Cyclooxygenases-2 (COX-2) and Prostaglandin E2 (PGE2) are important inflammatory factors, associated with survival, invasion, growth and immune escape of cancer cells. One of the “hallmarks” of cancer is chronic inflammatory disease, which often promotes tumorigenesis and tumor progression (1). For example, inflammatory bowel disease patients had a higher lifetime risk of colon cancer linked to colitis at a younger age than the general population, demonstrating that cancer is easily caused by chronic inflammation (2). Tumor-associated inflammation involves complex interactions between epithelial and mesenchymal cells and in some cases can lead to epigenetic alterations. More broadly, however, chronic inflammation can lead to the production of growth factors that support the development of emerging tumors and cause them to behave as “unhealable wounds” (3). Chronic inflammation can also promote tumor development by facilitating tumor immune escape and establishing an immunosuppressive microenvironment, both of which are cancer-related characteristics. Tumor immune escape occurs through a variety of immunosuppressive mechanisms, such as dysfunctional antigen-presenting cells (APCs), tumor cell resistance to immune attack, decreased cytotoxicity of CD8⁺ T cells and natural killer (NK) cells, induction of immunosuppressive cells such as myeloid-derived suppressor cells (MDSCs), transition of T helper (Th) cells from Th1 to Th2 and Th17, transition of macrophages from M1 to M2 (4).

Numerous studies have shown the importance of nonsteroidal anti-inflammatory drugs (NSAIDs) and EP receptor antagonists in reducing tumor incidence, metastasis and mortality. NSAIDs cause tumor regression and suppress tumor growth by inhibiting the COX-2-PGE2 signaling pathway in several ways: 1. activation of tumor epithelial cells; 2. inhibition of tumor epithelial cell survival and tumor immune surveillance; 3. establishment of tumor-supporting microenvironment; 4. alteration of DNA methylation mechanisms (2). However, the molecular mechanisms underlying the anti-tumor effects of NSAIDs and EP receptor antagonists have not yet been fully understood. More clinical trials and studies are required to explore this further.

In this review, we emphasize our current understanding of COX-2 and PGE2 regulation of tumor immunity. In addition, we are looking into how the COX-2-PGE2 pathway affects tumor-associated immune cells. Given the significance of this pathway in tumor immune escape, we will discuss how to target each component of this pathway as potential strategy for overcoming tumor immune escape while avoiding some serious adverse effects associated with the use of NSAIDs or COX-2 inhibitors.

Cyclooxygenases (COX) include COX-1 and COX-2, also known as prostaglandin G/H synthase-1 and -2, are membrane-bound enzymes that are mainly found on the nuclear membrane and luminal side of the endoplasmic reticulum (5). COX-1 is structurally expressed in many healthy tissues and provides steady-state levels of prostaglandins to perform “housekeeping functions”. In contrast, COX-2 is an inducible isoform produced by prostaglandin-like substances that is typically absent or expressed at low levels in normal organs and tissues, but is overexpressed in many tumor and inflammatory tissues (6). In addition to inducing inflammation, COX-2 can also promote cell survival and proliferation. Therefore, COX-2 may be associated with tumorigenesis and development. Indeed, COX-2 has been found to be highly expressed in different tumors, including breast cancer (7), the melanoma (8), and colorectal cancer (2). Its levels are associated with the development, aggressiveness and prognosis of many tumor entities. Regular use of NSAIDs dramatically decreased the incidence of sporadic colorectal cancer as well as breast, lung, and prostate cancers, according to a new analysis that searched all epidemiological studies (case-control and cohort studies) conducted since 1980 (9). According to one study, genetically removing COX enzymes from mouse melanoma (BRAFV600E or NrasG12D), colorectal cancer (CT26) or breast cancer (4T1) cell lines resulted in dramatic tumor eradication (10). When metastases were assessed 2 weeks after surgical resection of the primary breast cancer tumor, fibroblast-targeted Ptgs2-deficient mice (Ptgs2DFb) were found to develop fewer metastases than WT mice (11). Patients with gastric cancer who have low staging and higher levels of COX-2 expression are at greater risk of dying from the disease. In addition, COX-2 has been identified as an independent prognostic factor for gastric cancer (12). According to the findings mentioned above, COX-2 is thought to be crucial for tumor growth.

COX enzymes convert arachidonic acid to the endogenous peroxidation intermediate PGH2, which is modified by prostaglandin synthase to produce five structurally related prostaglandins, including prostaglandin D2, prostaglandin E2, prostaglandin I2, prostaglandin F2α, and thromboxane A2. The two essential enzymes that catalyze the initial and concluding phases of this synthetic pathway are COX-2 and prostaglandin E2 synthase (PGES), including microsomal PGES (mPGES) and cytosolic PGES. In addition, PGE2 is converted to its inactive form, 15-keto-PGE2 (PGEM), which is then further metabolized to a stable terminal metabolite by the enzyme 15-hydroxyprostaglandin dehydrogenase (15-PGDH). High levels of PGEM increase the risk of gastric and colorectal cancers (13, 14), implying that PGEM may be a biomolecular marker for cancer risk prediction. Prostaglandins control cellular processes by attaching to G protein-coupled receptors on the cell surface. These cell surface receptors are named EP (EP1, EP2, EP3, EP4), DP (DP1, DP2), FP, IP, and TP. PGE2 transmits signals by binding to four receptors, EP1 through EP4 (15) ( Figure 1 ). The expression of each EP receptor and the strength of each EP signal determine the PGE2 signal’s final output. Each EP interacts with its unique G protein to activate particular downstream signaling pathways such as PKA pathway, β-catenin pathway, NF-κB pathway and PI3K/AKT pathway, which have various functions in biological and pathological behavior. Activated EP1 can upregulate the level of intracellular calcium ion concentrations; EP2 and EP4 receptors are associated with cAMP stimulation and PKA signaling through sequential activation of Gas and adenylyl cyclase; EP3 is responsible for downregulating cAMP levels and lead to different cellular responses through different G proteins (16, 17).

The most abundant prostaglandin, PGE2, is frequently linked to a poor prognosis in a number of human cancers, including colon, head and neck, lung, and breast cancers (6, 17, 18). In immunologically active hosts, the formation of tumors by mutant BRAFV600E mouse melanoma cells requires the synthesis of prostaglandin E2, which inhibits immunity and fosters tumor inflammation (10). It has been shown that mPGES1 expression increased in human melanoma, and that elevated expression of this protein are linked to reduced patient survival (19). The mPGES-1 gene deletion significantly reduced the risk of developing colon cancer and resulted in reduced multiplicity of distal colon tumors as well as tumor load in mice treated with azoxymethane (AOM) (20). In addition, in ApcMin/+ and AOM mice models, elevated endogenous PGE2, caused by 15-PGDH gene deletion, encouraged the development of colon cancer (21). Taken together, PGE2 is essential in the growth of cancers ( Figure 2 ).

In addition, numerous evidence has shown that the COX-2-PGE2 pathway promotes tumor development. It has been shown that PGE2 promotes sporadic or colitis-associated colon carcinogenesis via EP1 and EP2 receptor in colon cancer cell lines (22). According to research, numerous tumors are linked to EP1 receptor. For example, selective EP1 antagonist (ONO-8713) can significantly reduce the number of tumor cells in mouse with UVB-induced acute skin inflammation (23). PGE2 upregulates anti-apoptosis protein expression in hepatocellular carcinoma cells via EP1 receptors (24). While EP3 receptors were downregulated, EP1, EP2 and EP4 receptors were all highly increased in COX-2-driven mammary cancers. Downregulation of EP3 receptors suggests a possible protective role against mammary tumors. EP2 and EP4 receptor levels were reduced in the mammary glands of anti-inflammatory pain-treated mice, while EP1 and EP3 levels were not altered (25). According to Fujino et al, EP3 receptors could boost vascular endothelial growth factor receptor-1 signaling and encourage tumor cell metastasis (26). Therefore, it may be debatable how EP3 contributes to carcinogenesis, and more research is required. The role of the EP4 receptor in tumors appears to be more clearly defined than that of other EP receptors. EP4 is a high-affinity EP receptor, and it is considered to be a pro-cancer mediator in many different types of malignancies due to its high expression. In colorectal cancer, deletion of EP4 can attenuates the abnormal AOM-induced crypt (27, 28), and therefore EP4 receptor can be used for prostate cancer immunotherapy. For example, YY001, an antagonist of the EP4 receptor, inhibits prostate cancer growth by modulating the tumor microenvironment (TME), leading to significant tumor regression, long-term survival and long-lasting immune memory (29). Besides, in mouse model, EP4 knockdown suppressed metastasis of oral cancer cells in the lung (30). Collectively, these findings imply a critical function for the PGE2/EP4 signaling pathway in tumor development.

The COX-2-PGE2 pathway induces tumor immune evasion by regulating myeloid-derived suppressor cells (MDSCs), lymphocytes (CD8⁺ T cells, CD4⁺ T cells and natural killer cells), and antigen presenting cells (APCs). Understanding the mechanisms of the COX-2-PGE2 pathway may provide a solid foundation for developing new methods to overcome tumor immune escape ( Figure 3 ).

Immunotherapy has grown in popularity as a cancer treatment option in recent years. Given the importance of the COX-2-PGE2 pathway in carcinogenesis and progression, there is a clear opportunity for therapeutic intervention. NSAIDs are a class of drugs that primarily suppress COX enzyme activity, which also lowers prostaglandins production. The use of NSAIDs to inhibit tumor growth has been considered. Combination therapy with NSAIDs and anti-PD-1 monoclonal antibody induces tumor eradication faster than anti-PD-1 alone (10). In addition, an in vivo investigation revealed that NSAIDs improved anti-tumor responses and reversed the imbalance between Th1 and Th2 in the metastatic spread of colorectal cancer (88). Some studies have demonstrated that NSAIDs can reduce cancer-associated mortalities and lower cancer incidence (89). For this reason, the FDA has approved celecoxib, a selective COX-2 inhibitor, for the treatment of population with familial adenomatous polyposis who wish to avoid developing colorectal polyps. In experimental animals, specific COX-2 inhibitors can prevent the development of mammary tumors (90). However, due to the risk of adverse cardiovascular events, celecoxib, along with other NSAIDs (except from aspirin), should not be used for an extended period of time, particularly in patients with a history of atherosclerotic heart disease (91). Excessive bleeding is the most serious side effect of aspirin, and both the risk and mortality rate increase with age (92). One way to prevent these negative consequences is to target only downstream PGE2 signaling. Cell surface inhibitory receptors (such as PD-1) are centrally involved in T cell exhaustion (93). And PGE2 could be one of the inducers of PD-L1 expression (94). We suppose that COX2-PGE2 pathway may result in T cell exhaustion by inducing PD-L1 expression. Emerging data from multiple solid tumor mouse models indicates that EP4 antagonists and PD1/PD-L1 blockade are effective in inhibiting primary tumor growth (49, 95, 96). Therefore, it may be a new potential strategy to combine EP antagonists with PD1/PD-L1 blockade in treating T cell exhaustion. In addition, combination immunotherapy with chimeric antigen receptor T (CAR-T) cells and checkpoint blockade is thought to be the next immunotherapy frontier (97). For instance, combination PD-L1 blockade and CD19 CAR-T cell therapy resulted in better outcomes in patients with heavily B-ALL (98). PGE2 signaling can activate protein kinase A, and then inhibits T-cell receptor activation, which inhibits the antitumor effect of CAR-T cell therapy (99). Thus, we conclude that targeting EP receptors (such as EP2 and EP4) could represent a novel strategy for improving the efficacy of adoptive immunotherapy. Some clinical trials are ongoing to evaluate the efficacy of these inhibitors. A phase I trial showed that EP4 antagonist (E7046) was safe in patients with advanced malignancies (ClinicalTrials.gov, Number NCT02540291). 1 clinical trial is recruiting patients to evaluate whether a dual EP2/EP4 antagonist (TPST-1495) suppresses tumor growth (ClinicalTrials.gov, Number NCT04344795). Although there is abundant EP receptor expression in fibroblasts and inflammatory cells, the role of the COX-2-PGE2 signaling pathway in TME is not fully understood, and more studies are needed to demonstrate this.

In this review, we discuss the role of the COX-2-PGE2 pathway in tumor immune evasion regulation. The idea that efficient therapeutic approaches should involve eliminating tumor cells and suppressing tumor immune evasion is being supported by a growing body of evidence, such as the use of checkpoint inhibitors to target immunosuppressive cells and reactivate immunosuppressive effector T cells. The COX-2-PGE2 signaling pathway may assist tumors in evading immune systems by inducing tumor-associated immune cells aggregation, impaired APC activity, a switch from Th1 to Th2 and Th17 immune responses or by suppressing CD8⁺ cytotoxic T cell and NK cell functions to promote tumor immune escape. In conclusion, the COX-2-PGE2 pathway not only is an effective target for tumor eradication, but it also suppresses tumor immune escape. Therefore, the addition of cyclooxygenase-2 inhibitors or EP antagonists to standard treatment may enhance the response of immunotherapy in anti-tumor immune escape.

KJ and CQ drafted the manuscript and revised the manuscript. JL and BL were in charge of the whole research conduction and paper writing. All authors contributed to the article and approved the submitted version.