RNA modification is an epigenetic change that alters the structure and function of RNA molecules by inserting, deleting, or substituting nucleotides at specific locations, playing a key role in cellular physiology and pathology. Currently, the roster of RNA modifications that have been pinpointed has extended to over 170 distinct subtypes (1), including methylation, acetylation, pseudouridinization, among others.

RNA methylation modifications are regulated by specific proteins, including methyltransferase “writers” that add marks, demethylase “erasers” that remove them, and “readers” that recognize these modifications. Among RNA modifications, N1-methyladenosine (m1A) (2), 5-methylcytosine (m5C) (3), N6-methyladenosine (m6A) (4), and 7-methylguanosine (m7G) have been the focus of comprehensive studies regarding their oncogenic properties and are viewed as key determinants in the occurrence (5), development, and treatment of tumors.

Immune cells within tumors play a key role in clearing the tumor. Activating immune cells, especially the tumor-clearing functions of T cells, can suppress tumor growth. However, immune cells in the tumor microenvironment are typically in a state of suppression, leading to immune evasion. We have found that RNA modifications are involved in the regulation of immune cell suppression in tumors. Immunotherapy, such as anti-PD-1/PD-L1 therapy, has been widely applied in clinical practice, but only a subset of patients is sensitive to immunotherapy. The efficacy of cancer immunotherapy is contingent upon a multitude of factors, including the type of tumor, tumor’s mutational burden, the stability of microsatellites, and the combined use of chemotherapy drugs. Notably, the presence of immune cell infiltration within the tumor microenvironment and the expression of immune checkpoints on the tumor’s surface play pivotal roles in the therapy’s success. We have found that RNA modifications, primarily the m6A modification, have a complex regulatory effect on the tumor infiltration of immune cells and the expression of PD-L1. By targeting the m6A regulators in tumors or immune cells, it is possible to increase immune cell infiltration, regulate the expression of PD-L1, sensitize patients to immunotherapy, inhibit tumor growth, and ultimately improve patient prognosis.

We have compiled a summary of the regulatory roles of four types of RNA methylation modifications, mainly m6A modification, in immune cells, as well as their regulatory effects on the expression of PD-L1 in tumor cells and the research progress on m6A regulators as therapeutic targets. We also reported on the predictive role of methylation modification-related genes in the efficacy of immunotherapy.

m6A, representing N6-adenosine methylation, is recognized as the most frequent and plentiful type of post-transcriptional RNA modification, predominantly occurring within mRNA in the nucleus (6). This modification contributes to essential life processes such as hematopoiesis, central nervous system development, and germ cell differentiation (4). Common regulatory enzymes include “writers” METTL3, METTL14, METTL16, “readers” YTHDF1, YTHDF2, YTHDF3, LRPPRC, IGF2BP1 and IGF2BP3, among others, and “erasers” FTO, ALKBH5. METTL3 is the most critical ‘writer’ enzyme for m6A modification, serving as the catalytic subunit of the methyltransferase complex that mediates N6-methyladenosine (m6A) methylation on mRNA. It commonly forms a heterodimer with METTL14, facilitating the process (7). Wilms Tumor 1 Associated Protein (Wtap), a regulatory subunit of the complex, enhances the binding affinity of METTL3 for mRNA (8). METTL3 and METTL14 are primarily located in the nucleus, while METTL16 is mainly found in the cytoplasm. Although METTL16 also mediates the m6A modification, its role in promoting tumorigenesis in various cancers is more noteworthy, which is related to the acceleration of mRNA translation by METTL16 binding to eIF3a/b in the cytoplasm (9). YTHDFs recognize mRNAs modified with m6A; YTHDF1 promotes the translation of these mRNAs, while YTHDF2 accelerates their degradation (10). YTHDF3 interacts with YTHDF1 to augment translation of m6A-modified mRNA and influences the RNA binding activity of YTHDF2 (11). And IGF2BPs enhance the stability of mRNA after recognizing m6A modification (12). FTO and ALKBH5 are the only two demethylases identified for m6A demethylation, which remove the m6A modification from RNA ( Figure 1 ). The role of m6A regulatory enzymes in cancer has been widely reported. Several preclinical experiments have demonstrated that the use of drugs targeting m6A regulators can inhibit tumor growth or enhance the therapeutic efficacy of immunotherapy ( Table 1 ).

m1A and m5C modifications predominantly occur in mRNA, tRNA, and rRNA. The m1A modification is mainly found in tRNA and shares some similarities with m6A modification in mRNA, although it occurs at a much lower frequency. The similarities between m1A and m6A modifications are also reflected in their shared binding proteins YTHDF1, YTHDF2, YTHDF3, and the demethylase FTO (2). m5C plays an important role in maintaining the structure and stability of tRNA and rRNA. m5C modification can occur in mRNA and can regulate the stability of mRNA, but the research on m5C in mRNA is not sufficiently in-depth and comprehensive (3). m7G is widely present in mRNA and is crucial in the translation process and is involved in many important physiological processes (5).

The emergence of new sequencing technologies and bioinformatics tools has facilitated the detection and in-depth understanding of RNA modifications. However, there are still many RNA modifications that have not yet been discovered, and research on many that have been confirmed is far from adequate. We have primarily focused on the function of m6A modification, as it is the most prevalent and well-studied modification in mRNA, but it is important to recognize that other RNA modifications also hold significant potential. The relationship between RNA modifications and the occurrence and development of tumors, as well as their therapy, is currently a hotbed of research. Multiple regulators of RNA modifications are considered oncogenes and have shown potential therapeutic value. The exploration of gene editing technology to enhance RNA modifications that combat tumors and reduce those that promote tumor growth also presents a valuable research avenue.

RNA modifications have been shown to be associated with the expression of various immune checkpoints, such as the upregulation of PDCD1 and KIR3DL1 and the downregulation of TIGIT, IDO1, and BTLA in the high-risk group established based on m6A scoring in diffuse large B-cell lymphoma (109). In breast cancer, the groups differentiated by m1A scoring exhibit differential expression of immune checkpoints, for example, TIGIT, IDO1, LAG3, and ICOS (110). We mainly summarize and introduce the research findings related to RNA modifications and PD-1/PD-L1 and CTLA-4.

RNA modifications exhibit differential impacts on PD-1 blockade therapy across various tumors. Targeting specific RNA modification regulators can enhance the efficacy of PD-1 blockade therapy ( Figure 3 ). In NSCLC, METTL3-driven m6A methylation leads to the destabilization of PD-L1 mRNA, which reduces the therapeutic efficacy of anti-PD-1/PD-L1 interventions; however, the knockout of METTL3 increases immune cell infiltration and enhances the therapeutic efficacy of anti-PD-1/PD-L1 (122). Similarly, the deletion of METTL3 or METTL14 in immune-resistant melanoma tumor cells makes the tumor sensitive to immunotherapy (79). Targeting METTL3 in NAFLD-HCC and NSCLC can improve the effectiveness of PD-1 therapy. In the NAFLD-HCC mouse model, the knockout of METTL3 in conjunction with anti-PD-1 therapy synergistically suppressed tumor growth, resulting in a reduction of over 90% in both tumor volume and weight (83). In NSCLC, the suppression of METTL3 can enhance the sensitivity of tumor-bearing mice to anti-PD-1 treatment, and patients with NSCLC exhibiting low METTL3 expression have a more favorable prognosis with immunotherapy (122). In thyroid cancer, high expression of METTL3 in tumor cells inhibits the demethylation of CD70mRNA, maintains the degradation of transcripts mediated by YTHDF2, thereby releasing T cells from suppression and enhancing the efficacy of PD-1 blockade (86). Similar observations have been made in melanoma and lung cancer, where high expression of METTL3 in macrophages is beneficial for immunotherapy (63). Targeting the YY1-CDK9 transcription elongation complex in glioblastoma results in lowered METTL3 expression, which in turn, enhances the therapeutic impact of PD-1 blockade (150). Abrine treatment inhibits IFN-γ-induced m6A modification, thereby regulating JAK1/STAT1 and suppressing the expression of PD-L1. Combined therapy with PD-1 blockade can inhibit tumor growth (151). Targeting YTHDF1 in colorectal cancer can relieve the inhibition of CD8⁺ T cells and enhance the efficacy of anti-PD-1. Targeting YTHDF1 significantly reduced the resistance to anti-PD-1 therapy in the MC38 tumor model, leading to better prognosis for tumor-bearing mice. Similar observations were made in mice with the CT26 tumor model, which is insensitive to PD-1 therapy; the deletion of YTHDF1 in CT26 cells followed by anti-PD-1 treatment markedly inhibited tumor growth (91). Knocking out ALKBH5 in glioma or targeting it with IOX1 reduces the expression of the PD-L1 protein, inhibits tumor growth, extends the survival of mice, and enhances these effects of PD-1 blockade therapy (105). Similarly, inhibiting ALKBH5 in melanoma enhances the efficacy of PD-1 blockade, patients with low expression of ALKBH5 are more likely to benefit from PD-1 blockade therapy (31). The use of FTO inhibitors in HCC and melanoma can enhance immune activation and sensitivity to anti-PD-1 treatment (99, 152). Targeting circular RNA circRHBDD1 can block its m6A-dependent mediated rapid translation of PIK3R1 and improve the efficacy of anti-PD-1 therapy (153). NSUN2-mediated m5C methylation modulates TREX2 expression, thereby suppressing the cGAS/STING pathway and contributing to resistance against PD-1 checkpoint blockade (51).

In this part, we have explored the regulatory role of RNA modifications on PD-1 blockade therapy across different types of tumors. By affecting the expression of genes involved in immune activation and suppression, RNA modifications have significantly impacted the therapeutic efficacy of immune checkpoint blockade and have provided ideas for specific treatment plans: by targeting specific RNA modification regulators, we may be able to increase the response rate to immunotherapy and overcome patient resistance to existing treatments.

Genes related to the RNA modifications m6A, m1A, m5C, and m7G can predict tumor sensitivity to immunotherapy, either individually or in combination. By conducting bioinformatics analysis on many samples in databases, key genes are screened, and corresponding scores are established. For instance, in HCC, an m6A score derived from m6A-associated feature genes categorizes patients into distinct risk groups. The high-risk cohort demonstrates increased immune evasion and immune system dysregulation, correlating with heightened responsiveness to anti-CTLA-4 and anti-PD-1 therapies (154). The m6A score has been studied in multiple tumors but varies depending on the analysis method and selected genes. Patients with oral squamous cell carcinoma exhibiting a high m6A score are more likely to experience enhanced efficacy from treatments targeting the PD-1 and CTLA-4 pathways (155), lung squamous cell carcinoma (156), soft tissue sarcoma (157), gastric cancer (102), and hepatocellular carcinoma (158), while a low m6A score suggests greater sensitivity to immunotherapy in thyroid cancer (159), NSCLC (160), follicular lymphoma (161), breast cancer (162), and head and neck squamous cell carcinoma (163). Additionally, the score of m6A methyltransferase regulators can effectively predict the efficacy of immunotherapy in urothelial cancer patients (164).

A prognostic model established using m6A-related lncRNA suggests that esophageal cancer in the low-risk group responds better to immunotherapy (165). Similarly, a model established using m5C-related genes can also evaluate prognosis and immune therapy efficacy, with liver and pancreatic cancers with lower m5C scores being more sensitive to anti-CTLA-4 therapy, and pancreatic cancer also being sensitive to anti-PD-1 therapy (166). A score based on m7G indicates that colorectal cancer and rectal cancer in the low-scoring group are more sensitive to anti-PD-1 therapy (119, 121). Lung adenocarcinoma with a low score has a higher immune prognostic score (120). Interestingly, in low-grade glioma, the high m7G score group is sensitive to anti-PD-1 treatment, while the low m7G score group is more sensitive to anti-PD-L1 (167). In colorectal cancer, a low m1A score suggests a better prognosis with anti-PD-L1 treatment (77). Lung adenocarcinoma with a low m1A score has a lower TIDE score, indicating greater sensitivity to immunotherapy (78).

In addition to establishing models based on the scores of single RNA modifications, analyzing multiple RNA modification genes simultaneously can also establish effective predictive models. In cervical cancer, a prognostic model established using m6A/m5C/m1A indicates that the high-risk group is more sensitive to anti-CTLA-4 treatment (168). In colon cancer, the low-risk group is more sensitive to anti-CTLA-4 and anti-PD-1 treatments (169). In HCC, a methylation score composed of m1A/m5C/m3C/m6A suggests that a low score is sensitive to anti-PD-L1 therapy (170). The Writer-Score established according to m1A and m6A RNA modification enzymes indicates that a low score is associated with better outcomes in immunotherapy (171).

Through bioinformatics analysis of large-sample databases, we have been able to screen for key genes and establish corresponding scoring systems. These scoring systems have demonstrated the potential to predict responses to immunotherapy across various types of tumors, offering new tools for personalized medicine. However, many scoring systems still require further validation and research to explore their applicability in oncology and immunotherapy strategies.

Therapeutic interventions targeting RNA modification-related genes or proteins have been extensively studied, and numerous effective drugs have been developed. For instance, metformin can specifically inhibit FTO and block its demethylation effect on m6A modification (172). Targeting specific regulatory genes or proteins of RNA modification can enhance the therapeutic efficacy of immunotherapy in certain types of cancer. For example, targeting ALKBH5 in melanoma can enhance the effectiveness of PD-1 blockade (31). Lower levels of METTL3 in NSCLC are also associated with better outcomes from anti-PD-1 therapy (122). Moreover, targeting specific RNA modification regulators’ genes or proteins can directly exert tumor-suppressive effects. For example, miR-4429, which targets METTL3 in gastric cancer, can inhibit the proliferation of GC cells and induce apoptosis (14). Although the vast majority of drug treatments target m6A regulatory genes or proteins, we found that targeting the m1A methyltransferase complex TRMT6/TRMT61A in HCC with Hiram can effectively inhibit the progression of HCC (32), indicating that targeting other RNA modification regulatory genes or proteins also has therapeutic significance. We have summarized the specific targeted drugs and their mechanisms of action ( Table 1 ).

This section of our study extensively explores therapeutic interventions targeting genes or proteins associated with RNA modifications and outlines the development of a series of effective drugs. Targeting regulators such as METTL3, METTL14, IGF2BP3, or YTHDF1 can alleviate T-cell suppression in melanoma, colorectal cancer, non-small cell lung cancer (NSCLC), and breast cancer. Targeting regulators like METTL3, METTL14, the IGF2BP family, ALKBH5, or FTO can enhance the efficacy of PD-1 blockade therapy in renal clear cell carcinoma, colorectal cancer, melanoma, NSCLC, breast cancer, hepatocellular carcinoma, and intrahepatic cholangiocarcinoma. Inhibition of tumor cell proliferation can be achieved by blocking METTL3, TRMT6/TRMT61A, the IGF2BP family, YTHDF2, or ALKBH5. Targeting “writers” such as METTL3 or METTL14 blocks the formation of m6A and thus can play a role in multiple tumor-suppressing or immune therapy-enhancing effects; however, this may also greatly impact normal physiological functions and lead to severe side effects. RNA modifications have a considerable number of “readers,” so targeting specific “readers” in specific tumors may achieve precise therapeutic effects with lower side effects, but it may need to be specific to certain tumors and may not be universally applicable. Targeting “erasers,” mainly FTO and ALKBH5, seems to primarily enhance PD-1 blockade therapy and inhibit tumor cells, being key to combined immunotherapy. However, targeting FTO and ALKBH5 results in the inability to remove m6A, which may also lead to severe side effects.

In this review, we primarily discuss the relationship between m6A, m1A, m5C, and m7G modifications and tumor immunity and immunotherapy, and summarized the regulation of immune cells and immune checkpoints, as well as drug treatment targeting RNA modification regulators and immune prediction based on the four RNA modifications. Research related to RNA modifications has the following limitations: m6A is the most abundant modification in mRNA and has been extensively studied; however, the physiological processes in which m6A is involved are not fully understood, and there may be undiscovered regulatory proteins related to m6A modification. The specific roles of m1A, m5C, and m7G in tumors have been insufficiently studied, possibly due to variations in modification sites and abundance. There are also limitations and challenges in tumor therapy related to RNA modifications: some proteins among RNA modification-related proteins have dual identities, and the same regulatory factor may have multiple roles, possessing both oncogenic and tumor-suppressive effects, requiring specific research to provide solutions. For example, knocking out METTL3 in melanoma can inhibit tumor development and increase the infiltration of CD8⁺ T cells (79), but another study has shown that the absence of METTL3 expression in macrophages can promote the growth and metastasis of melanoma and weaken the efficacy of PD-1 blockade (63). Additionally, RNA modifications are widely involved in physiological activities, and it is challenging to target RNA modifications in the tumor microenvironment without affecting normal physiological processes. Researchers have obtained drugs with high specificity through complex drug design and optimization, but this issue has not been resolved. Moreover, although many RNA modification regulators have been identified, only a few can be used as therapeutic targets for cancer treatment. Finally, current research is mainly focused on mouse models, and there may be differences in the physiological activities and tumor therapy based on mRNA between mice and humans, requiring more research to clarify the specific situation.

Despite certain achievements of drugs targeting RNA modifications in preclinical models, no related drugs have yet entered the clinical research phase. Future research in the following areas may be helpful: designing drugs to specifically target RNA modification regulators, especially in the tumor microenvironment, to reduce the impact on normal tissues; exploring appropriate drug dosages to balance efficacy and side effects; continuing drug safety research to assess the side effects of different targeted drugs; developing new therapeutic targets or drugs to improve treatment safety; and exploring the combined application of RNA modification-targeting drugs with other drugs in tumor therapy. In addition, tumor vaccines are also related to RNA modifications such as m6A and have therapeutic potential. Introducing encoded mRNA molecules into the bodies of tumor patients, translating them into proteins with anti-tumor effects in the body to trigger anti-tumor immune responses is a new treatment method (173). RNA modifications play a key role in cancer vaccines. Through RNA modifications, the immunogenicity of RNA vaccines can be eliminated, and the rapid translation of anti-tumor proteins may be promoted, improving anti-tumor effects. This direction has considerable potential for development. Furthermore, lifting the immune suppression caused by RNA modifications may help improve the efficacy of adoptive immunotherapy. By lifting the toxicity suppression and infiltration suppression of T and NK cells caused by RNA modifications, the corresponding tumor-killing ability can be restored, which can be used for ex vivo expansion and then re-introduced into the patient’s body to play an anti-tumor role. Lifting the immune suppression of the tumor microenvironment is beneficial for adoptive immune cells to clear the tumor. This could greatly enhance adoptive immunotherapy and holds value for research.