Tumor-Associated Macrophages Regulate PD-1/PD-L1 Immunosuppression

1Cancer is a worldwide health problem, with an increasing number of confirmed cases and a high mortality. Immune escape is one of the most important characteristics of cancers. Tumors reduce immunogenicity as they divide and proliferate, leading to immune escape. Immune checkpoint inhibitors (ICIs) are a new method for tumor immune escape that yields survival benefits for tumor patients. The first ICI that was developed targeted the protein cytotoxic T-lymphocyte antigen 4 (CTLA-4)-ipilimumab, which increased survival by 3.7 months in patients with advanced melanoma and boosted the field of cancer treatment. ICIs bind to CTLA-4 or PD-1 and its ligand PD-L1, the key targets related to T-cell activation and exhaustion, and then eliminate immune suppression by tumors.

2In the tumor microenvironment (TME), PD-L1 is expressed on the surface of tumors and binds to PD-1 on T cells to resist the killing effect of T cells, ultimately causing tumor immune escape. The application of anti-PD-1/PD-L1 monoclonal antibodies (mAbs) to block the PD-1/PD-L1 signaling pathway has shown excellent antitumor efficacy in a variety of solid tumors. However, clinical studies have demonstrated that some patients do not respond to the therapy, and some patients even exhibit tumor recurrence after a period of remission. Drug resistance is a crucial factor that determines the efficiency of anti-PD-1/PD-L1 ICIs. Therefore, a deeper understanding of the regulation of the PD-1/PD-L1 axis is essential for the improvement of antitumor immunotherapy.

3The mechanisms of resistance to PD-1/PD-L1 blockade mainly include dysfunction or activation disorder of T cells, depletion or reduced infiltration of T cells, and changes in PD-L1 expression. The infiltration of T cells in the TME is the precondition of antitumor immunity, while the infiltration of immunosuppressive cells is the premise of tumor immune escape. Tumor-associated macrophages (TAMs) are immunosuppressive cells that induce drug resistance to PD-1/PD-L1 therapy. As one of the most abundant cell types in solid tumors, TAMs contribute to T-cell dysfunction and exhaustion through the secretion of cytokines and metabolic products and increase PD-L1 expression in tumor cells and other immunosuppressive cells. In diagnosed cancers, high macrophage infiltration is often closely related to the occurrence of drug resistance to PD-1/PD-L1 immune suppressants. Therefore, TAMs have been suggested as important targets to reverse the resistance to anti-PD-1/PD-L1 therapy. In this review, we highlight the recent findings of the suppressive effects of TAMs on PD-1/PD-L1 checkpoint inhibitors. To facilitate precision medicine and expand the target population, we further discuss combination therapies that may improve the efficacy of ICIs targeting PD-1/PD-L1.

4Macrophages have powerful functions in identifying, phagocytosing, and removing bacteria and foreign bodies in the immune system. In the process of tumorigenesis, macrophages evolve, resulting in the properties of TAMs that promote tumor growth. Mounting evidence suggests that secretions or exosomes from tumor cells shift the transcriptional program of TAMs from the M1-like phenotype to the M2-like phenotype. In a variety of cancers, the infiltration of M2 TAMs is significantly related to poor prognosis, tumor progression, and other adverse clinical outcomes. Moreover, in the process of anti-PD-1/PD-L1 immunotherapy, M2 TAMs can also suppress immunotherapy efficacy by inhibiting T-cell activity and enhancing the expression of PD-L1 in the TME. Specifically, M2 TAMs inhibit the function of PD-1/PD-L1 blockers by secreting anti-inflammatory cytokines and exosomes, increasing superficial immune checkpoint ligands ( Figure 1 ).

5TAMs are capable of secreting several cytokines, which mediate tumor-promoting activity and immunosuppression in the TME. Among them, transforming growth factor-β (TGF-β) and prostaglandin E2 (PGE2) are noteworthy due to their strong immunosuppressive effect and high correlation with TAM.

6TGF-β has been shown to affect anti-PD-1/PD-L1 immunotherapy by inhibiting the activation of T-cells and the expression of PD-L1. TGF-β derived from TAMs inhibits T-cell activity by decreasing the expression levels of IFN-γ and Granzyme B (indicating cytotoxic activity) in T cells through phosphorylation of the Smad2/3 protein and the inhibition of mitochondrial respiration. The expression of TGF-β was associated with the infiltration of T cells. Tumors with higher expression of TGF-β presented lower infiltration of CD8+ T cells. TGF-β can also modulate the function of PD-1/PD-L1 by regulating PD-L1 expression. In solid tumors such as breast cancer, TGF-β can induce the upregulation of PD-L1 in tumor cells and tumor-associated angiogenesis, which may be associated with the accumulation of succinate in tumor cells. Increased TGF-β levels in the TME not only promoted T-cell exclusion and accumulation of regulatory T cells (Tregs) but also blocked the acquisition of the Th1 effector phenotype. In addition, since TGF-β in the TME originates from a variety of cells, immunosuppression induced by crosstalk between these cells should also be noted.

7PGE2 can inhibit T-cell activation and function by increasing the expression of PD-L1. As a downstream of cyclooxygenase 2 (COX-2), the level of PGE2 in the TME is regulated by the expression of COX-2 and microsomal PGE2 synthase 1 (mPGES1). In bladder cancer, TAMs can increase the expression of PD-L1 in tumor-infiltrating myeloid cells through the COX-2/mPGES1/PGE2 pathway, which leads to the exclusion of CD8+ T cells. Similarly, PGE2 upregulates PD-L1 expression in ovarian cancer cells by activating the PI3K-AKT-mTOR pathway. Moreover, PGE2 can induce the expression of Forkhead Box P3 (Foxp3) to stimulate the differentiation of immunosuppressive Tregs from naïve T cells. Of particular note is that therapies targeting PGE2 with NSAIDs or COX-2 inhibitors fail in clinical trials due to global prostaglandin inhibition, which in turn could cause serious side effects. Therefore, precision therapy targeting macrophages should be proposed, which may be the next step in reversing drug resistance to PD-1/PD-L1 therapy.

8In addition to the expression of PD-L1 on the surface of TAMs, there are also homologous immune checkpoint ligands that can block anti-PD-1/PD-L1 immune efficacy. V-domain Ig-containing suppressor of T-cell activation (VISTA), an immune checkpoint ligand expressed by TAMs, is an immunosuppressive molecule that reduces T-cell proliferation and cytokine production while sustaining Treg function. The expression of VISTA is not only positively correlated with the expression of PD-L1 on the surface of tumor cells but also correlated with the patient’s poor prognosis, pathological grade, and lymph node status. In fact, a recent study demonstrated a strong correlation between VISTA expression and tumor infiltration by myeloid cells and PD-1+ inflammatory cells. Targeting VISTA antibodies can regulate innate immunity and adaptive immunity by promoting T-cell infiltration, thereby slowing tumor growth in mouse cancer models. However, VISTA is also highly expressed in hematopoietic and microglial cells, and more studies on the systemic responses to VISTA-targeted therapies are needed. We look forward to its future application in combination with anti-PD-1/PD-L1 mAbs.

9Moreover, TAMs can prevent the interaction of anti-immune checkpoint mAbs with the targets through the Fcγ receptor present on cell surfaces. Indeed, studies have demonstrated that after administration, anti-PD-1 mAb binds to tumor-infiltrating T cells at an early stage but is subsequently captured by TAMs due to the presence of Fcγ receptors, which ultimately leads to drug failure. More importantly, activation of Fcγ receptors by an anti-PD-1 mAb results in depletion of activated CD8+ T cells in vitro and in vivo, reducing the therapeutic effect. Therefore, the design of FC-null anti-PD-1 mAbs or specific competitive inhibitors is one of the future strategies necessary to block Fcγ receptor-mediated resistance and increase T-cell infiltration.

10Exosomes in the TME have been reported as a medium of communication between cells for the occurrence and invasion of tumors. Exosomes are able to promote the migration of cancer cells through the PI3K-AKT signaling pathway activated by apolipoprotein E. The effect of M2 TAM-derived exosomes on drug resistance has also been reported, and it has been verified that microRNAs in exosomes are key regulators of resistance to gemcitabine. Analogously, miRNAs of M2 TAM-derived exosomes have also been implicated in the regulation of anti-PD-1/PD-L1 immunotherapy. MicroRNA-21 (MiR-21) expression is relatively high in glioma and associated with low infiltration of CD8+ T cells. Inhibiting miR-21 in exosomes not only improves the proliferation and cytotoxic activity of CD8+ T cells but also reduces the level of TGF-β1, which prevents immune escape of glioma cells. In addition, another study showed that the combination of miR-21 deletion and anti-PD-1 treatment demonstrates better antitumor activity than either drug alone. Moreover, in vivo, miR-155-5p in exosomes secreted by M2 TAMs can promote the expression of interleukin-6 (IL-6) in tumor cells, thereby inhibiting the T-cell immune response.

11As mentioned above, TAMs have a multichannel inhibitory effect on anti-PD-1/PD-L1 immunotherapy. Therefore, targeting TAMs is of great significance to improve the efficacy of anti-PD-1/PD-L1 immunotherapy. Currently, strategies are designed to deplete TAMs, re-educate TAMs, and intervene with chemokines secreted by TAMs. The combination of the exosomes secreted by M1 macrophages and other nanoimmunotherapy strategies provides more potential options to reduce the occurrence of PD-1/PD-L1 immunosuppression ( Figure 2 ). In addition, we summarize the current clinical trials on different targets of TAMs in combination with anti-PD-1/PD-L1 mAbs ( Table 1 ).

12As the infiltration of TAMs limits clinically relevant immune responses, depleting TAMs seems to be an attractive strategy. Colony-stimulating factor 1 receptor (CSF-1R) blockers are the main method by which to deplete M2 TAMs. CSF-1 binds to its receptor, and the latter then undergoes autophosphorylation, which plays an important role in the proliferation, differentiation, and maintenance of macrophages. Therefore, a CSF-1R inhibitor (CSF-1Ri) may improve the efficacy of anti-PD-1/PD-L1 immunity. PLX3397 (pexidartinib), a CSF-1R kinase inhibitor, can increase the infiltration and antitumor function of CD8+ T cells in tumors when combined with anti-PD-1 therapy. At the same time, it can also effectively reduce the appearance of tumor neovascularization and ascites. BLZ945, another CSF-1Ri, combined with a PD-1/PD-L1 blocking antibody is also effective in controlling tumor growth. Inherent antitumor drugs can also target CSF-1R to provide a new method for combined immunotherapy. Erlotinib, a first-generation small-molecule inhibitor targeting the epidermal growth factor receptor (EGFR) tyrosine kinase, is suitable for the first-line treatment of EGFR mutation-positive non-small cell lung cancer (NSCLC). Combination therapy consisting of its derivative TD-92 and anti-PD-1 contributes to reduced tumor growth and increased survival in vivo.

13Recently, several findings have suggested that re-education of M2 TAMs rather than depletion may represent a more effective strategy. Previous studies have reported that the protumor M2 phenotype can be re-educated to the tumoricidal M1 phenotype, thereby inhibiting the supporting role of TAMs in tumors. BRD4, a bromodomain and extraterminal (BET) family protein, can enhance the expression of CCL2 by activating the NF-κB signaling pathway, which, in turn, causes the recruitment of macrophages in tumors. The BRD4 inhibitor AZD5153 can re-educate TAMs from M2 to M1 and promote the secretion of proinflammatory cytokines, thereby activating cytotoxic T lymphocytes (CTLs) in vitro. More importantly, AZD5153 was proven to render ovarian cancer sensitive to anti-PD-L1 treatment through a 3-D microfluidic model. SF2523, another BRD4 inhibitor that can block the polarization of TAMs, restores the activity of CD8+ T cells and then stimulates the antitumor immune response. Furthermore, the BET inhibitor JQ1 can significantly reduce PD-L1 expression on tumor cells and TAMs and limit tumor progression in a cytotoxic T-cell-dependent manner.

14Ubiquitin-specific protease 7 (USP7), a deubiquitinating enzyme, is considered a promising therapeutic target because of its regulatory role in DNA damage and epigenetic inheritance. USP7 has been identified as a highly expressed M2 TAM gene, and specific inhibition of USP7 can reprogram M2 TAMs into M1 through the P38 MAPK pathway. In addition, targeting USP7 promoted the infiltration and cytotoxicity of CD8+ T cells in the TME and decreased PD-L1 expression in tumor cells.

15In addition to targeting M2 TAMs themselves, the immunosuppressive effects of M2 TAMs can also be reversed by inhibiting cytokines secreted by M2 TAMs.

16As one of the most important immunosuppressive cytokines secreted by M2 TAMs, TGF-β changes the TME by limiting the infiltration of T cells to inhibit antitumor immunity. Combination therapy with a TGF-β inhibitor (1D11 and galunisertib) and anti-PD-1/PD-L1 resulted in the upregulation of immune response genes, restoring the cytotoxic activity of T cells and the antitumor activity of anti-PD-L1. In addition to the above effects, combinatorial treatment consisting of tranilast, a TGF-β inhibitor, with Doxil nanomedicine has been shown to improve M1 macrophage content in the tumor tissue, which results in the increased efficacy of anti-PD-1. Phase II research is ongoing to investigate the combined effect of TGF-β inhibitor (vactosertib) with anti-PD-L1 (durvalumab) (NCT04064190), while another phase II study is assessing the efficacy and safety of NIS793 with and without spartalizumab (NCT04390763) in untreated metastatic pancreatic ductal adenocarcinoma (mPDAC). In addition, M7824 is a bifunctional fusion protein comprising a mAb against PD-L1 fused to the extracellular domain of TGF-β receptor 2. Compared with targeting TGF-β alone, M7824 has been proven to reduce tumor burden and improve overall survival (OS). A phase II trial was performed to determine the efficacy of M7824 plus topotecan or temozolomide in recurrent SCLC (NCT03554473).

17CCL2 is a chemokine that attracts a number of CCR2-high-expressing monocytes to the tumor site. The role of C-C motif chemokine receptor 2 (CCR2) seems to influence TAM recruitment at the tumor site. In a recent study, it was proven that CCR2 is involved in the recruitment and initiation of tumor-promoting inflammation. Many preclinical studies showed high efficacy of CCL2/CCR2 antagonists; for instance, targeting CCR2 with a small-molecule inhibitor not only reduced recruitment of M2-type macrophages but also induced tumor infiltration of activated CD8+ T cells. Many other preclinical studies on different tumor types showed that either depletion of CCR2 or disruption of the CCL2-CCR2 interaction has an impact on the inhibition of TAM recruitment and tumor regression or inhibition of metastasis. A phase I/II trial of combination immunotherapy with nivolumab and a CCR2/CCR5 dual antagonist (BMS-813160) is in progress to evaluate whether this therapy is safe in patients with locally advanced pancreatic cancer (LAPC) (NCT03767582).

18CXC-motif chemokine ligand 12 (CXCL12) is another chemokine that regulates the migration of monocytes. Elevated C-X-C chemokine receptor 4 (CXCR4) is correlated with the tumorigenesis of NSCLC. CXCL12 secretion could be induced in response to radiation therapy and cause the accumulation of TAMs in the tumor. BL-8040 (motixafortide), one of the CXCR4 antagonists, plus the anti-PD-1 pembrolizumab in the COMBAT trial contributes to the improvement found in pancreatic ductal adenocarcinoma (PDAC) patients. After using a new CXCR4 inhibitor peptide R, the expression of CD73, CD38, and IL-10 in non-small cell lung cancer is reduced, which can rescue the cytotoxic activity of T cells and prevent TAM polarization. Another CXCR4 antagonist, Pep R, demonstrated efficacy in combination with nivolumab in melanoma. In addition, there are already 2 observational studies in progress to study whether Pep R can reverse anti-PD1 resistance (NCT03891485 and NCT03628859).

19Many people have recognized the potential of TGF-β, and there are currently many clinical trials combining TGF-β targeting with PD-1/PD-L1 treatment. However, most studies have stalled due to serious adverse events or the observation of minimal clinical benefit, which may be associated with higher global drug levels. As CXCR4 is widely expressed in hematopoietic cells and a variety of stem cells, therapies targeting CXCR4 face a similar dilemma. In contrast, therapies that block the CCL2–CCr2 axis appear to be safer and are expected to elucidate the mechanism of action in different cancer species. In addition, for the current targeted drugs yielding a poor systemic response, the combination of cell-targeted drug delivery systems is an ideal choice to counteract their limitations.

20Compared with traditional delivery systems, nanoparticles that can specifically deliver drugs to TAMs and modulate their polarized states may be an effective method in cancer immunotherapy. For example, tumor cell-derived microparticles containing the chemotherapeutic drug methotrexate (TMP-MTX), nanoparticles delivering shikonin and PD-L1 knockdown siRNA (SK/siR-NPs), and Gadofullerene (GF-Ala) nanoparticles can all reprogram M2 TAMs to an M1-like phenotype and increase the infiltration of CTLs, thereby effectively inhibiting tumor growth. CMPB90-1, a new natural polysaccharide from Cordyceps, converts immunosuppressive TAMs by binding to toll-like receptor 2 (TLR2), polarizes TAMs to the M1 phenotype, and has antitumor effects and a better safety profile. These studies may provide a promising strategy for the development of high-efficiency, low-toxicity immunotherapy based on nanotechnology.

21Due to the unpredictable toxicity and poor scalability of nanocarriers in the human body, other advanced drug carriers have gradually attracted attention, such as hydrogels, exosomes, and enucleated cells. A Melittin-Rada32 hybrid peptide hydrogel loaded with doxorubicin (DOX) was designed to reshape the tumor immune microenvironment in the treatment of melanoma, which specifically consumes M2 TAMs and increases activated CTL infiltration. Dai et al. designed a hydrogel scaffold loaded with KN93, a Ca2+/calmodulin-dependent protein kinase II (CAMKII) inhibitor, which can reprogram TAMs into the M1 phenotype. After this hydrogel treatment, CTL infiltration in the TME increased, and the expression of macrophage PD-L1 increased, suggesting that it has good prospects for anti-PD-1 treatment.

22Studies have reported that exosomes derived from macrophages have immunomodulatory effects. Exosomes secreted by M1 macrophages are reported to inhibit the development of gastric cancer and activate T-cell-dependent immune responses. Endogenous macrophage exosomes have been shown to have absolute advantages over their safety. Therefore, M1 macrophage-derived exosomes (M1-exos) can be used to deliver various anticancer drugs for tumor therapy. Mannose-modified macrophage-derived microparticles (Man-MPs) loaded with metformin have been developed to efficiently target M2-like TAMs to repolarize them into the M1-like phenotype. More importantly, the collagen-degrading capacity of Man-MPs contributes to the infiltration of CD8+ T cells into tumor interiors and enhances tumor accumulation and penetration of anti-PD-1. Macrophage-derived exosomes loaded with PTX and Dox were developed to treat triple-negative breast cancer (TNBC) in vivo. With the development of nanotechnology, exosome-mimetic nanovesicles derived from M1 macrophages (M1NVs) were designed to repolarize M2 TAMs to M1 macrophages. Moreover, injection of a combination of M1NVs and anti-PD-L1 further reduced the tumor size compared with the injection of either M1NVs or aPD-L1 alone.

23As a conventional means of tumor treatment, radiotherapy leads to increased expression of PD-L1 in tumor cells, which is one of the markers of anti-PD-1/PD-L1 mAb therapy. Interestingly, in a model of malignant pleural effusions, microparticles released by radiated tumor cells (RT-MPs) can precisely locate M2-TAMs in the TME and convert the latter into M1-TAMs by activating the JAK-STAT and MAPK pathways. It is noteworthy that the combination of RT-MPs and anti-PD-1 exhibits good biocompatibility and memory immune response, which TMP-MTX cannot match. Since hypoxia has been proven to be an important factor affecting the clinical outcome after radiotherapy, strategies to interfere with hypoxia have been developed to optimize radiotherapy. Oxygen microcapsules are designed to rapidly increase the oxygen concentration in the TME, resulting in reduced infiltration of TAMs while enhancing the efficacy of radiotherapy. Of particular note is the microcapsules’ ability to repolarize TAMs into the M1 phenotype, which, in turn, activates the T-cell-mediated antitumor immune response. Obviously, this nanotherapy can further enhance the efficacy of radiotherapy combined with anti-PD-1/PD-L1 mAbs.

24The application of ICIs in clinical practice has completely changed the therapeutic strategies for cancer patients by improving the prognosis and reducing the impact on their quality of life (QoL) compared with standard approaches. However, the high cost of anti-PD-1 and PD-L1 agents highlights the need to select patients who will benefit most from the treatment early, supporting the research on predictive biomarkers of response and strategies to overcome resistance and optimize the efficacy of these drugs.

25Targeting macrophages to treat cancer is a young but rapidly developing area of research and therapy. Despite great interest, the optimum therapeutic approach has not yet been identified because TAMs represent a heterogeneous population, and their role in tumors varies depending on many environmental conditions. The other difficulty arises from the TME, which is a very dynamic tissue and contains various infiltrating immune cells and external factors that influence tumor progression, macrophage polarization, and therapeutic response. Some macrophage-targeting therapeutics are effective as monotherapies. However, more evidence exists that targeting TAMs could improve the efficacy of conventional therapies and immune therapeutics. Currently, two main approaches that target TAMs with apparent opposite effects have been developed. One approach is to deplete macrophages, and the other is to re-educate them to kill cancer cells. Depending on the macrophage infiltration status and the chosen therapy as a combination treatment, various approaches will be chosen. For example, through their Fcγ receptors, macrophages were shown to take up therapeutic antibodies such as anti-PD-L1, limiting the efficacy of such therapeutic modalities in animal models. In fact, in several recent studies, it was shown that depleting macrophages with the use of CCL2/CCR2 antagonists improves the efficacy of PD-L1-targeting antibodies and possibly other ICIs.

26In addition, some important issues should be resolved before TAM antagonists are used to overcome resistance to immunotherapy. More convincing clinical studies are needed to confirm the correlation of macrophage infiltration or phenotype with the outcomes of patients under anti-PD-1/PD-L1 therapy. It is essential to identify subpopulations that have the potential to benefit from different therapies targeting macrophages. Despite these difficulties, there is still great potential to harness macrophage biology to improve the efficiency of anti-PD-1/PD-L1 ICIs in oncology.

27YP and QJ conceived the structure of manuscript and revised the manuscript. YP created the figures. QJ reviewed the manuscript. All authors contributed to the article and approved the submitted version.

28This work was supported by the National Science Foundation of China (82074225 to QJ).

29The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

30All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.