Dual PD-1 and VEGFR-2 blockade promotes vascular normalization and enhances anti-tumor immune responses in HCC

Abstract

Activation of anti-tumor immune response using programmed death receptor-1 (PD-1) blockade showed benefit only in a fraction of hepatocellular carcinoma (HCC) patients. Combining PD-1 blockade with antiangiogenesis has shown promise in substantially increasing the fraction of HCC patients who respond to treatment, but the mechanism of this interaction is unknown. We recapitulated these clinical outcomes using orthotopic—grafted or induced—murine models of HCC. Specific blockade of vascular endothelial receptor 2 (VEGFR-2) using a murine antibody significantly delayed primary tumor growth but failed to prolong survival, while anti-PD-1 antibody treatment alone conferred a minor survival advantage in one model. However, dual anti-PD-1/VEGFR-2 therapy significantly inhibited primary tumor growth and doubled survival in both models. Combination therapy reprogrammed the immune microenvironment by increasing CD8⁺ cytotoxic T cell infiltration and activation, shifting the M1/M2 ratio of tumor-associated macrophages and reducing T regulatory cell (Treg) and CCR2+ monocyte infiltration in HCC tissue. In these models, VEGFR-2 was selectively expressed in tumor endothelial cells. Using spheroid cultures of HCC tissue, we found that PD-L1 expression in HCC cells was induced in a paracrine manner upon anti-VEGFR-2 blockade in endothelial cells in part via interferon-gamma expression. Moreover, we found that VEGFR-2 blockade increased the PD-1 expression in tumor-infiltrating CD4+ cells. We also found that under anti-PD-1 therapy, CD4+ cells promote normalized vessel formation in the face of antiangiogenic therapy with anti-VEGFR-2 antibody.

Conclusion:

We show that dual anti-PD-1/VEGFR-2 therapy has a durable vessel fortification effect in HCC and can overcome treatment resistance to either treatment alone and increase overall survival in both anti-PD-1 therapy resistant and responsive HCC models.

Introduction

Hepatocellular carcinoma (HCC) is the second most common cause of cancer-related death worldwide (1). HCC is a malignant disease that develops predominantly in patients with underlying chronic liver disease and cirrhosis (2, 3). Therefore, inflammation and angiogenesis are critical during the process of hepatocarcinogenesis, when immune evasive mechanisms are established. These mechanisms include upregulation of immune checkpoints on HCC cells as well as CD8+ cytotoxic T lymphocytes (CTLs), antigen presenting cells (macrophages and dendritic cells) and immunosuppressive cells such as T regulatory cells (Tregs) or myeloid-derived suppressor cells (MDSCs) (4-7). Dysregulation of immune checkpoints is critical for immune response evasion in HCC (5).

Programmed death receptor-1 (PD-1)/PD ligand 1 (PD-L1) pathway has been the main target for immune checkpoint blockade (ICB) in HCC, as both the receptor and the ligand are often expressed in human HCC (5, 6). PD-L1 upregulation in the tumor is reported to correlate with alphafetoprotein (AFP), a well-known oncofetoprotein in HCC (8). High PD-1 expression is marker of immune cell activation but also of “exhausted” phenotype of effector lymphocytes. PD-L1 expression on immune, endothelial and cancer cells is largely viewed as a marker of immune suppression, which could be induced by microenvironmental factors (for example, hypoxia or IFN-γ) (9-11). PD-1 and PD-L1 expression correlate with CTL infiltration in HCC, and these parameters associated with survival in patients undergoing surgical resection (12, 13). ICB such as using anti-PD-1 antibodies has shown promise in HCC (5, 6). Clinical data showed a response in approximately 15-20% of HCC patients, with some long survivors (14, 15). As PD-1 blockade is being tested in phase III trials, it is clear that the current challenge will be to address the resistance to ICB seen in the majority of HCC patients.

A critical question is whether combining anti-PD-1/PD-L1 therapy with other treatments could enhance its effectiveness in HCC (16). HCCs are highly vascularized tumors (17), and clinical successes indicated a key role for targeting VEGFR-2-driven angiogenesis in HCC. Unfortunately, the responses to anti-VEGFR-2 drugs are rare and often short-lived (18). This may be because antiangiogenesis can excessively prune vessels and increase hypoxia (4, 9). Interestingly, a recent study has shown that anti-PD-1 therapy can increase blood perfusion of tumors by normalizing vessels in breast and colorectal cancer models (19). However, the effect of anti-PD-1 therapy—alone or with anti-VEGF/R therapy—on HCC vasculature is unknown. Understanding this effect is critically important, as combination therapy may yield outcomes superior to either monotherapy (20, 21). Importantly, a recent study demonstrated the reciprocal regulation of vascular normalization and immune responses, including in HCC (22). Tian et al. reported that CD4+ type 1 helper T cells (Th1) activated with ICB play pivotal role in tumor vessel normalization.

We have previously shown that there is a “window” of normalization depending on the time and dose of anti-angiogenic treatment (23). For example, titration of anti-VEGFR-2 treatment extended the window, which enhanced the efficacy of vaccine therapy in mouse models of breast cancer (24). Here, we hypothesized that combining anti-PD-1 treatment with VEGFR-2 blockade extends the window of normalization and enhances anti-tumor immune response. The hypothesis that dual PD-1/VEGFR-2 blockade can be effective in HCC is supported by early clinical data on combinations of lenvatinib and pembrolizumab and of bevacizumab with atezolizumab showing unexpectedly high overall response rates (25, 26). This has led to a breakthrough designation by the FDA for these combination therapies, and to three ongoing phase III trials of these combinations.

Materials and Methods

Cells and culture conditions.

We used 2 murine cell lines: HCA-1, established in our laboratory (27, 28), and RIL-175 (a p53/Hras mutant HCC cells from C57Bl/6 mice, a kind gift from Dr. Tim Greten, NIH) (29). HCA-1 was maintained in Dulbecco’s modified essential medium (DMEM) with 10% fetal bovine serum (FBS) and pyruvic acid. RIL-175 was maintained in DMEM with 20% FBS with pyruvic acid.

Hypoxic condition cell culture was performed in a Brunswick Galaxy Incubator at 1% for 48hr. Organoid culture was performed using primary cells from tumor tissue. Tumor tissues were resected and minced with scalpel. Minced tissue fragments were incubated in Hank’s buffered salt solution (HBSS) with 15μg/ml of collagenase and 1.5mg/ml of hyaluronidase for 30min in 37μC. Digested tissues were passed through 70μm cell strainer and washed twice with HBSS. After counting the cell concentration, cells were diluted in 1 million cells/ml with DMEM with 10% FBS. The cell suspension was poured in Nano culture plate (MH type, SCIVAX, Organogenix, Japan) for 72hr to form organoids in 37μC with 5%CO₂. After the confirmation of organoid formation by light microscope, 50μg/ml of DC101 and/or 20μg/ml of anti-IFN-γ blocking antibody was added to the medium and organoids were placed in the specific experimental culture condition. Primary endothelial cells were obtained from single cell suspension of RIL-175 tumors. CD31 positive cells were sorted using biotin-conjugated CD31 antibody (cloneMEC13.3, Biolegend) and anti-biotin micro beads (Miltenyi, Germany) by magnetic sorting. Endothelial cells were cultured on the type I collagen coated cell culture dish, using Endothelial Cell Growth Medium MV2 (Promo Cell).

HCC models.

The orthotopic HCC mouse models were described elsewhere (30). The HCC cells (one million 1:1 in Matrigel, Mediatech/Corning, Manassas, VA) were grafted in mice matching their genetic background—HCA-1 in C3H mice and RIL-175 in C57BL/6 mice. Tumor growth was monitored by high-frequency ultrasonography twice a week. In the Mst1^–/–Mst2^f/– model (31), adeno-Cre at a concentration of 10^7 pfu was injected via the tail vein in 3-week-old mice. To induce liver damage, 100μl of 20% carbon tetrachloride (CCl4)(Sigma-Aldrich, Saint Louis, MO) was administrated orally from 8-12wk (30). When tumors reached ~5mm in diameter, mice were randomly assigned to a treatment group (n=6-8). Treatments were administered i.p. thrice/wk for 14 days at a dose of 10mg/kg (anti-PD-1 antibody, anti-CD4-antibody and IgG control), and 10mg/kg (low) or 40mg/kg (standard) (anti-VEGFR-2 antibody DC101). All animal experiments were performed after approval by the Institutional Animal Care and Use Committee of the Massachusetts General Hospital.

Reagents.

DC101 antibody was kindly provided by Eli Lilly and Company (Indianapolis, IN). Mouse anti-PD-1 antibodies (clone RMP-014) and anti-CD4 antibodies was purchased from BioXcel (West Lebanon, NH). Control rat IgG was purchased from Jackson ImmunoResearch (West Grove, PA). Anti-IFN-γ antibody (clone XMG1.2) was purchased from Thermo Fisher Scientific (Waltham, MA).

Flow cytometry.

Tumor tissues were resected and minced, and fragments were incubated in HBSS with 15μg/ml of collagenase and 1.5mg/ml of hyaluronidase for 30min at 37°C. Digested tissues were passed through 70μm cell strainer and washed twice with PBS/ 0.5% BSA. Single cell suspensions were incubated with anti-mouse CD16/32 antibody (clone 93, Biolegend, San Diego, CA) for 5min prior to staining for immune cell markers for 15min at room temperature. For the intracellular markers, cells were fixed and permeabilized with either FoxP3/Transcription Factor Staining Buffer Set (eBioscience/Thermo Fisher Scientific, Waltham, MA) or BD Cytofix/Cytoperm™ Kit (BD Biosciences, San Jose, CA), according to manufacturers’ protocols. For cytokine staining, harvested cells were incubated in RPMI with cell activation cocktail with brefeldin A (Biolegend) for 6hr at 37°C, and stimulated cells were stained same as above. Monoclonal antibodies used for flow cytometric analysis were, CD45 (30-F11), CD3e (145-2C11), CD4 (RM4-5), CD8 (53-6,7), PD-1 (29F.1A12), IFN-γ (XMG1.2), and FoxP3 (FJK-165).

Cell sorting.

For the magnetic bead sorting of endothelial cells or myeloid cells, cells were first stained with either biotin-conjugated anti-CD31 (cloneMEC13.3), or anti-CD11b (clone M1/70) and anti-CD11c antibodies (clone N418) (all Biolegend). Biotin-labeled cells were separated by magnetic sorting using anti-biotin microbeads (Miltenyi, Germany). For macrophage collection, CD45+CD11b+Ly6C-F4/80+ population was sorted by using a FACS Aria cell sorter.

Immunohistochemistry (IHC) and immunofluorescence (IF).

Resected tumor tissues were either fixed in 4% paraformaldehyde and embedded in paraffin or embedded in OCT and frozen. Vascular endothelial staining was performed using anti-CD31 antibody (clone DIA-310, Dianova, Germany for IHC and Millipore, Saint Louis, MO for IF), pericytes were identified with antibodies to α-smooth muscle actin (SMA)(Sigma-Aldrich, Saint Louis, MO), PD-L1 expression was detected using anti-PD-L1 antibodies (eBioscience), and hypoxia was detected with anti-Carbonic Anhydrase IX (CA-IX) antibody (Abcam plc, Cambridge, UK). Specimens were incubated with fluorescence (Cy3, and Alexa 647)-conjugated anti-hamster or anti-rabbit secondary antibodies, as appropriate (Jackson ImmunoResearch). Cell nuclei were identified with DAPI (ProLong™ Gold Antifade Mountant with DAPI) (Thermo Fisher Scientific). The total number of vessels and pericyte-covered vessels, and the fraction of hypoxic tumor area were counted in five random fields using a ×400 magnification lens. These data were analyzed using ImageJ (US NIH) and PhotoShop (Adobe Systems Inc., San Jose, CA) software. IF or organoids was performed after fixation in 0.4% formaldehyde for 15min. Then, organoids were washed 3 times with PBS and immunostained. IF images were analyzed using a confocal microscope using (FLUOVIEW FV100, OLYMPUS, Center Valley, PA).

RNA sequencing analysis.

When tumors reached 200mm³, 3-5 samples per group were collected from mice treated with control IgG, anti-PD-1 antibody alone, anti-PD-1 antibody in combination with DC101, and anti-PD-1 antibody in combination with DC101 and anti-CD4 antibodies. Mice were sacrificed and total RNA was isolated from the freshly isolated tumors tissues using Qiagen kits. Total RNA was sequenced at Molecular Biology Core Facilities, Dana Farber Cancer Institute (Boston, MA), and used for Gene Set Enrichment Analysis (GSEA).

Cytokine analysis.

Culture supernatants were collected after the specific experimental treatment. Samples were assayed in duplicate using the MSD proinflammatory Panel I, a highly sensitive multiplex enzyme-linked immunosorbent assay (ELISA) for quantitatively measuring 10 cytokines: IFN-γ, interleukin (IL)-1β, IL-2, IL-4, IL-6, IL-8, IL-10, IL-12p70, IL-13, and tumor necrosis factor (TNF)-α from a single small sample volume (25μl) using electrochemiluminescence-based detection (MesoScale Discovery, Gaithersburg, MD).

Statistical analysis.

All statistical analyses were performed using Stata software, version 14.1 (Stata Corp LLC, College Station, TX). Error bars indicate standard error of mean. Differences were considered significant when p values were less than 0.05. Quantitative variables were compared using Student’s t test. Analysis of experiments with more than two groups was performed using one-way ANOVA with Scheffe’s correction for multiple comparisons. Median overall survival (OS) was estimated using the Kaplan–Meier method. Statistical analyses in the survival experiments were performed by Cox proportional hazard model and hazard ratio (HR) and 95% CI were calculated as well. Statistical significance was indicated in the figures as follows: *P<0.05; **P<0.01; ***P<0.001.

Results

Dual VEGFR-2/PD-1 blockade is effective in orthotopic HCC models in mice.

We tested combination therapy in orthotopic models of HCC in mice with (chemically induced) underlying liver damage and examined the impact of the dose of the antiangiogenic agent. We treated mice with established HCC (5–6mm in diameter) with the anti-VEGFR-2 antibody DC101 at two different doses (AA-low, 10mg/kg and standard AA, 40mg/kg, both thrice a week), anti-PD-1 antibody (ICB), or their combination. Using HCA-1 in C3H mice, we found that in the single agent treatment groups, there was a significant growth delay only in the AA group compared to control, and non-significant delays in the others; however, combinations of AA and ICB therapy—irrespective of the dose—induced comparable significant growth delays (Fig. S1A). In addition, there was a significant increase in median OS in the combination therapy groups, which was not seen in any of the monotherapy arms (Fig. 1A). The lack of survival advantage in AA group was due to lack of control of metastatic disease, despite delay of the primary tumor growth; the metastatic burden was reduced in all the other groups at the experimental endpoint (Fig. S1B). Tumor growth inhibition was also seen in the genetically engineered RIL-175 (p53/Hras) and Mst-mutant HCC models (Fig. S1C,D). Next, we tested the impact of dual AA/ICB in C57Bl/6 mice bearing RIL-175 HCCs—sensitive to anti-PD-1 therapy. We found that the combination therapy produced the longest median OS (Fig. 1B).

Fig. 1: Efficacy of dual anti-PD-1/DC101 treatment in HCC models.

A) Kaplan-Meier survival distribution in the orthotopic HCA-1 model after immune checkpoint blockade (ICB) with anti-PD-1 antibody, anti-VEGFR-2 blockade with DC101 low-dose (AA-low, 10mg/kg) or high-dose (AA, 40mg/kg) or their combinations compared with Control (C) IgG group. Combination of anti-PD-1 treatment with either low-dose (AA-low/IBC) or high-dose (AA/IBC) of DC101 showed significantly longer survival compared with Control IgG group [AA-low/IBC vs C: Hazard Ratio (HR) = 0.13, 95% confidence interval (CI) 0.04, 0.42, p=0.001; AA/IBC vs C: HR=0.17, 95%CI 0.05, 0.52, p=0.002]. B) Kaplan-Meier survival distribution in the orthotopic RIL-175 model after ICB with anti-PD-1 antibody, anti-VEGFR-2 blockade with DC101 (AA) or their combinations compared with IgG group (C). The AA/ICB combination showed significantly longer survival compared with control, with a HR=0.054, 95%CI 0.006, 0.46, p=0.007. *p<0.05; **p<0.01; ***p<0.001.

Dual VEGFR-2/PD-1 blockade reprograms the immune microenvironment and promotes anti-tumor immunity in HCC.

Having reproduced the clinical benefit seen with combination immunotherapy/antiangiogenesis in mouse models, we first examined how therapy impacted lymphocyte infiltration. We first analyzed the number of tumor-infiltrating leukocytes as a fraction of total CD45⁺ immune cells by flow cytometry. The number of tumor-infiltrating CTLs was not affected by treatment with ICB alone or AA-low/ICB (Fig. S2). AA decreased the proportion of infiltrating CTLs, even when combined with ICB in RIL-175 HCCs (Fig. 2A and Fig. S2A, B). However, the overall fraction of infiltrating CD4+ T cells and the CD8+/CD4+ T-cell ratio were not affected in any of the combination groups (Fig. S2C, D). Moreover, the fraction of FoxP3+ Tregs was significantly decreased in all treatment groups, which translated into increased CTL/Treg ratios in the ICB-containing groups but not in the AA alone groups (Fig. 2B, C and Fig. S2E, F). These findings were confirmed in the Mst-mutant model of HCC, where we found a decrease in Tregs, and an increase in CD8+/CD4+ and CTL/Treg ratios (Fig. S3G-I). Furthermore, when evaluating CTL function by evaluating IFN-γ expression, we found it to be significantly increased in all ICB-containing groups (Fig. S2J). Finally, using IHC we found that the number of tumor-infiltrating CTLs was significantly increased in the combination treatment groups compared to AA treatments alone (Fig. S2L).

Fig. 2: Effect of dual anti-PD-1/VEGFR-2 blockade on immune stimulation in HCC.

A) Changes in CD8+ cytotoxic T lymphocytes (CTLs) among tumor-infiltrated immune cells (shown as fractions of CD45+ positive cells) in RIL-175 tumors treated with control IgG, anti-PD-1 antibody (ICB), anti-VEGFR-2 antibody (AA), or their combination, measured by flow cytometry at day 8. AA decreases the fraction of CTLs among HCC-infiltrating immune cells. B) All treatments decreased the fraction of infiltrating CD4+FoxP3+ T regulatory cells (Tregs). C) As a result, ICB-treated groups showed favorable ratios of CTLs to Tregs after treatment. D) Similarly, the fraction of interferon gamma (IFN-γ)-positive CD8+ T cells increased in both ICB-treated groups. E, Changes in immune transcriptome from RNA-sequencing analysis (complete dataset in Table S1) showing gene expression patterns consistent with the activation of anti-tumor immunity. *p<0.05; **p<0.01; ***p<0.001.

We also evaluated the changes in myeloid cell populations post-treatment using flow cytometry. AA-low treatment increased the fraction of F4/80⁺CD80⁺ and/or CD86⁺ anti-tumor “M1-type activated” tumor-infiltrating macrophages (TAMs) (Fig. S3). ICB treatment alone did not change the faction on these populations; however, AA/ICB combinations increased the faction of M1-TAMs (Fig. S3A). Moreover, AA/ICB combinations also decreased the fraction of both pro-tumor F4/80⁺CD206⁺ “M2-type activated” TAMs (Fig. S3B), and CD11b⁺CCR2⁺ monocytic cells (Fig. S4C); these effects appeared to be primarily mediated by AA (Fig. S4D-F).

To examine more broadly the immune effects of AA/IBC therapy, we performed RNA-sequencing and found that ICB (alone or with AA) significantly upregulated immune pathways (including proliferation, function/activity, and migration), compared to control-treated HCC tissues. Moreover, AA/IBC therapy selectively upregulated pathways related to myeloid cells and B cells that were not enhanced by ICB alone (Fig. 2E and Table S1).

PD-L1 and PD-1 expression is increased after VEGFR-2 blockade in HCC in vivo.

We have previously reported that VEGFR2 inhibition with the multitargeted TKI sorafenib increases PD-L1 expression in these HCC models (9). Thus, we examined whether specific VEGFR-2 blockade has any impact of PD-L1 expression in the HCC tissue in vivo. Using flow cytometry, we found that the number of non-immune (CD45^–) cells expressing PD-L1 was increased in tumors after VEGFR-2 blockade alone or in combination with ICB (Fig. 3A and Fig. S4A). The high expression of PD-L1 in both cancer and stromal cells was confirmed by IHC (Fig. 3B and Fig. S4B). Of note, double staining using anti-PD-L1 and anti-F4/80 antibodies showed that PD-L1 was predominantly expressed in TAMs in control and AA-low treated tumors, while AA treated tumors exhibited PD-L1 expression both in TAMs and non-immune cells (Fig. 3C). Interestingly, AA significantly increased the fraction of CD4+ cells expressing PD-1 and the PD-1 expression level (Fig. 3D-F).

Fig. 3: PD-L1 and PD-1 upregulation after VEGFR-2 blockade in HCC.

A) Fraction of PD-L1 positive cells by flow cytometry in RIL-175 HCC tissues. Fraction of PD-L1 positive cells is increased in anti-VEGFR-2 antibody (AA)-treated groups irrespective of dose compared to control or compared to anti-PD-1 therapy (ICB) alone, respectively. B) Representative PD-L1 IHC in spontaneous HCCs from Mst1/2-mutant mice: AA treatment increased PD-L1 expression. C) Double-immunofluorescence for PD-L1 and the TAM marker F4/80; PD-L1 expression is predominantly seen in TAMs, with the exception of high-dose DC101 treatment of mice bearing HCC, where PD-L1 expression is increased in tumor and other stromal cells (arrows). D-F) AA treatment does not significantly change the fraction of CD4+ cells in HCC tissue (D) but increases the fraction of PD1+CD4+ cells (E), and also the PD-1 staining intensity – mean fluorescence intensity or MFI (F) – measured by flow cytometry. **p<0.01; ***p<0.001.

PD-L1 expression in HCC cells is upregulated upon VEGFR-2 blockade in endothelial cells in part via IFN-γ production.

To gain insights into the mechanism of PD-L1 upregulation after DC101 treatment, we used organoid 3-D cultures in vitro. Single suspension of cells was obtained after enzymatic digestion of tumor tissues, and then incubated in Nano-culture plated under normoxic conditions. The cells formed heterotypic tumor organoids consisting of multiple cell types, including endothelial cells, after 3 days, and then organoids were cultured in hypoxic conditions for 48hr. While PD-L1 expression was detectable, it was only slightly increased in hypoxic conditions (Fig. S4C, D). However, when we treated the organoids with blocking doses of DC101 (50μg/ml) (32), PD-L1 expression was substantially upregulated in tumor and stromal cells (Fig. 4A, B). In this HCC model, VEGFR-2 expression was restricted to endothelial cells and to a subset of myeloid cells (Fig. S5). To elucidate the nature of the paracrine interaction leading to PD-L1 upregulation after VEGFR-2 blockade, we next depleted certain cell lineage from the original single cell suspension prior to organoid formation. First, we depleted all VEGFR-2+ cells from the single cell suspension using magnetic beads and treated the resulting organoids with DC101 under hypoxic conditions. DC101 treatment failed to upregulate PD-L1 expression in these conditions (Fig. S4E). Next, we depleted endothelial cells using anti-CD31 antibodies or myeloid cells using anti-CD11b and anti-CD11c antibodies, and generated organoids that were treated with DC101 in hypoxic conditions. In these conditions, PD-L1 upregulation was not seen in organoids where endothelial cells were depleted but was readily detectable in organoids where myeloid cells were depleted (Fig. S4E). Collectively, these results indicate that upregulation of PD-L1 in HCC is dependent on VEGFR-2 blockade in endothelial cells, particularly in hypoxic conditions, which mimic the in vivo effects of AA.

Fig. 4: VEGFR-2 blockade in endothelial cells increases PD-L1 expression in HCC cells.

A, B) VEGFR-2 blockade with DC101 induces high PD-L1 expression in heterotypic tumor organoids cultured in normoxic (A) as well as hypoxic (B) conditions (see controls in Fig. S4C, D). C, D) Heterotypic organoid culture of primary HCC-derived cells were cultured with or without DC101 and anti-IFN-γ antibody (C). IFN-γ blockade partially prevented the increase in PD-L1 expression induced by DC101 treatment independently of hypoxia (D). Quantification was performed using Image J and in five random fields. ***p<0.001.

To examine the mechanism of PD-L1 expression regulation in a paracrine manner by the endothelial cells, CD31 positive cells were sorted from single cell suspension of RIL-175 tumors and cultured in vitro. We collected supernatant of endothelial cell culture treated with or without DC101 and evaluated inflammatory cytokine production by multiplexed ELISA. DC101 treatment enhanced most notably IFN-γ production by the endothelial cells (Fig. S4F). Since IFN-γ is a key inducer of PD-L1 expression (33), we treated the organoids with DC101 and blocked IFN-γ using a blocking antibody. IFN-γ blockade decreased PD-L1 expression by half in the HCC endothelial cells (Fig. 4C, D), indicating the that HCC cell PD-L1 upregulation by DC101 targeting of VEGFR-2 in endothelial cells is in part mediated by IFN-γ expression by endothelial cells.

ICB therapy promotes vascular normalization when combined with AA treatment.

Since immunostimulatory reprogramming and vascular normalization may be reciprocally regulated in HCC (22), we next examined the effects of treatment on vascular structure and function in size-matched tumors. We found that ICB increased the total and pericyte-covered microvessel density (MVD) in established HCCs (Fig. 5A-C). Surprisingly, dual AA/ICB treatment further increased total and pericyte-covered MVD, and prevented an increase in hypoxia in HCC, in contrast to the effect of AA alone (Fig. 5D, E). These data demonstrate that ICB promotes durable vascular formation and normalization even when used in combination with AA.

Fig. 5: Effects of dual anti-PD-1/VEGFR-2 blockade on HCC vessel formation, structure and function and the role of CD4+ cells.

A) Representative immunofluorescence (CD31, α-SMA, and DAPI counterstaining) of endothelial cells and pericytes in size-matched HCC tissues after treatment with immune checkpoint blockade (ICB) alone, ICB + anti-VEGFR-2 therapy (AA) and, ICB + AA + anti (a)CD4 antibody therapy. B, C) Dual PD-1/VEGFR-2 blockade significantly increases microvessel density (MVD in B) and pericyte-covered MVD (C) in HCC, and this effect is prevented by CD4 cell depletion. D) Representative immunofluorescence for CA-IX (to detect hypoxia), CD31 (for endothelial cells), and DAPI counterstaining in RIL-175 HCC tissues after treatment. E, ICB treatment decreased tissue hypoxia, both alone and combined with AA, and this effect was prevented by CD4+ cell depletion. *p<0.05; **p<0.01; ***p<0.001.

Since combination treatment did not increase CTL infiltration or activation over ICB alone, but increased PD-1 expression in CD4+ cells, we next sought to decipher whether CD4+ cells play a role in the modulation of vascular structure and function in HCC. Of note, a recent study showed that CD4+ cells mutually regulate vascular and immune effects in developing tumors in an IFN-γ-dependent manner (22). We found that CD4+ cell depletion abrogated the normalized vessel formation seen after AA/ICB treatment and resulted in increased hypoxia (Fig. 5). Strikingly after CD4+ cell depletion during AA/ICB compared to AA/ICB with intact CD4+ cells, RNA-sequencing revealed that 25/235 significantly downregulated pathways were related to blood vessel formation, maturation and normalization (Fig. S6 and Table S1). Thus, CD4+ cells promote vascular formation and remodeling after ICB despite potent AA in HCC (Fig. 6).

Fig. 6: Schematic of the mechanism of interaction between anti-PD-1 and anti-VEGFR-2 therapy in HCC.

The benefit of dual anti-PD-1/VEGFR-2 treatment is due to CD4+ cell-mediated normalized vessel formation; normalized vessels and blockade of PD-1/PD-L1 axis lead to reduction in Tregs and CCR2+ monocytes, shift from M2– to M1-type in tumor-associated macrophages, and promotion of cytotoxic T lymphocyte (CTL) infiltration and activation.

Discussion

The effects of AA—a treatment modality widely used in oncology currently—depend on the dose, timing and specificity of the drugs as well as on the tumor type (4). These effects may involve transient normalization of the vascular structure and function as well as pronounced vascular rarefaction, which have been proven critical for the metastatic progression of tumors as well as for AA interactions with cytotoxic drugs (4). These effects remain unclear in HCC, a disease where several AAs are standard (34). The potential of AAs to enhance anti-tumor immunity when combined with immunotherapy in patients has been postulated by others and us by diverse mechanisms based on data from preclinical models, which all converged to normalization of tumor vessels and microenvironment as a principle (4, 10, 20, 23, 24). On the other hand, an increasing body of evidence describes the complex effects of immune cell activation on tumor vessel normalization. For example, Tian et al. (22) demonstrated that—in a preventative setting—normalized vessel formation and anti-tumor immune responses were reciprocally regulated by CD4+ and not CD8+ cells in an IFN-γ-dependent manner in breast cancer models. This contrasted with data generated in interventional setting, where Zhang et al. (19) showed that CD8+ T cell activation by ICB alone mediated the normalization of tumor vessels, albeit also in an IFN-γ-dependent manner.

Other reports showed that the benefit of combinations of AAs with ICB was associated with substantial pruning of the tumor vasculature. Thus, beyond the questions on the specific role of combining AA with ICB for HCC therapy, these reports also raise the question on the roles of passive (pruning of abnormal vessels) versus active (formation of normalized vessels via increased pericyte coverage) normalization, and which is more critical for the efficacy of AA/ICB therapy. For example, Schmittnaegel et al. (10) showed in the same breast cancer model (MMTV-PyVT) as well as in pancreatic neuroendocrine tumors (Rip1-Tag2) and melanomas that dual blockade of angiopoietin 2 (Ang-2) and VEGF promoted vascular regression, tumor necrosis, and antigen presentation by intratumoral phagocytes. Dual Ang-2/VEGF blockade also structurally normalized the remaining blood vessels and facilitated the extravasation and perivascular accumulation of activated CTLs. This effect was further enhanced by PD-1 blockade which counteracted the increased endothelial PD-L1 expression induced by IFN-γ. Allen et al. (20) showed that the efficacy of AA/ICB treatment with anti-VEGFR-2 and anti–PD-L1 antibodies was associated with significant pruning of vessels and induction of structural normalization and “high endothelial venule” phenotype in the remaining tumor vessels, and a concomitant increase in T cell recruitment in MMTV-PyVT, Rip1-Tag2 and glioblastoma models.

Here, using clinically relevant mouse models of HCC, we directly addressed two critical issues. First, we demonstrate that a VEGFR-2-targeted agent can be successfully combined with ICB to increase its efficacy in HCC. For HCC, this question is timely and is in line with promising phase I clinical trial data in HCC patients, which showed unexpectedly high response rates and led to randomized phase III trials (e.g., NCT03744247, NCT03434379) and the “breakthrough” designation by the US FDA. Moreover, dual VEGF/PD-L1 blockade has already shown increased overall survival in lung cancer patients (35). Second, we unravel an unexpected mechanism of benefit for specific blockade of PD-1 and VEGFR-2 with antibodies in HCC. In size-matched tumor samples, we show here that dual AA/ICB therapy not only does not cause pruning of the abnormal vessels, but rather promotes normalized vessel formation mediated by CD4+ cells. These treatment interactions led to reprogramming of the immune microenvironment of HCC (Fig. 6). This is in contrast to pruning of vessels in these tumors after AA alone, which increases the expression of PD-1 on CD4+ cells.

It remains currently unknown whether titrating the dose of an anti-VEGFR TKIs could also induce changes in the tumor microenvironment of HCC that will lead to benefits when combined with ICB. This will be critical in this disease, where several multi-targeted TKIs are FDA approved as monotherapy. In addition to potential effects on the microenvironment, reducing the dose of the TKIs may also mitigate concerns related to toxicity of such combinations in HCC patients, which often have poor liver function (15). Also, other mechanisms may play a role in this treatment interaction, as VEGF/VEGFR-2 signaling blockade does not only affect tumor angiogenesis but also regulates key anti-tumor immune response through mechanisms such as suppression of CTL activity through inhibition of dendritic cell maturation (36), inactivation of STAT3 (37, 38), or induction of exhaustion in intratumoral CTLs (39). Terme et al. showed that VEGF pathway inhibition decreased Treg infiltration in a mouse model of colorectal cancer (40). These findings highlight the complexity and the intricate mechanisms of interaction between these treatments in different tumors. They also underscore the potential for dual AA/ICB therapy across tumor types, including in subsets known to be refractory to either one or both treatment types.

While the predictive value of PD-L1 expression for response to anti-PD-1/PD-L1 therapy is being debated, a consistent finding emerging from multiple reports is the upregulation of PD-L1 in cancer and stromal cells after AA treatment (9, 10, 20, 41). This upregulation should be examined as a “dynamic biomarker” of response to AA/ICB. The mechanisms of this upregulation appear to be tumor, antiangiogenic agent, dose and context dependent. PD-L1 upregulation by hypoxia has been inconsistent between studies (10, 20, 39); we found that hypoxia may lead to a modest increase in PD-L1 expression in HCC, but that agents such as sorafenib could increase PD-L1 expression even in normoxic conditions via MAPK and NF-kB activation (41). Here, we used an HCC organoid culture model to demonstrate that PD-L1 upregulation is mediated by VEGFR-2 blockade in endothelial cells through paracrine effects on cancer and stromal cells, mediated at least in part by IFN-γ production.

Future studies should establish the extent by which immune cell activation improve the cytostatic effect of VEGFR2 blockade or VEGFR2 blockade improve immune cell trafficking and activation; they should also validate our findings using human tumor specimens—if available post-treatment. However, the available clinical evidence supports the further development of dual AA/ICB therapy. High circulating plasma levels of VEGF associated with treatment resistance in melanoma patients treated with anti-CTLA-4 antibodies (42). Hodi et al. reported promising efficacy signals with combination of anti-VEGF and anti-CTLA-4 therapy in melanoma patients (43). Moreover, the anti-VEGFR-2 human antibody ramucirumab has shown improved survival in gastric, colorectal and lung cancer patients. In HCC, ramucirumab initially failed to increase survival despite showing favorable safety and tolerability profiles and inducing a significant delay in tumor progression compared with placebo among the patients previously treated with sorafenib (REACH trial, NCT01140347) (44). However, a follow-up biomarker-driven randomized phase III clinical trial showed significant survival benefit in HCC patients with high baseline serum AFP level or preserved liver function (REACH-2 trial, NCT02435433) (45). Moreover, a recent study of ramucirumab with pembrolizumab showed a manageable safety profile with favorable antitumour activity in patients with previously treated advanced gastric or gastro-esophageal junction adenocarcinoma, non-small-cell lung cancer, and urothelial carcinoma (46).

In conclusion, we show here that dual PD-1/VEGFR-2 antibody blockade is effective in HCC models. The efficacy was seen both when using “vascular normalizing” and anti-vascular doses of AA and was mediated by the vascular normalizing effects of anti-PD-1 therapy itself mediated by CD4+ cells. AA/ICB therapy counteracted the immunosuppressive cues in the HCC microenvironment: PD-L1 upregulation in HCC induced by VEGFR-2 blockade in tumor endothelial cells, and Treg and CCR2+ monocyte infiltration. Because of ICB-induced vascular normalization, the dose of anti-VEGFR-2 antibody therapy may not be as critical as when it is given alone. However, our results also indicate that AA-low therapy—seldom tested in preclinical or clinical studies—may be sufficient to appropriately reprogram the immune microenvironment and enhance ICB efficacy. These findings are important for future testing of these clinically relevant drugs in HCC and other cancers.

List of Abbreviations

AA

antiangiogenic

AFP

alphafetoprotein

CTL

cytotoxic T lymphocyte

CTLA-4

cytotoxic T lymphocyte antigen 4

DMEM

Dullbeco’s modified essential medium

ELISA

enzyme-linked immunosorbent assay

FACS

flow-activated cell sorting

FBS

fetal bovine serum

GSEA

Gene Set Enrichment Analysis

HBSS

Hank’s buffered salt solution

HCC

Hepatocellular carcinoma

HR

hazard ratio

IF

immunofluorescence

IFN

interferon

IHC

immunohistochemistry

IL

interleukin

ICB

immune checkpoint blockade

MDSCs

myeloid-derived suppressor cells

MVD

microvessel density

OS

overall survival

PD-1

programmed death receptor-1

PD-L1

PD ligand 1

VEGF

vascular endothelial growth factor

VEGFR

VEGF receptor

TAM

tumor-infiltrating macrophages

Th1

type 1 helper T cells

TKI

tyrosine kinase inhibitor

TNF

tumor necrosis factor

Treg

T regulatory cell