. Lord JM
      . The effect of ageing of the immune system on vaccination responses. Hum Vaccin Immunother 2013;9:1364–7. doi:10.4161/hv.24696
      OpenUrlCrossRefPubMed
    37. ↵
      1. Wong SK,
      2. Nebhan CA,
      3. Johnson DB
      . Impact of patient age on clinical efficacy and toxicity of checkpoint inhibitor therapy. Front Immunol 2021;12:786046. doi:10.3389/fimmu.2021.786046
    38. ↵
      1. Sceneay J,
      2. Goreczny GJ,
      3. Wilson K, et al
      . Interferon signaling is diminished with age and is associated with immune checkpoint blockade efficacy in triple-negative breast cancer. Cancer Discov 2019;9:1208–27. doi:10.1158/2159-8290.CD-18-1454
      OpenUrlAbstract/FREE Full Text
    39. ↵
      1. Briceño O,
      2. Lissina A,
      3. Wanke K, et al
      . Reduced naïve CD8(+) T-cell priming efficacy in elderly adults. Aging Cell 2016;15:14–21. doi:10.1111/acel.12384
      OpenUrlCrossRefPubMed
    40. ↵
      1. Hurez V,
      2. Daniel BJ,
      3. Sun L, et al
      . Mitigating age-related immune dysfunction heightens the efficacy of tumor Immunotherapy in aged mice. Cancer Res 2012;72:2089–99. doi:10.1158/0008-5472.CAN-11-3019
      OpenUrlAbstract/FREE Full Text
    41. ↵
      1. Fane M,
      2. Weeraratna AT
      . How the ageing microenvironment influences tumour progression. Nat Rev Cancer 2020;20:89–106. doi:10.1038/s41568-019-0222-9
      OpenUrlCrossRefPubMed

    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

    • X @NCICCR_VB

    • Contributors WJB, JAB and PO designed the study. JAB is guarantor of the work. WJB, NS and PAM collected and analyzed the data. WJB, JAB and PO interpreted the data and discussed the conclusions. KCG provided bioinformatics support and generated the heatmap. HM provided consultations on clinical relevance and the sequence of therapies. WJB wrote the manuscript, which all authors critiqued.

    • Funding This study was supported by the NCI, NIH Intramural Research Program (NIA-SC-004020).

    • Competing interests None declared.

    • 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.