Cancers evade immune surveillance, which can be reversed through immune-checkpoint therapy in a small subset of cases. Here, we report that the MYC oncogene suppresses innate immune surveillance and drives resistance to immunotherapy. In 33 different human cancers, MYC genomic amplification and overexpression increased immune-checkpoint expression, predicted nonresponsiveness to immune-checkpoint blockade, and was associated with both Th2-like immune profile and reduced CD8 T-cell infiltration. MYC transcriptionally suppressed innate immunity and MHCI-mediated antigen presentation, which in turn impeded T-cell response. Combined, but not individual, blockade of PDL1 and CTLA4 could reverse MYC-driven immune suppression by leading to the recruitment of proinflammatory antigen-presenting macrophages with increased CD40 and MHCII expression. Depletion of macrophages abrogated the antineoplastic effects of PDL1 and CTLA4 blockade in MYC-driven hepatocellular carcinoma (HCC). Hence, MYC is a predictor of immune-checkpoint responsiveness and a key driver of immune evasion through the suppression of proinflammatory macrophages. The immune evasion induced by MYC in HCC can be overcome by combined PDL1 and CTLA4 blockade.

Significance:

Macrophage-mediated immune evasion is a therapeutic vulnerability of MYC-driven cancers, which has implications for prioritizing MYC-driven hepatocellular carcinoma for combination immunotherapy.

The MYC oncogene is one of the most common drivers of human neoplasia, with a majority of human cancers having genetic alterations either in MYC or in one of the members of the proximal MYC network (1). This makes MYC a very desirable therapeutic target. However, no existing therapies directly target MYC. MYC contributes to tumorigenesis both by causing tumor-intrinsic effects such as autonomous cellular proliferation, as well as by tumor extrinsic effects such as remodeling the tumor microenvironment via immune evasion (2). MYC inhibits antitumor immunity by multiple mechanisms, including impeding CD4+ T-cell activation, promoting expression of immune checkpoints like PDL1 and CD47, and blocking natural killer (NK) cell activation (2–7). These immune effects suggest that MYC-driven tumors may be vulnerable to specific combinations of immune therapies.

MYC activation is a major oncogenic event in many types of epithelial tumors including hepatocellular carcinoma (HCC). HCC is an aggressive malignancy with dismal prognosis and the MYC pathway activated in 70% of these tumors (1). Moreover, several studies have established that MYC plays a crucial pathogenic role in HCC progression (8–12). Yet, MYC-driven HCCs remain a therapeutically challenging subset of HCCs. Immune-checkpoint inhibitors (ICI) have shown promise in many human cancers but, notably, have failed as first-line monotherapy for HCC (13, 14). Recently, the combination of PD1 and CTLA4 blockade has shown promise in HCC. However, only around 20% of patients showed an objective response, whereas around 50% of the patients experienced serious adverse events (15). Understanding the mechanisms by which individual or combined ICIs are effective in subsets of patients could help identify the right patient for the right treatment and lead to the development of further therapies targeting the mechanisms of resistance.

We hypothesize that MYC is a major driver of immune evasion and dictates response to immune-checkpoint inhibition in HCC. Here we demonstrate the combination of checkpoint inhibitors PDL1 and CTL4 can overcome MYC immune evasion by restoring macrophage activation via CD40 and MHCII expression, thereby eliciting robust antitumor CD8+ T-cell responses.

Human pan-cancer TCGA analysis

We used the TumorMap (https://tumormap.ucsc.edu/; ref. 16), a tool that enables grouping The Cancer Genome Atlas (TCGA) samples in a visually accessible way. TumorMap uses dimensionality reduction methods to condense high-dimensional omics data to a two-dimensional space. In the TumorMap, each node is a sample, and clusters of samples indicate groups with similar genomic alteration events or oncogenic signatures. We used the TCGA pan-cancer data to visualize samples with MYC amplification (17). We visualized the MYC gene signature derived from gene set enrichment analysis of MYC amplification across the TCGA pan-cancer samples (http://www.broadinstitute.org/gsea/msigdb/cards/chr8q24). The Spearman test was used for correlation analysis. Kaplan–Meier analysis was performed for survival analysis. We used the immune signatures and immune deconvolution analysis published in the pan-cancer TCGA immune landscape data (18).

RNA-seq

RNA-seq of MYC-HCC was performed at the Beijing Genomics Institute (BGI) using their BGIseq 500 platform single end 150 bp, 20 million reads per sample. MYC-activated tumors (MYC-On, n = 3) and tumors 4 days after MYC inactivation (MYC-Off, n = 3) underwent whole-transcriptome sequencing. Gene expression level was quantified by a software package called RSEM. For each RNA-seq data sample deposited in Gene Expression Omnibus (GEO), we counted the number of identified expressed genes and calculated its proportion to the total gene number in the database. DEseq software was used to perform differential expression analysis. A random forest classifier algorithm was used to derive a 56-gene signature by comparing mouse MYC-HCC On vs. Off, which was used to stratify human HCC in the TCGA cohort. Qlucore Omics Explorer v. 2.2 was used to create heat maps for the visual representation of data.

Ingenuity Pathway Analysis (IPA; Qiagen) was used to perform functional pathway analysis and similarity analysis using “Analysis Match.” To compare the MYC-HCC transcriptional changes to other analyses, “Analysis Match” builds a signature from the highest confidence predictions for MYC-HCC and compares it to a curated list of analyses from public gene expression data sets. A signature in the MYC-HCC analysis was considered to match a signature from another human HCC transcriptome analysis if the predicted activated and inhibited entities strongly overlap.

The CIBERSORT gene expression deconvolution package was used to estimate the immune cell composition in the MYC-HCC (19). The LM22 signature was used as the immune cell gene signature after the careful modification of genes in the signature to their respective mouse orthologs. The settings for the run were: 1,000 permutations with quantile normalization disabled. The Student t test was used to infer the statistical significance of the predicted immune cell populations where P < 0.05 was considered significant.

The immunotherapy-treated MYC-HCC were sequenced using the NanoString IO360 data using software developed by ROSALIND, Inc following the nCounter Advanced Analysis protocol (20). Housekeeping probes to be used for normalization were selected based on the geNorm algorithm as implemented in the NormqPCR R library (20). Significant P values (P < 0.05) were adjusted using the Benjamini–Hochberg method of estimating false discovery rates. We used the meta-analysis pipeline from ROSALIND to compare the transcriptional changes induced by the combination therapy and the individual monotherapies versus control antibody-treated MYC-HCC. ROSALIND meta-analysis clusters the data using K-means cluster for value k = 2 to 15, then calculates the optimal number of clusters using the PAM method from R package fpc with method “clara,” and runs enrichment analysis using hypergeometric tests. Hypergeometric distribution was used to analyze the enrichment of pathways, gene ontology, domain structure, and other ontologies. The topGO R library was used to determine local similarities and dependencies between GO terms to perform Elim pruning correction. Enrichment was calculated relative to a set of background genes relevant for the experiment.

Transgenic mice

Animals were housed in a pathogen-free environment at Stanford University, and all procedures were performed in accordance with Stanford's Administrative Panel on Laboratory Animal Care protocols and approved by the Institutional Animal Care and Use Committee. LAP-tTA/tet-O-MYC transgenic lines were used, as previously described (21). Mice were administered weekly doses of 0.1 mg/mL doxycycline (Sigma) in drinking water during mating and until 4 weeks of age. At 4 weeks, mice were taken off doxycycline. Mice were screened for tumors via magnetic resonance imaging (MRI) at 2 to 3 months of age, at which time, they developed tumors between 50 and 150 mm3.

MRI

MRI was performed using a 7T small animal MRI (Agilent conversion) with a 40-mm Varian Millipede RF coil (ExtendMR LLC) at the Stanford Small Animal Imaging Facility as previously described (22, 23). Briefly, animals were anesthetized with 1% to 3% isoflurane and placed into the MRI scanner containing a 40-mm Varian Millipede RF coil (ExtendMR LLC). ParaVision (PV6.01) was used to acquire the DICOM images, and tumor volumes were quantified from images using Osirix image processing software (Osirix). Tumor sizes in mice were monitored weekly, and once before and after the completion of treatments.

Immunotherapy treatment

Two days following tumor detection, mice were enrolled into one of four treatment groups. Control Rat IgG (Bio X Cell) and αPDL1 (clone 10 F.9G2, Bio X Cell) antibodies were given intraperitoneally (100 μg/mouse) every other day, αCTLA4 antibody (clone UC10-4 F10-11, Bio X Cell) was given i.p. (100 μg/mouse) every 3 days twice. For the long-term treatment, mice were treated for 2 weeks, and for the mass cytometry experiment, mice were treated for 1 week. For the macrophage-depletion experiments, anti-CSF1R antibody (#BE0213, clone AFS98, BioCXell) was administered (400 μg/mouse) thrice a week for 2 weeks (24, 25).

IHC

Tissues were fixed in 10% paraformaldehyde and embedded in paraffin for sectioning. Sections were deparaffinized by incubation in xylene and rehydrated by sequential incubation in 100%, 95%, 80%, and 60% ethanol. The sections were incubated with primary antibody overnight at 4°C (Supplementary Table S1) and subsequently incubated with biotinylated anti-mouse (1:300, Vector Lab) or biotinylated anti-rabbit (1:300, Vector Lab) for 30 minutes at room temperature. Sections were incubated for 30 minutes at room temperature in an ABC reagent (1:300, Vectastain ABC kit, Vector Lab). Sections were developed using 3,3′-diaminobenzidine (DAB) for 30 to 60 seconds, counterstained with hematoxylin, and mounted with Permount. The stained sections were scanned on Digital Pathology Slide Scanner (Philips) and quantified on ImageJ and Qupath software (NIH).

Mass cytometry analysis

Infiltration of immune cells in the tumor was assessed using mass cytometry (CyTOF). After euthanization, the mice were perfused by injecting PBS into the left ventricle. The dissected tumor was cut into small pieces and processed into a single-cell suspension using the Tumor Dissociation Kit (MACS Miltenyi) according to the manufacturer's instructions. The cell suspension was filtered through a 70-μm cell strainer and treated with the ACK lysis buffer (Thermo Fisher). Approximately 5 × 106 cells were stained for CyTOF analysis. In brief, the cells were incubated for 15 minutes at room temperature with 1 μmol/L Intercalator-Rh (Fluidigm) diluted in Maxpar Cell Staining Buffer (Fluidigm) for dead cell discrimination. Surface Fc receptors were blocked prior to staining by adding 2.5 μg Rat anti-mouse CD16/32 Fc block (BD Pharmingen). Without washing out the blocking antibody, the cells were stained with metal-conjugated antibodies according to the list in Supplementary Table S2 in a total volume of 100 μL per sample and incubated for 45 minutes at room temperature. Antibodies were washed out using the Maxpar Cell Staining Buffer and the cells were resuspended in 125 nmol/L Intercalator-Ir (Fluidigm) diluted in Maxpar Fix and Perm Buffer (Fluidigm) and incubated overnight at 4°C. The intercalator was washed out using the Maxpar Cell Staining Buffer following a washing step in Maxpar water (Fluidigm). The cells were kept as a pellet on ice until analysis. Immediately before analysis, the cells were resuspended to a concentration of approximately 0.5 × 106 cells/mL in Maxpar water containing 10% (v/v) EQ Four Element Calibration Beads (Fluidigm). The samples were analyzed on the DVS CyTOF 2 instrument (Fluidigm). Data analysis was performed using FlowJo v10 and Cytosplore (26). The cells were first gated on “Live CD45+ single cells” in FlowJo and then tSNE analysis was performed in Cytosplore. The different immune cell populations were defined based on the expression of the immune markers. Some antibodies were labeled with metal conjugates in-house using the Maxpar X8 Multimetal Labeling Kit (Fluidigm) according to the manufacturer's instructions. For each labeling, 100 μg pure antibody was used, and the metal-conjugated antibody was subsequently resuspended in 120 μL of antibody stabilization buffer (Candor Bioscience). The metal-conjugated antibodies were titrated prior to use for staining of cells.

Data availability statement

The RNA-seq data for this project are publicly available in the GEO database at GSE215010 and GSE215006.

The MYC oncogene drives human tumorigenesis associated with immune evasion

We evaluated the role of MYC in the immune status of human cancers using TCGA pan-cancer studies (n = 11,069 samples, n = 33 cancers; Fig. 1A). First, MYC was commonly activated by genomic amplification (n = 889, 8%), especially in solid tumors such as HCC (11%; Fig. 1A; Supplementary Fig. S1A; ref. 27). MYC amplification was strongly associated with MYC gene overexpression (P < 0.0001; Supplementary Fig. S1B). Even tumors without MYC amplification showed overexpression of a gene signature associated with MYC amplification (Fig. 1A; ref. 17), which also correlated with MYC mRNA expression (P < 0.0001, r = 0.241). This highlights the ubiquitous nature of MYC activation in cancers. Both MYC amplification and MYC mRNA overexpression were associated with worse overall- and progression-free survival (Fig. 1B). Further, MYC-amplified tumors had higher tumor grade (P = 7.0 × 10−8) and belonged to more advanced stages (P = 2.7 × 10−7; Supplementary Fig. S1C–S1D). Thus, MYC activation by amplification and/or overexpression is a global driver of tumorigenesis in human cancers.

Figure 1.

The MYC oncogene drives human tumorigenesis associated with immune evasion. A, TumorMap rendering of the Pan-Cancer-Atlas of 33 cancers; tumors are defined by different tissues of origin in the first box. The second box shows the prevalence of MYC genomic amplification. Epithelial tumors with high rates of MYC amplification are annotated. The third box shows the gene expression of a signature derived from MYC amplification. LAML, acute myeloid leukemia; ACC, adrenocortical carcinoma; BLCA, bladder urothelial carcinoma; LGG, brain lower grade glioma; BRCA, breast invasive carcinoma; CESC, cervical squamous cell carcinoma and endocervical adenocarcinoma; CHOL, cholangiocarcinoma; LCML, chronic myelogenous leukemia; COAD, colon adenocarcinoma; ESCA, esophageal carcinoma; GBM, glioblastoma multiforme; HNSC, head and neck squamous cell carcinoma; KICH, kidney chromophobe; KIRC, kidney renal clear cell carcinoma; KIRP, kidney renal papillary cell carcinoma; LIHC, liver hepatocellular carcinoma; LUAD, lung adenocarcinoma; LUSC, lung squamous cell carcinoma; DLBC, lymphoid neoplasm diffuse large B, cell lymphoma; MESO, mesothelioma; OV, ovarian serous cystadenocarcinoma; PAAD, pancreatic adenocarcinoma; PCPG, pheochromocytoma and paraganglioma; PRAD, prostate adenocarcinoma; READ, rectum adenocarcinoma; SARC, sarcoma; SKCM, skin cutaneous melanoma; STAD, stomach adenocarcinoma; TGCT, testicular germ cell tumors; THYM, thymoma; THCA, thyroid carcinoma; UCS, uterine carcinosarcoma; UCEC, uterine corpus endometrial carcinoma; UVM, uveal melanoma. B, Overall and progression-free survival in patients with MYC amplification and MYC overexpression (top and bottom quartiles). C, Correlation of MYC expression with gene expression of immune checkpoints PDL1 and CTLA4 (top) and with gene signature of immune-checkpoint blockade [ICB; Jiang et al. 2018 (28); bottom]. D, Expression of gene signatures associated with T helper cells 1 (Th1), Th17, and Th2 compared between tumors with (1) and without (0) MYC amplification. E, Expression of gene signatures associated with CD8 T cells, NK cells, and monocytes compared between tumors with (1) and without (0) MYC amplification. F, Immune deconvolution analysis showing a comparison of abundance of different immune subsets between tumors with (1) and without (0) MYC amplification. G, IHC showing CD8 T-cell infiltration in representative HCC tumors from the TCGA cohort with (1; n = 20) and without MYC amplification (0; n = 8). *, P < 0.05; ***, P < 0.001.

Figure 1.

The MYC oncogene drives human tumorigenesis associated with immune evasion. A, TumorMap rendering of the Pan-Cancer-Atlas of 33 cancers; tumors are defined by different tissues of origin in the first box. The second box shows the prevalence of MYC genomic amplification. Epithelial tumors with high rates of MYC amplification are annotated. The third box shows the gene expression of a signature derived from MYC amplification. LAML, acute myeloid leukemia; ACC, adrenocortical carcinoma; BLCA, bladder urothelial carcinoma; LGG, brain lower grade glioma; BRCA, breast invasive carcinoma; CESC, cervical squamous cell carcinoma and endocervical adenocarcinoma; CHOL, cholangiocarcinoma; LCML, chronic myelogenous leukemia; COAD, colon adenocarcinoma; ESCA, esophageal carcinoma; GBM, glioblastoma multiforme; HNSC, head and neck squamous cell carcinoma; KICH, kidney chromophobe; KIRC, kidney renal clear cell carcinoma; KIRP, kidney renal papillary cell carcinoma; LIHC, liver hepatocellular carcinoma; LUAD, lung adenocarcinoma; LUSC, lung squamous cell carcinoma; DLBC, lymphoid neoplasm diffuse large B, cell lymphoma; MESO, mesothelioma; OV, ovarian serous cystadenocarcinoma; PAAD, pancreatic adenocarcinoma; PCPG, pheochromocytoma and paraganglioma; PRAD, prostate adenocarcinoma; READ, rectum adenocarcinoma; SARC, sarcoma; SKCM, skin cutaneous melanoma; STAD, stomach adenocarcinoma; TGCT, testicular germ cell tumors; THYM, thymoma; THCA, thyroid carcinoma; UCS, uterine carcinosarcoma; UCEC, uterine corpus endometrial carcinoma; UVM, uveal melanoma. B, Overall and progression-free survival in patients with MYC amplification and MYC overexpression (top and bottom quartiles). C, Correlation of MYC expression with gene expression of immune checkpoints PDL1 and CTLA4 (top) and with gene signature of immune-checkpoint blockade [ICB; Jiang et al. 2018 (28); bottom]. D, Expression of gene signatures associated with T helper cells 1 (Th1), Th17, and Th2 compared between tumors with (1) and without (0) MYC amplification. E, Expression of gene signatures associated with CD8 T cells, NK cells, and monocytes compared between tumors with (1) and without (0) MYC amplification. F, Immune deconvolution analysis showing a comparison of abundance of different immune subsets between tumors with (1) and without (0) MYC amplification. G, IHC showing CD8 T-cell infiltration in representative HCC tumors from the TCGA cohort with (1; n = 20) and without MYC amplification (0; n = 8). *, P < 0.05; ***, P < 0.001.

Close modal

Next, we found that MYC activation influences the immune state of human cancers. First, the expression of MYC strongly correlated with the expression of the immune checkpoints PDL1 and CTLA4 (P = 9.4 × 10−64; Fig. 1C), with higher quartiles of MYC mRNA expression showing higher expression of both PDL1 and CTLA4 (Supplementary Fig. S2A). Second, MYC overexpression was associated with a gene signature predictive of T-cell dysfunction and resistance to immune-checkpoint blockade (9 = 3.7 × 10−91; Fig. 1C; ref. 28). Third, MYC-amplified tumors and MYC-overexpressing tumors had lower expression of antitumor Th1 and Th17 signatures and higher expression of protumorigenic Th2 immune signature (P < 0.001; Fig. 1D; Supplementary Fig. S2B). Fourth, MYC-amplified tumors had lower infiltration with antitumor immune subsets such as CD8+ T cells, NK cells, and monocytes (all P < 0.001), with no major differences in other subsets, upon immune deconvolution analysis (Fig. 1E and F). Finally, we examined if MYC overexpression in HCC was associated with immune evasion because MYC is frequently amplified and overexpressed in HCC (Supplementary Fig. S3A–S3B) and HCC exhibits limited responsiveness to immune-checkpoint blockade (13, 14). Indeed, by IHC, we found that MYC amplification was associated with lower CD8+ T-cell infiltration in HCC (Fig. 1G). Thus, MYC overexpression is common in human cancer and is associated with a poor prognosis. Moreover, MYC-driven tumors are immune evasive and less likely to respond to immune-checkpoint blockade.

MYC causes reversible suppression of antitumor immune response

To examine how MYC suppresses the immune response, we used an MYC-driven conditional transgenic mouse model of HCC (21). The transcriptional profile of our transgenic mouse model of MYC-HCC was similar to the transcriptome to multiple human HCC cohorts including 10 human HCC data sets and the TCGA data (similarity z-score 27.2, P = 1 × 10−15; Fig. 2A). Specifically, there was an overlap between mouse MYC-HCC transcriptome and the human HCC immune signatures of inactivated 41BB PD1+ T cells in human HCC (29; similarity z-score 3.2, P = 1 × 10−9) and exhausted PD1+ CD8 T cells (similarity z-score 16.7, P = 1 × 10−21; Fig. 2B; ref. 30). Cross-species analysis by classification of human HCC in the TCGA cohort using a 56-gene orthologous signature derived from the mouse model of MYC-HCC (mMYCOn) found that 16% (n = 58/373) of human HCC showed strong enrichment on the mMYCOn gene signature (Fig. 2C). We confirmed the mMYCOn. signature did indeed identify MYC-driven human HCCs by its association with MYC molecular pathway activation (NES = 2.0, P = 0.001) and strong enrichment for MYC gene amplification (P < 0.001; Fig. 2D). Human HCC with high expression of mMYCOn signature showed enrichment of a gene program expressed by cancer cells associated with T-cell exclusion and resistance to anti–PD-1 checkpoint therapy (31), and also showed strong upregulation of CTLA4 (Fig. 2E). Thus, the mouse MYC-HCC and MYC-driven human HCC exhibit similar genetic and immune profiles associated with exhausted T cells, and resistance to immune-checkpoint monotherapy.

Figure 2.

MYC causes reversible transcriptional repression of antitumor immune responses. A, Heat map shows similarity between MYC-driven murine HCC (MYC-On, n = 3 vs. MYC-Off, n = 3) and 11 human HCC transcriptome data sets including the TCGA cohort; the top 10 upregulated and downregulated genes with enrichment z-scores across the different data sets are shown. Bar plot at the bottom shows log of P value of similarity between the individual cohorts and murine MYC-HCC. B, Heat map shows z-scores of enrichment of canonical pathways and biological functions in murine MYC-HCC transcriptome and the human HCC immune signatures of exhausted PD1+ CD8 T cells (30) and inactivated 41BB PD1+ T cells in human HCC (29). C, Gene signature derived from murine MYC-HCC used to classify human HCC in the TCGA cohort to identify MYC-driven human HCCs. D, Human HCCs enriched in mMYCOn signature show MYC molecular pathway activation and strong association with MYC gene amplification. E, Human HCCs with high expression of mMYCOn signature showed enrichment of a gene program of T-cell exclusion and resistance to anti–PD-1 checkpoint therapy (31), and upregulation of CTLA4.F, Principal component analysis (PCA) plot comparing the transcriptional profile of a transgenic model of HCC overexpressing wild-type MYC (MYCOn, n = 16), a mutant MYC defective in Miz1 binding (V394D; MYCVd, n = 11), and tumors upon MYC inactivation for a short time point of 16 hours (MYCOff, n = 8; ref. 35) to identify MYC/Miz-1–repressed genes. G, Heat map shows genes that are reversibly repressed by MYC in the MHC protein complex, complement system, and cytokines. H, IHC shows the abundance of macrophages (F4/80+) by day 1 and CD4 T-cell recruitment by day 4 of MYC inactivation in HCC. I, Immune deconvolution analysis showing comparison of abundance of different immune subsets in murine MYC-HCC upon MYC inactivation. *, P < 0.05; **, P < 0.01; ***, P < 0.001; ****, P < 0.0001; ns, nonsignificant. (C, Created with BioRender.com.)

Figure 2.

MYC causes reversible transcriptional repression of antitumor immune responses. A, Heat map shows similarity between MYC-driven murine HCC (MYC-On, n = 3 vs. MYC-Off, n = 3) and 11 human HCC transcriptome data sets including the TCGA cohort; the top 10 upregulated and downregulated genes with enrichment z-scores across the different data sets are shown. Bar plot at the bottom shows log of P value of similarity between the individual cohorts and murine MYC-HCC. B, Heat map shows z-scores of enrichment of canonical pathways and biological functions in murine MYC-HCC transcriptome and the human HCC immune signatures of exhausted PD1+ CD8 T cells (30) and inactivated 41BB PD1+ T cells in human HCC (29). C, Gene signature derived from murine MYC-HCC used to classify human HCC in the TCGA cohort to identify MYC-driven human HCCs. D, Human HCCs enriched in mMYCOn signature show MYC molecular pathway activation and strong association with MYC gene amplification. E, Human HCCs with high expression of mMYCOn signature showed enrichment of a gene program of T-cell exclusion and resistance to anti–PD-1 checkpoint therapy (31), and upregulation of CTLA4.F, Principal component analysis (PCA) plot comparing the transcriptional profile of a transgenic model of HCC overexpressing wild-type MYC (MYCOn, n = 16), a mutant MYC defective in Miz1 binding (V394D; MYCVd, n = 11), and tumors upon MYC inactivation for a short time point of 16 hours (MYCOff, n = 8; ref. 35) to identify MYC/Miz-1–repressed genes. G, Heat map shows genes that are reversibly repressed by MYC in the MHC protein complex, complement system, and cytokines. H, IHC shows the abundance of macrophages (F4/80+) by day 1 and CD4 T-cell recruitment by day 4 of MYC inactivation in HCC. I, Immune deconvolution analysis showing comparison of abundance of different immune subsets in murine MYC-HCC upon MYC inactivation. *, P < 0.05; **, P < 0.01; ***, P < 0.001; ****, P < 0.0001; ns, nonsignificant. (C, Created with BioRender.com.)

Close modal

Next, we examined the immune-related transcriptional changes induced upon MYC inactivation in the mouse model of MYC-driven HCC (MYC-On (n = 3) vs. Off (4 days, n = 3; DE genes 756 upregulated, 1,783 downregulated; Padj < 0.05; Supplementary Table S3). Inactivating MYC in the MYC-HCC resulted in the upregulation of antitumor immune pathways including the Il6, Tnfa, and Ifng pathways, and activation of inflammatory biological processes such as phagosome formation, production of ROS by macrophages, and phagocyte migration (Supplementary Fig. S4). This led us to hypothesize that MYC transcriptionally represses macrophage-mediated antitumor immunity.

To test this hypothesis, we further analyzed transcriptional data from a transgenic model of HCC expressing a mutant MYC defective in Miz1 binding (V394D; ref. 32), because interaction with Miz1 is a well-characterized mechanism of transcriptional repression by MYC (33, 34). We compared gene expression between HCC with wild-type MYC overexpression (MYCOn, n = 16), HCC with MYC-V394D-mutant (MYCVd, n = 11), and tumors upon MYC inactivation for a short time point of 16 hours (MYCOff, n = 8; Fig. 2F; ref. 35). Genes that were overexpressed in tumors expressing MYC Vd-mutant vs. MYC-WT HCC, and also upregulated upon short-term MYC inactivation were identified as the ones specifically transcriptionally repressed by the MYC/Miz1 interaction (n = 1028, Padj < 0.05, Fig. 2F; Supplementary Table S4). Pathway analysis of the MYC/Miz1 repressed gene program revealed enrichment of several critical immune-related pathways particularly involving antigen presentation, MHC I protein complex, and T-cell–mediated cytotoxicity (Supplementary Fig. S5A–S5B). Multiple genes related to antigen presentation, particularly belonging to the MHCI protein complex pathway such as Cd74, B2 m, and MHC I genes were found to be strongly transcriptionally repressed by the MYC-Miz1 complex (Fig. 2G). Even short-term MYC inactivation was noted to activate the innate immune response in a Miz1-dependent manner with upregulation of multiple components of the complement pathway (C2, C3, C4a, C4b, C8, and C9) and cytokines that recruit innate immune cells Il1b, Ccl15, Cxcl10, and Cxcl16 (Fig. 2G). Thus, we show that MYC interacts with Miz1 to transcriptionally repress genes and programs specifically essential to invoke antitumor innate immunity.

To confirm the activation of antitumor immune responses upon MYC inactivation, we performed a temporal analysis of immune changes in the tumor microenvironment. We found that MYC inactivation was associated with early macrophage recruitment by day 1, especially in the peritumoral regions (Fig. 2H). This was followed by the subsequent infiltration of CD4T cells by day 4 (Fig. 2H). These findings are consistent with our earlier work showing that CD4+ T cells play a causal role in tumor regression upon oncogene inactivation (36). Additionally, immune deconvolution analysis of transcriptional changes noted at day 4 of MYC inactivation demonstrated recruitment of proinflammatory M1 macrophages and CD4+ T cells, along with a decrease in the proportion of protumorigenic immune cells such as regulatory T cells (Treg; Fig. 2I).

Thus, the mouse MYC-HCC and human HCC exhibit activation of gene programs of immune evasion. Further, in the mouse model, MYC inactivation reverses immune evasion by activating innate and adaptive antitumor immunity.

Combined anti-PDL1 and anti-CTLA4 overcome MYC-induced immune evasion

We evaluated if inhibiting the immune checkpoints PDL1 and/or CTLA4 could overcome MYC-induced immune evasion (Fig. 3A). In mouse MYC-HCC, only mice treated with both anti-PDL1 and anti-CTLA4, but not either alone, showed significantly delayed progression on 3D tumor volume assessment by MRI compared with mice treated with control antibody [mean tumor volume (858.31 μm3, SEM = 53.60) vs. (1,897.67 μm3, SEM = 258.62); P = 0.004; Fig. 3B and C]. Mice treated with combined PDL1 and CTLA4 blockade (n = 5) had smaller and fewer liver tumors than mice treated with control antibody [n = 5 (858.31 μm3, SEM = 53.60) vs. (1,897.67 μm3, SEM = 258.62); P = 0.004; Fig. 3D and E]. However, there were no differences in final tumor volumes of mice treated with anti-PDL1 (n = 5; 1,235.81 μm3, SEM = 288.87) or anti-CTLA4 alone (n = 5; 1,922.64 μm3, SEM = 509.45) compared with IgG-treated mice (P = 0.126; P = 0.966, respectively; Fig. 3D and E). The combination therapy led to a decrease in the proliferative index of the tumor as measured by phosphohistone 3 staining without affecting MYC expression levels in the tumor cells (Supplementary Fig. S6), and also led to increased cell death as shown by cleaved caspase-3 staining (Supplementary Fig. S7). Thus, only combined, but not individual, inhibition of PDL1 and CTLA4, impeded tumor progression in MYC-HCC.

Figure 3.

Combined anti-PDL1 and anti-CTLA4 delay tumor progression in MYC-HCC. A, Experimental scheme of MYC-HCC treatment with IgG control (n = 5), PDL1 antibody (n = 5), CTLA4 antibody (n = 5), or their combination (n = 5). B, Weekly MRI showing tumor progression in representative MYC-HCC mice in the four treatment groups (n = 5 each group). C, Quantification of volumetric tumor measurement using MRI of MYC-HCC mice in the four treatment groups (n = 5 each group). D, End-of-treatment gross appearance and histology of representative MYC-HCC mice in the four treatment groups (n = 5 each group). H&E, hematoxylin and eosin. E, Quantification of liver tumor burden at end of treatment of MYC-HCC mice in the four treatment groups (n = 5 each group). **, P < 0.01; ***, P < 0.001; ns, nonsignificant. (A, Created with BioRender.com.)

Figure 3.

Combined anti-PDL1 and anti-CTLA4 delay tumor progression in MYC-HCC. A, Experimental scheme of MYC-HCC treatment with IgG control (n = 5), PDL1 antibody (n = 5), CTLA4 antibody (n = 5), or their combination (n = 5). B, Weekly MRI showing tumor progression in representative MYC-HCC mice in the four treatment groups (n = 5 each group). C, Quantification of volumetric tumor measurement using MRI of MYC-HCC mice in the four treatment groups (n = 5 each group). D, End-of-treatment gross appearance and histology of representative MYC-HCC mice in the four treatment groups (n = 5 each group). H&E, hematoxylin and eosin. E, Quantification of liver tumor burden at end of treatment of MYC-HCC mice in the four treatment groups (n = 5 each group). **, P < 0.01; ***, P < 0.001; ns, nonsignificant. (A, Created with BioRender.com.)

Close modal

Treatment with combination therapy did not result in weight loss or hepatotoxicity in mice (Fig. 4A). End-of-treatment analysis revealed slight lymphopenia with individual or combined treatments. Treatment with anti-CTLA4 alone and in combination with anti-PDL1 resulted in lower peripheral blood monocyte count (Fig. 4B). Histologic evaluation of liver, colon, and lungs by a blinded pathologist showed no areas of severe autoimmune organ damage with the combination therapy even though IHC analysis revealed higher levels of CD8+ T-cell infiltration with combination therapy (Fig. 4C). Thus, the combination of anti-PDL1 and anti-CTLA4 shows synergistic efficacy in overcoming MYC-induced immune evasion without severe toxicity.

Figure 4.

Safety profile of combined anti-PDL1 and anti-CTLA in MYC-HCC. A, Body weight and liver tests at end of treatment of MYC-HCC bearing mice treated with either IgG control (n = 5) or PDL1 antibody (n = 5) or CTLA4 antibody (n = 5) or their combination (n = 5). B, Peripheral white cell counts at end of treatment of MYC-HCC–bearing mice treated with either IgG control (n = 5) or PDL1 antibody (n = 5) or CTLA4 antibody (n = 5) or their combination (n = 5). C, Histology and CD8 T-cell infiltration in normal surrounding liver and colon tissue of mice treated with the four treatment groups. *, P < 0.05; **, P < 0.01; ns, nonsignificant.

Figure 4.

Safety profile of combined anti-PDL1 and anti-CTLA in MYC-HCC. A, Body weight and liver tests at end of treatment of MYC-HCC bearing mice treated with either IgG control (n = 5) or PDL1 antibody (n = 5) or CTLA4 antibody (n = 5) or their combination (n = 5). B, Peripheral white cell counts at end of treatment of MYC-HCC–bearing mice treated with either IgG control (n = 5) or PDL1 antibody (n = 5) or CTLA4 antibody (n = 5) or their combination (n = 5). C, Histology and CD8 T-cell infiltration in normal surrounding liver and colon tissue of mice treated with the four treatment groups. *, P < 0.05; **, P < 0.01; ns, nonsignificant.

Close modal

Dual targeting of PDL1 and CTLA4 restores macrophage-mediated antitumor immunity

Next, we evaluated how dual targeting of anti-PDL1 and CTLA4 remodeled the immune microenvironment of the tumors. IHC analysis showed that the combination of PDL1 and CTLA4 blockade elicited the recruitment of a higher number of CD4+ T cells (P = 0.005) and CD8+ T cells (P < 0.0001) than individual monotherapies (Fig. 5A). Interestingly, we observed significantly increased macrophage infiltration in tumors treated with the combination of PDL1 and CTLA4 compared with IgG-treated mice (P = 0.005) but not in mice treated with anti-PDL1 (P = 0.31) or anti-CTLA4 (P = 0.06) as monotherapy (Fig. 5A). Despite an overall increase in macrophage infiltration, the proportion of immunosuppressive PDL1+ macrophages was lower in mice treated with combined anti-PDL1 and anti-CTLA4 compared with mice treated with IgG control antibody (P = 0.001; Supplementary Fig. S8). Thus, dual targeting of PDL1 and CTLA4 elicits a more robust T-cell immune response and macrophage infiltration than either monotherapy.

Figure 5.

Dual targeting of PDL1 and CTLA4 restores macrophage-mediated antitumor immunity. A, IHC for CD4+ T cells, CD8+ T cells, and F4/80+ macrophages in MYC-HCC–bearing mice treated with IgG control (n = 5), PDL1 antibody (n = 5), CTLA4 antibody (n = 5), or their combination (n = 5). Boxplots show the quantification of cell counts. B, Violin plot showing differentially expressed genes between MYC-HCC treated with combination therapy (n = 3) versus control IgG (n = 3). Macrophage-related genes are upregulated with treatment with anti-PDL1 + CTLA4. C, Pathway analysis of differentially expressed genes between MYC-HCC treated with combination therapy (n = 3) versus control IgG (n = 3). D, Enrichment of inflammatory signature of macrophages stimulated with lipopolysaccharide (LPS) in tumors treated with combination therapy (n = 3) versus control IgG (n = 3). E, Pathways analysis of meta-analysis comparing the transcriptional changes induced by the combination therapy (n = 3) with those induced by anti PDL1 (n = 3) or CTLA4 monotherapies (n = 3). *, P < 0.05; ***, P < 0.001; ns, nonsignificant.

Figure 5.

Dual targeting of PDL1 and CTLA4 restores macrophage-mediated antitumor immunity. A, IHC for CD4+ T cells, CD8+ T cells, and F4/80+ macrophages in MYC-HCC–bearing mice treated with IgG control (n = 5), PDL1 antibody (n = 5), CTLA4 antibody (n = 5), or their combination (n = 5). Boxplots show the quantification of cell counts. B, Violin plot showing differentially expressed genes between MYC-HCC treated with combination therapy (n = 3) versus control IgG (n = 3). Macrophage-related genes are upregulated with treatment with anti-PDL1 + CTLA4. C, Pathway analysis of differentially expressed genes between MYC-HCC treated with combination therapy (n = 3) versus control IgG (n = 3). D, Enrichment of inflammatory signature of macrophages stimulated with lipopolysaccharide (LPS) in tumors treated with combination therapy (n = 3) versus control IgG (n = 3). E, Pathways analysis of meta-analysis comparing the transcriptional changes induced by the combination therapy (n = 3) with those induced by anti PDL1 (n = 3) or CTLA4 monotherapies (n = 3). *, P < 0.05; ***, P < 0.001; ns, nonsignificant.

Close modal

To identify mechanisms by which combination therapy is effective in MYC-HCC, we evaluated the transcriptomic differences between the tumors treated with the four different therapies using a targeted immune-oncology RNA-seq panel. We identified 87 genes that were differentially expressed (DEG) between tumors treated with the anti-PDL1 + CTLA4 combination therapy versus control antibody (Padj < 0.05; Fig. 5B) with upregulation of multiple genes in the innate immune response including Cd84, Cd74, and S100A9 and upregulation of MHC II genes (Fig. 5B). Pathway analysis of the DEG revealed that apart from the activation of the interferon gamma pathway, as expected (Fig. 5C), a top pathway upregulated in tumors treated with anti PDL1 + CTLA4 was macrophage activation, with enrichment of a signature associated with inflammatory macrophages stimulated with lipopolysaccharide (LPS; Fig. 5D; ref. 37).

Next, we performed a meta-analysis comparing the transcriptional changes induced by the combination therapy to those induced by anti-PDL1 or CTLA4 monotherapies (Supplementary Fig. S9A). The pathways that were preferentially upregulated in the MYC-HCC treated with combination therapy included innate immune response, antigen processing, and presentation and MHC class II pathway (Fig. 5E; Supplementary Fig. S9B). Crucial genes in the antigen presentation pathway were preferentially upregulated in tumors treated with combination therapy including B2m, H2-Aa, H2-Ab1, Cd74, and Tap1. Moreover, multiple macrophage-related genes such as Mmp13 and Atf3 and monocyte chemoattractants such as Ccl5 and Ccl8 were differentially upregulated in tumors treated with combination therapy (Supplementary Fig. S9C). Thus, the combination therapy leads to differential activation of inflammatory macrophages and upregulation of the antigen presentation pathway, which is transcriptionally repressed by MYC.

Combined PDL1 and CTLA4 inhibition elicits early repolarization of myeloid cells

To determine the sequence of macrophage and T-cell activation with immune-checkpoint inhibition, we evaluated immune responses at an earlier time point of 1 week after treatment with PDL1 and/or CTLA4 inhibitors. We used high-dimensional single-cell profiling of the tumor immune infiltrates with mass cytometry (CyTOF; Fig. 6A). After dimensionality reduction of the high-order parametric data, 27 unique immune subsets were identified (Fig. 6B; Supplementary Fig. S10). The most significant changes in the mice treated with combined anti-PDL1 and anti-CTLA4, compared with the individual monotherapies, were found in the myeloid compartment. The frequency of proinflammatory macrophages (Ly6CHigh/CCR2High) was significantly increased, and immunosuppressive macrophages (PDL1+/Ly6CLow/CCR2Low) was significantly decreased in the tumors treated with both anti-PDL1 and anti-CTLA4, and not either alone, compared with mice treated with IgG control (Fig. 6C and D). We further confirmed this on immunofluorescent staining, which showed higher M1-like CCR2+ macrophages at the outer edges of the tumors (Supplementary Fig. S11) with depletion of intratumoral M2-like CD206+ macrophages (Supplementary Fig. S12)

Figure 6.

Mass cytometry reveals a unique early immune response to dual targeting of PDL1 and CTLA4. A, Experimental scheme of MYC-HCC treatment with IgG control (n = 3), PDL1 antibody (n = 3), CTLA4 antibody (n = 2), or their combination (n = 4) for 1 week, followed by mass cytometry analysis. B, tSNE plot grouped by major immune subsets identified in the tumor immune microenvironment. C, tSNE plot showing the visual representation of myeloid marker expression between MYC-HCC treated with IgG control (n = 3), PDL1 antibody (n = 3), CTLA4 antibody (n = 2), or their combination (n = 4). D, Boxplots show quantification of CCR2+/Ly6High M1-like macrophages, Ly6C expression in the macrophages, abundance of PDL1+/Ly6CLow/CCR2Low M2-like macrophages, and myeloid-derived suppressor cells (MDSC) in MYC-HCC treated with IgG control (n = 3), PDL1 antibody (n = 3), CTLA4 antibody (n = 2), or their combination (n = 4), respectively. E, Boxplots show quantification of CD40 expression in different myeloid subsets in MYC-HCC treated with IgG control (n = 3), PDL1 antibody (n = 3), CTLA4 antibody (n = 2), or their combination (n = 4), respectively. F, Boxplots show quantification of MHCII expression in CD39+ and PDL1+ M2-like macrophages in MYC-HCC treated with IgG control (n = 3), PDL1 antibody (n = 3), CTLA4 antibody (n = 2), or their combination (n = 4), respectively. *, P < 0.05; ***, P < 0.001; #, P < 0.1; ns, nonsignificant. (A, Created with BioRender.com.)

Figure 6.

Mass cytometry reveals a unique early immune response to dual targeting of PDL1 and CTLA4. A, Experimental scheme of MYC-HCC treatment with IgG control (n = 3), PDL1 antibody (n = 3), CTLA4 antibody (n = 2), or their combination (n = 4) for 1 week, followed by mass cytometry analysis. B, tSNE plot grouped by major immune subsets identified in the tumor immune microenvironment. C, tSNE plot showing the visual representation of myeloid marker expression between MYC-HCC treated with IgG control (n = 3), PDL1 antibody (n = 3), CTLA4 antibody (n = 2), or their combination (n = 4). D, Boxplots show quantification of CCR2+/Ly6High M1-like macrophages, Ly6C expression in the macrophages, abundance of PDL1+/Ly6CLow/CCR2Low M2-like macrophages, and myeloid-derived suppressor cells (MDSC) in MYC-HCC treated with IgG control (n = 3), PDL1 antibody (n = 3), CTLA4 antibody (n = 2), or their combination (n = 4), respectively. E, Boxplots show quantification of CD40 expression in different myeloid subsets in MYC-HCC treated with IgG control (n = 3), PDL1 antibody (n = 3), CTLA4 antibody (n = 2), or their combination (n = 4), respectively. F, Boxplots show quantification of MHCII expression in CD39+ and PDL1+ M2-like macrophages in MYC-HCC treated with IgG control (n = 3), PDL1 antibody (n = 3), CTLA4 antibody (n = 2), or their combination (n = 4), respectively. *, P < 0.05; ***, P < 0.001; #, P < 0.1; ns, nonsignificant. (A, Created with BioRender.com.)

Close modal

CD40 expression on myeloid cells is known to activate the tumoricidal activities of macrophages (38, 39). The combination therapy, but not either anti-PDL1 or anti-CTLA4 alone, significantly increased CD40 expression on multiple myeloid cells including inflammatory macrophages (Ly6CHigh/CCR2High), conventional dendritic cells (CD11c+/CD11b/CD103+), myeloid dendritic cells (CD11c+/CD11b+/CD103), and MDSCs (Fig. 6CE). On the other hand, immunosuppressive macrophage subsets such as PDL1+ TAMs and CD39+ TAMs showed significantly increased expression of MHC II with combination anti-PDL1 and anti-CTLA4, but not with individual monotherapies (Fig. 6CF). We did find a trend toward increased effector CD8+ T cells (LyCHigh/PD1Low) with combination therapy but no other significant differences in the proportion of T-cell subpopulations at this early stage of treatment (Supplementary Fig. S13A–S13B). There were no significant changes in other innate immune cells such as NK cells, eosinophils, or neutrophils (Supplementary Fig. S13A–S13B). Thus, we show that combined blockade of both PDL1 and CTLA4, and not either alone, elicits an early immune response with the expansion of the proinflammatory myeloid compartments and repolarization to an antitumor profile with upregulation of CD40 and MHC II.

Macrophages are essential for the antitumor efficacy of immune-checkpoint therapy

To evaluate the causal role of macrophages, we depleted macrophages prior to treatment with combined PDL1 and CTLA4 blockade (Fig. 7A). We used the anti-CSF1R antibodies to deplete macrophages (24, 25), after first confirming that the CSF1R antibody effectively depleted macrophages, without impacting tumor progression in MYC-driven HCC (Supplementary Fig. S14A–S14C).

Figure 7.

Macrophages are essential for the antitumor efficacy of combined immune-checkpoint therapy in MYC-HCC. A, Experimental scheme of macrophage depletion in MYC-HCC followed by treatment with either IgG control (n = 4) or dual PDL1 and CTLA4 antibodies (n = 3). B, Quantification of tumor burden at end of treatment with either IgG control (n = 4) or dual PDL1 and CTLA4 antibodies (n = 3) in macrophage-depleted MYC-HCC mice. C, End-of-treatment gross appearance, histology, and immunofluorescence for CD8 T cells in representative macrophage-depleted MYC-HCC–bearing mice treated with IgG control (n = 4) or dual PDL1 and CTLA4 antibodies (n = 3). H&E, hematoxylin and eosin. D, Quantitation of CD8 T-cell infiltration in macrophage-depleted MYC-HCC–bearing mice treated with IgG control (n = 4) or dual PDL1 and CTLA4 antibodies (n = 3). E, Figurative representation of the mechanism of efficacy of combined PDL1 and CTLA4 therapy in MYC-HCC. MYC-driven cancers are immune evasive with MHCI repression and PDL1 overexpression on cancer cells and macrophages, which in turn lead to T-cell exhaustion. Treatment with CTLA4 and PDL1 inhibitors leads to repolarization of macrophages to the M1-like phenotype, with increased expression of CD40 and MHCII, which leads to enhanced antigen presentation and robust T-cell activation, resulting in delayed tumor progression. ns, nonsignificant. (A and E, Created with BioRender.com.)

Figure 7.

Macrophages are essential for the antitumor efficacy of combined immune-checkpoint therapy in MYC-HCC. A, Experimental scheme of macrophage depletion in MYC-HCC followed by treatment with either IgG control (n = 4) or dual PDL1 and CTLA4 antibodies (n = 3). B, Quantification of tumor burden at end of treatment with either IgG control (n = 4) or dual PDL1 and CTLA4 antibodies (n = 3) in macrophage-depleted MYC-HCC mice. C, End-of-treatment gross appearance, histology, and immunofluorescence for CD8 T cells in representative macrophage-depleted MYC-HCC–bearing mice treated with IgG control (n = 4) or dual PDL1 and CTLA4 antibodies (n = 3). H&E, hematoxylin and eosin. D, Quantitation of CD8 T-cell infiltration in macrophage-depleted MYC-HCC–bearing mice treated with IgG control (n = 4) or dual PDL1 and CTLA4 antibodies (n = 3). E, Figurative representation of the mechanism of efficacy of combined PDL1 and CTLA4 therapy in MYC-HCC. MYC-driven cancers are immune evasive with MHCI repression and PDL1 overexpression on cancer cells and macrophages, which in turn lead to T-cell exhaustion. Treatment with CTLA4 and PDL1 inhibitors leads to repolarization of macrophages to the M1-like phenotype, with increased expression of CD40 and MHCII, which leads to enhanced antigen presentation and robust T-cell activation, resulting in delayed tumor progression. ns, nonsignificant. (A and E, Created with BioRender.com.)

Close modal

There was no difference in liver tumor burden between macrophage-depleted mice treated with combination PDL1 and CTLA4 inhibition (n = 3) or control antibody (n = 4; liver/body ratio 0.25 vs. 0.17, P = 0.23; Fig. 7B). Further, we confirmed that the combined blockade of PDL1 and CTLA4 was not able to elicit antitumor CD8 T-cell response in macrophage-depleted mice (Fig. 7C and D). Together, these data show that macrophage-depletion blocks the antitumor efficacy of combined PDL1 and CTLA4 inhibition in MYC-HCC. Thus, our data demonstrate that combined PDL1 and CTLA4 blockade requires the activation of macrophages to overcome MYC-induced immune evasion (Fig. 7E).

We show that MYC is involved in immune evasion during HCC tumorigenesis, which can be overcome through the combined blockade of immune checkpoints PDL1 and CTLA4. A pan-cancer analysis of 33 human tumors reveals that not only is MYC commonly amplified and overexpressed, but it is also associated with increased immune-checkpoint expression, immunotherapy resistance, a Th2-like immune profile, and reduced antitumor CD8 T-cell infiltration. We demonstrate that MYC transcriptionally represses MHC class I genes, thus impeding antigen presentation in a murine model of MYC-driven HCC. Notably, MYC immune evasion can be abrogated through combined immune-checkpoint blockade with anti-PDL1 and CTLA4, which was associated with the causally required activation of macrophages to a proinflammatory phenotype with enhanced antigen presentation. The restoration of immune responsiveness was associated with increased expression of CD40 and MHCII on antigen-presenting cells (APC) and robust antitumor T-cell responses (Fig. 7E). We conclude that MYC is a key driver of immune evasiveness in human cancer, blocking both innate and adaptive antitumor immune responses. Further, we identify MYC-driven HCC as a specific subset of cancers that are unlikely to respond to monotherapy with ICI and need combination immunotherapy to achieve tumor remission.

The MYC oncogene has been shown to play a pivotal role in remodeling the tumor microenvironment (4, 7, 36, 40, 41). In transgenic mouse models of MYC-driven pancreatic and lung cancer, inactivation of MYC led to rapid efflux of M2-like CD206+ macrophages with the recruitment of T cells (40, 41). However, in a different mouse model of pancreatic adenocarcinoma MYC inactivation did not influence macrophage infiltration, but instead reprogrammed macrophages to secrete Cxcl13, leading to the recruitment of B and NK cells (42). We show here that MYC reversibly transcriptionally repressed several crucial genes involved in antigen presentation including MHC class I genes thus handicapping both innate and adaptive immunity. MYC inactivation derepressed MHC I expression and also led to the secretion of monocyte chemoattractants such as IL1β (43, 44) and CXCL10 (45) associated with rapid recruitment of M1-like macrophages to the peritumoral areas and subsequent recruitment of CD4 T cells. In this context, we show that combined inhibition of PDL1 and CTLA4 in MYC-HCC led to the loss of M2-like CD206+ macrophages, recruitment of CCR2+ inflammatory macrophages, and increased expression of CD40 and MHC II on multiple proinflammatory myeloid cells, thus relieving the MYC-induced block on antigen presentation and resulting in enhanced T-cell cytotoxicity.

Inhibition of immune checkpoints such as PDL1 or CTLA4 has shown promising efficacy in many cancers (46, 47). However, less than a third of patients respond to ICI monotherapy, making it critical to understand mechanisms of resistance to immunotherapy. Although the role of MYC in driving immune evasion has been demonstrated by previous studies (4, 5, 40, 48, 49), its functional relevance in modulating response to immune-checkpoint inhibition is not understood. Our study identifies MYC oncogene amplification or overexpression as a major driver of tumor-intrinsic resistance to immune-checkpoint inhibition. We demonstrate that MYC-driven HCC exhibits primary resistance to monotherapy with PDL1 or CTLA4 blockade, implying that additional aspects of the immune response are modulated by MYC. These results imply that MYC amplification or MYC overexpression can potentially serve as a biomarker to prioritize patients to be considered for combination immunotherapies, rather than individual checkpoint inhibitors.

Our results identify a unique mechanistic basis for the synergistic efficacy of combined PDL1 and CTLA4 blockade, beyond just additive effects of their individual mechanisms. We suggest a model whereby the combination of anti-PDL1 and CTLA4 restores immune responsiveness by the activation of proinflammatory CCR2+ macrophages, which we show are causally required, as macrophage depletion abrogated the antitumor efficacy of PDL1 and CTLA4 blockade (Fig. 7E). These macrophages appear to be involved in the subsequent recruitment of adaptive immune cells including CD4+ T cells and cytotoxic CD8+ T cells. We suggest that the upregulation of at least 2 ligands on APCs, CD40, and MHCII, is involved in the mechanism of activating adaptive immunity. CD40 expression is known to license APCs to activate cytotoxic T cells and invigorate the tumoricidal activity of macrophages (50). Increased MHCII expression enables efficient antigen presentation by APCs to CD4+ T cells (51). The upregulation of both CD40 and MHCII on APCs, seen uniquely with combined PDL1 and CTLA4 blockade and not with the monotherapies, could be a key mechanism that contributes to the antitumor efficacy of the combination.

Finally, no existing therapies that directly target the MYC oncogene exist. Several strategies to indirectly target MYC are currently under evaluation (52, 53), and our data suggest that such MYC inhibitors will potentially synergize with immunotherapy. In the existing treatment paradigm, our results show that combined PDL1 and CTLA4 inhibition may be effective in restoring macrophage-mediated antitumor immune response in MYC-driven human HCC.

No disclosures were reported.

R. Dhanasekaran: Conceptualization, resources, data curation, formal analysis, funding acquisition, investigation, visualization, methodology, writing–original draft, writing–review and editing. A.S. Hansen: Data curation, formal analysis, visualization, writing–review and editing. J. Park: Data curation, formal analysis, investigation. L. Lemaitre: Data curation, formal analysis, visualization, writing–review and editing. I. Lai: Data curation, formal analysis, investigation. N. Adeniji: Formal analysis, investigation, visualization. S. Kuruvilla: Formal analysis, validation, investigation. A. Suresh: Investigation, visualization, writing–review and editing. J. Zhang: Data curation, formal analysis, writing–review and editing. V. Swamy: Formal analysis, investigation, visualization. D.W. Felsher: Conceptualization, resources, project administration, writing–review and editing.

R. Dhanasekaran was supported by NIH grant CA222676 from the NCI and the American College of Gastroenterology Junior Faculty Career Development Grant. D.W. Felsher was supported by NIH grants CA208735 and CA253180 from the NCI. The authors thank Pauline Chu for help with mouse histology services and Rachana Baskar for help with IHC. Human HCC TCGA samples were obtained from the Mayo Clinic Hepatobiliary SPORE support (Award Number P50 CA210964) from the NCI.

The publication costs of this article were defrayed in part by the payment of publication fees. Therefore, and solely to indicate this fact, this article is hereby marked “advertisement” in accordance with 18 USC section 1734.

Note: Supplementary data for this article are available at Cancer Research Online (http://cancerres.aacrjournals.org/).

1.
Schaub
FX
,
Dhankani
V
,
Berger
AC
,
Trivedi
M
,
Richardson
AB
,
Shaw
R
, et al
.
Pan-cancer alterations of the MYC oncogene and its proximal network across the cancer genome atlas
.
Cell Syst
2018
;
6
:
282
300
.
2.
Dhanasekaran
R
,
Deutzmann
A
,
Mahauad-Fernandez
WD
,
Hansen
AS
,
Gouw
AM
,
Felsher
DW
, et al
.
The MYC oncogene: the grand orchestrator of cancer growth and immune evasion
.
Nat Rev Clin Oncol
2022
;
19
:
23
36
.
3.
Rakhra
K
,
Bachireddy
P
,
Zabuawala
T
,
Zeiser
R
,
Xu
L
,
Kopelman
A
, et al
.
CD4 T cells contribute to the remodeling of the microenvironment required for sustained tumor regression upon oncogene inactivation
.
Cancer Cell
2010
;
18
:
485
98
.
4.
Casey
SC
,
Tong
L
,
Li
Y
,
Do
R
,
Walz
S
,
Fitzgerald
KN
, et al
.
MYC regulates the antitumor immune response through CD47 and PD-L1
.
Science
2016
;
352
:
227
31
.
5.
Xu
Y
,
Poggio
M
,
Jin
HY
,
Shi
Z
,
Forester
CM
,
Wang
Y
, et al
.
Translation control of the immune-checkpoint in cancer and its therapeutic targeting
.
Nat Med
2019
;
25
:
301
11
.
6.
MYC-mediated translation of PD-L1 promotes liver cancer immune escape
.
Cancer Discov
2019
;
9
:
317
.
7.
Swaminathan
S
,
Hansen
AS
,
Heftdal
LD
,
Dhanasekaran
R
,
Deutzmann
A
,
Fernandez
WDM
, et al
.
MYC functions as a switch for natural killer cell-mediated immune surveillance of lymphoid malignancies
.
Nat Commun
2020
;
11
:
2860
.
8.
Xin
B
,
Yamamoto
M
,
Fujii
K
,
Ooshio
T
,
Chen
X
,
Okada
Y
, et al
.
Critical role of Myc activation in mouse hepatocarcinogenesis induced by the activation of AKT and RAS pathways
.
Oncogene
2017
;
36
:
5087
97
.
9.
Li
L
,
Jin
R
,
Zhang
X
,
Lv
F
,
Liu
L
,
Liu
D
, et al
.
Oncogenic activation of glypican-3 by c-Myc in human hepatocellular carcinoma
.
Hepatology
2012
;
56
:
1380
90
.
10.
Wang
H
,
Wang
P
,
Xu
M
,
Song
X
,
Wu
H
,
Evert
M
, et al
.
Distinct functions of transforming growth factor-β signaling in c-MYC driven hepatocellular carcinoma initiation and progression
.
Cell Death Dis
2021
;
12
:
200
.
11.
Wangensteen
KJ
,
Wang
YJ
,
Dou
Z
,
Wang
AW
,
Mosleh-Shirazi
E
,
Horlbeck
MA
, et al
.
Combinatorial genetics in liver repopulation and carcinogenesis with a in vivo CRISPR activation platform
.
Hepatology
2018
;
68
:
663
76
.
12.
Dhanasekaran
R
,
Baylot
V
,
Kim
M
,
Kuruvilla
S
,
Bellovin
DI
,
Adeniji
N
, et al
.
MYC and Twist1 cooperate to drive metastasis by eliciting crosstalk between cancer and innate immunity
.
Elife
2020
;
9
:
e50731
.
13.
Yau
T
,
Park
JW
,
Finn
RS
,
Cheng
A-L
,
Mathurin
P
,
Edeline
J
, et al
.
CheckMate 459: A randomized, multicenter phase III study of nivolumab (NIVO) vs sorafenib (SOR) as first-line (1L) treatment in patients (pts) with advanced hepatocellular carcinoma (aHCC)
.
Ann Oncol
2019
;
v874
5
.
14.
Finn
RS
,
Ryoo
B-Y
,
Merle
P
,
Kudo
M
,
Bouattour
M
,
Lim
HY
, et al
.
Pembrolizumab as second-line therapy in patients with advanced hepatocellular carcinoma in KEYNOTE-240: a randomized, double-blind, Phase III trial
.
J Clin Oncol
2020
;
38
:
193
202
.
15.
Abou-Alfa
GK
,
Lau
G
,
Kudo
M
,
Chan
SL
,
Kelley
RK
,
Furuse
J
, et al
.
Tremelimumab plus durvalumab in unresectable hepatocellular carcinoma
.
NEJM Evidence
.
Massachusetts Medical Society
;
2022
;
1
.
Available from
: https://evidence.nejm.org/doi/10.1056/EVIDoa2100070
16.
UCSC TumorMap
[
Internet
]. [
cited 2022 Oct 12
].
Available from
: https://tumormap.ucsc.edu/
17.
Hoadley
KA
,
Yau
C
,
Hinoue
T
,
Wolf
DM
,
Lazar
AJ
,
Drill
E
, et al
.
Cell-of-origin patterns dominate the molecular classification of 10,000 tumors from 33 types of cancer
.
Cell
2018
;
173
:
291
304
.
18.
Thorsson
V
,
Gibbs
DL
,
Brown
SD
,
Wolf
D
,
Bortone
DS
,
Ou Yang
T-H
, et al
.
The immune landscape of cancer
.
Immunity
2019
;
51
:
411
2
.
19.
Newman
AM
,
Liu
CL
,
Green
MR
,
Gentles
AJ
,
Feng
W
,
Xu
Y
, et al
.
Robust enumeration of cell subsets from tissue expression profiles
.
Nat Methods
2015
;
12
:
453
7
.
20.
Perkins
JR
,
Dawes
JM
,
McMahon
SB
,
Bennett
DLH
,
Orengo
C
,
Kohl
M
.
ReadqPCR and NormqPCR: R packages for the reading, quality checking and normalisation of RT-qPCR quantification cycle (Cq) data
.
BMC Genomics
2012
;
13
:
296
.
21.
Shachaf
CM
,
Kopelman
AM
,
Arvanitis
C
,
Karlsson
A
,
Beer
S
,
Mandl
S
, et al
.
MYC inactivation uncovers pluripotent differentiation and tumour dormancy in hepatocellular cancer
.
Nature
2004
;
431
:
1112
7
.
22.
Lai
I
,
Swaminathan
S
,
Baylot
V
,
Mosley
A
,
Dhanasekaran
R
,
Gabay
M
, et al
.
Lipid nanoparticles that deliver IL-12 messenger RNA suppress tumorigenesis in MYC oncogene-driven hepatocellular carcinoma
.
J Immunother Cancer
2018
;
6
:
125
.
23.
Dhanasekaran
R
,
Park
J
,
Yevtodiyenko
A
,
Bellovin
DI
,
Adam
SJ
,
Kd
AR
, et al
.
MYC ASO impedes tumorigenesis and elicits oncogene addiction in autochthonous transgenic mouse models of HCC and RCC
.
Mol Ther Nucleic Acids
2020
;
21
:
850
9
.
24.
MacDonald
KPA
,
Palmer
JS
,
Cronau
S
,
Seppanen
E
,
Olver
S
,
Raffelt
NC
, et al
.
An antibody against the colony-stimulating factor 1 receptor depletes the resident subset of monocytes and tissue- and tumor-associated macrophages but does not inhibit inflammation
.
Blood
2010
;
116
:
3955
63
.
25.
Gordon
SR
,
Maute
RL
,
Dulken
BW
,
Hutter
G
,
George
BM
,
McCracken
MN
, et al
.
PD-1 expression by tumour-associated macrophages inhibits phagocytosis and tumour immunity
.
Nature
2017
;
545
:
495
9
.
26.
Höllt
T
,
Pezzotti
N
,
van Unen
V
,
Koning
F
,
Eisemann
E
,
Lelieveldt
B
, et al
.
Cytosplore: interactive immune cell phenotyping for large single-cell data sets
.
Comput Graphics Forum
2016
.
171
80
.
27.
Newton
Y
,
Novak
AM
,
Swatloski
T
,
McColl
DC
,
Chopra
S
,
Graim
K
, et al
.
TumorMap: exploring the molecular similarities of cancer samples in an interactive portal.
Cancer Res
2017
;
77
:
e111
4
.
28.
Jiang
P
,
Gu
S
,
Pan
D
,
Fu
J
,
Sahu
A
,
Hu
X
, et al
.
Signatures of T-cell dysfunction and exclusion predict cancer immunotherapy response
.
Nat Med
2018
;
24
:
1550
8
.
29.
Kim
H-D
,
Park
S
,
Jeong
S
,
Lee
YJ
,
Lee
H
,
Kim
CG
, et al
.
4-1BB delineates distinct activation status of exhausted tumor-infiltrating CD8 T cells in hepatocellular carcinoma
.
Hepatology
2020
;
71
:
955
71
.
30.
Kim
H-D
,
Song
G-W
,
Park
S
,
Jung
MK
,
Kim
MH
,
Kang
HJ
, et al
.
Association between expression level of PD1 by tumor-infiltrating CD8 T cells and features of hepatocellular carcinoma
.
Gastroenterology
2018
;
155
:
1936
50
.
31.
Jerby-Arnon
L
,
Shah
P
,
Cuoco
MS
,
Rodman
C
,
Su
M-J
,
Melms
JC
, et al
.
A cancer cell program promotes T-cell exclusion and resistance to checkpoint blockade
.
Cell
2018
;
175
:
984
97
.
32.
Herold
S
,
Wanzel
M
,
Beuger
V
,
Frohme
C
,
Beul
D
,
Hillukkala
T
, et al
.
Negative regulation of the mammalian UV response by Myc through association with Miz-1
.
Mol Cell
2002
;
10
:
509
21
.
33.
Peukert
K
.
An alternative pathway for gene regulation by Myc
.
EMBO J
1997
;
16
:
5672
86
.
34.
Herkert
B
,
Eilers
M
.
Transcriptional repression: the dark side of myc
.
Genes Cancer
2010
;
1
:
580
6
.
35.
Kress
TR
,
Pellanda
P
,
Pellegrinet
L
,
Bianchi
V
,
Nicoli
P
,
Doni
M
, et al
.
Identification of MYC-dependent transcriptional programs in oncogene-addicted liver tumors
.
Cancer Res
2016
;
76
:
3463
72
.
36.
Rakhra
K
,
Bachireddy
P
,
Zabuawala
T
,
Zeiser
R
,
Xu
L
,
Kopelman
A
, et al
.
CD4(+) T cells contribute to the remodeling of the microenvironment required for sustained tumor regression upon oncogene inactivation
.
Cancer Cell
2010
;
18
:
485
98
.
37.
Bordano
V
,
Kinsella
GK
,
Cannito
S
,
Dianzani
C
,
Gigliotti
CL
,
Stephens
JC
, et al
.
G protein-coupled receptor 21 in macrophages: an in vitro study
.
Eur J Pharmacol
2022
;
926
:
175018
.
38.
Vonderheide
RH
,
Glennie
MJ
.
Agonistic CD40 antibodies and cancer therapy
.
Clin Cancer Res
2013
;
19
:
1035
43
.
39.
Piechutta
M
,
Berghoff
AS
.
New emerging targets in cancer immunotherapy: the role of cluster of differentiation 40 (CD40/TNFR5)
.
ESMO Open
2019
;
4
:
e000510
.
40.
Kortlever
RM
,
Sodir
NM
,
Wilson
CH
,
Burkhart
DL
,
Pellegrinet
L
,
Brown Swigart
L
, et al
.
Myc cooperates with Ras by programming inflammation and immune suppression
.
Cell
2017
;
171
:
1301
15
.
41.
Sodir
NM
,
Kortlever
RM
,
Barthet
VJA
,
Campos
T
,
Pellegrinet
L
,
Kupczak
S
, et al
.
MYC instructs and maintains pancreatic adenocarcinoma phenotype
.
Cancer Discov
2020
;
10
:
588
607
.
42.
Muthalagu
N
,
Monteverde
T
,
Raffo-Iraolagoitia
X
,
Wiesheu
R
,
Whyte
D
,
Hedley
A
, et al
.
Repression of the type I interferon pathway underlies MYC- and KRAS-dependent evasion of NK and B cells in pancreatic ductal adenocarcinoma
.
Cancer Discov
2020
;
10
:
872
87
.
43.
Rider
P
,
Carmi
Y
,
Guttman
O
,
Braiman
A
,
Cohen
I
,
Voronov
E
, et al
.
IL-1α and IL-1β recruit different myeloid cells and promote different stages of sterile inflammation
.
J Immunol
2011
;
187
:
4835
43
.
44.
Kaplanov
I
,
Carmi
Y
,
Kornetsky
R
,
Shemesh
A
,
Shurin
GV
,
Shurin
MR
, et al
.
Blocking IL-1β reverses the immunosuppression in mouse breast cancer and synergizes with anti–PD-1 for tumor abrogation
.
Proc Natl Acad Sci U S A.
2019
;
116
:
1361
9
.
45.
Liu
M
,
Guo
S
,
Stiles
JK
.
The emerging role of CXCL10 in cancer (review)
.
Oncol Lett
2011
;
2
:
583
9
.
46.
Bagchi
S
,
Yuan
R
,
Engleman
EG
.
Immune-checkpoint inhibitors for the treatment of cancer: clinical impact and mechanisms of response and resistance
.
Annu Rev Pathol
2021
;
16
:
223
49
.
47.
Kalbasi
A
,
Ribas
A
.
Tumour-intrinsic resistance to immune-checkpoint blockade
.
Nat Rev Immunol
2020
;
20
:
25
39
.
48.
Maeda
T
,
Hiraki
M
,
Jin
C
,
Rajabi
H
,
Tagde
A
,
Alam
M
, et al
.
MUC1-C induces PD-L1 and immune evasion in triple-negative breast cancer
.
Cancer Res
2018
;
78
:
205
15
.
49.
Vartuli
RL
,
Zhou
H
,
Zhang
L
,
Powers
RK
,
Klarquist
J
,
Rudra
P
, et al
.
Eya3 promotes breast tumor-associated immune suppression via threonine phosphatase-mediated PD-L1 upregulation
.
J Clin Invest
2018
;
128
:
2535
50
.
50.
Vonderheide
RH
.
CD40 agonist antibodies in cancer immunotherapy
.
Annu Rev Med
2020
;
71
:
47
58
.
51.
Roche
PA
,
Furuta
K
.
The ins and outs of MHC class II-mediated antigen processing and presentation
.
Nat Rev Immunol
2015
;
15
:
203
16
.
52.
Meyer
N
,
Penn
LZ
.
Reflecting on 25 years with MYC
.
Nat Rev Cancer
2008
;
8
:
976
90
.
53.
Allen-Petersen
BL
,
Sears
RC
.
Mission possible: advances in MYC therapeutic targeting in cancer
.
BioDrugs
2019
;
33
:
539
53
.