m⁶A RNA modification is a dynamic reversible chemical modification that is regulated by “writers”, “erasers”, and “readers” ( Figure 1 ) (8). The core components of methyltransferase complex (MTC) include methyltransferase like 3 (METTL3), methyltransferase like 14 (METTL14), and Wilms tumor 1-associating protein (WTAP), which is the main factor of “writers” in inducing m⁶A methylation (9). In the process of MTC-induced mRNA methylation, METTL3 has an S-adenosyl methionine (SAM)-binding sequence. METTL14, as a structural support of RNA binding, binds H3K36me3 and enhances the catalytic ability of METTL3 (3, 4, 22–24). METTL3 and METTL14 form heterodimers that WTAP recruits into specific RNA sequences to methylate mRNA (25–27). Intriguingly, METTL3 acts as a “reader” to recruit eIF3, thereby promoting mRNA translation in the cytoplasm (28). Furthermore, the MTC contains RNA-binding motif protein 15 and 15B (RBM15/15B), Cbl proto-oncogene like 1 (CBLL1/HAKAI), zinc-finger CCCH domain-containing protein 13 (ZC3H13), methyltransferase like 5 (METTL5), tRNA methyltransferase activator subunit 11-2 (TRMT112), and KIAA1429 (25, 29–31). KIAA1429 is the most studied “writer” among them. KIAA1429, also known as vir-like m⁶A methyltransferase-associated protein (VIRMA), mediates m⁶A methylation preferentially at the 3’ UTR and near the stop codon (32, 33).

In contrast, the role of “writers” and “erasers” remove m⁶A modifications from mRNA. Fat mass and obesity-associated (FTO) and AlkB homolog 5 (ALKBH5) are both Fe²⁺ and 2-oxoglutarate (2OG)-dependent AlkB dioxygenases that mediate demethylation by recognizing the methylation sites on m⁶A and subsequently regulating the mRNA function (34, 35). The difference between the two is that FTO oxidizes m⁶A to N6-hydroxymethyladenosine (hm6A) and N6-formyladenosine (f6A) and then hydrolyzes it to adenosine, while ALKBH5 locates the nuclear speck and directly eliminates mRNA m⁶A methylation on the mRNA (24, 30, 36). ALKBH3 is like a demethylase, but it is preferable to demethylate tRNA (3, 13). “Writers” and “erasers” interact to regulate the abundance of m⁶A, while “readers”—as special RNA-binding proteins—regulate the stability, splicing, nuclear output, translation, or degradation of m⁶A-modified mRNA (37, 38).

“Readers” are mainly divided into three classes, YT521-B homology (YTH) domain family, heterogeneous nuclear ribonucleoproteins (hnRNPs), and insulin-like growth factor 2 mRNA-binding proteins (IGF2BPs). YTH domain family includes YTHDF1, YTHDF2, YTHDF3, YTHDC1, and YTHDC2 (10). YTHDF1 binds the eukaryotic initiation factor (eIF3) to the m⁶A site on mRNA and promotes mRNA translation. YTHDF2, on the contrary, recognizes m⁶A modification sites on the RNA 3’ UTR and recruits the CCR4-NOT deadenylase complex to promote mRNA degradation. YTHDC2 is also associated with RNA hydrolases to promote mRNA degradation (3, 24). YTHDF3 functions as an accelerator, promoting mRNA translation or degradation in collaboration with YTHDF1 or YTHDF2 (24, 34, 39–42). Intriguingly, fragile-X mental retardation protein (FMRP) has been shown to bind to YTHDF2, which inhibits the degradation of mRNA and promotes its stability (43). YTHDC1 binds the nuclear export protein serine/arginine splicing factor 3 (SRSF3) and inhibits SRSF10 in the nucleus, thereby facilitating mRNA splicing (37, 44). HNRNPs are a class of RNA-binding proteins that act as “m⁶A switches” to regulate RNA abundance. HNRNPA2B1 enhances the stability of m⁶A-modified mRNA, and both HNRNPC and HNRNPG can regulate pre-mRNA alternative splicing (23, 45–47). The IGF2BPs are the third type of “readers”. Unlike YTHDF2, IGF2BP1-3 promotes mRNA stability and translation by recruiting ELAV-like RNA-binding protein 1 (ELAVL1), which is also known as HuR (23, 38, 48). Furthermore, LRPPRC leucine-rich PPR-motif-containing protein (LRPPRC) can promote mRNA stability in the mitochondria, whereas Staphylococcal Nuclease and Tudor Domain Containing 1 (SND1) and Proline-Rich Coiled-Coil 2A (Prrc2a) such as IGF2BPs promote mRNA stability in m⁶A modification (24, 34, 49, 50).

LncRNA can be categorized into 5 categories in accordance with the different positions of its transcription genes: internal lncRNA (lincRNA), internal lncRNA, sense lncRNA, antisense lncRNA, and bidirectional lncRNA. Although lncRNAs lack ORF and cannot be used as templates for protein synthesis, they do play an important regulatory role in RNA transcription and post-transcriptional modification (52, 53). Past studies have demonstrated that m⁶A modification can regulate the function of lncRNA. On one hand, m⁶A modification regulates the interaction between lncRNA and specific DNA sites via the RNA-DNA triple helix structure; on the other hand, m⁶A modification affects the binding of “readers” to lncRNA (54–56). YTHDC1 binds m⁶A methylated long-noncoding RNA X inactivity-specific transcripts (XIST) and promotes XIST-mediated gene silencing through the modification of WTAP and RBM15/15B (57, 58). LncRNA metastasis-associated lung adenocarcinoma transcript 1 (MALAT1) positively correlates with the methylation modification of METTL3/METTL14. The binding of “m⁶A switch” HNRNPC to MALAT1 reduced with reducing expression of METTL3 or METTL14. In addition, MALAT1’s binding ability to HNRNPC/G was significantly enhanced following m⁶A methylation at A2577 of MALAT1. Both HNRNPC and HNRNPG were found to be capable of regulating MALAT1 in alternative splicing and in promoting the expression of post-lncRNA (59–61). METTL16 is a methyltransferase of U6 spliceosome small nuclear RNA (snRNA), but it directly interacts with the MALAT1 triple helix to promote the cleavage of MALAT1 (62). m⁶A-modified lncRNAs play an important function in different cancers. ALKBH5, for example, promotes tumorigenesis in GC by reversing the methylation of nuclear para-peckle assembly transcript 1 (NEAT1) (63). YTHDF2 promotes the degradation of FOXF1-adjacent noncoding developmental regulatory RNA (FENDRR) in endometrioid endometrial carcinoma (EEC) (64). The co-action of METTL3 and METTL14 with m⁶A modification promotes the stability of lncRNA LNCAROD in head and neck squamous cell carcinoma (HNSCC). The lncRNA differentiation antagonizing nonprotein coding RNA (DANCR) is essential in cancer development. IGF2BP2 has been demonstrated to recognize the m⁶A modification site on lncRNA DANCR, which promotes its stability (65).

MiRNA is a single-stranded non-coding RNA of length 21–25 nucleotides. MiRNA can form the RNA-induced silencing complex (RISC), which binds the 3’ UTR of target mRNA, resulting in translational inhibition or degradation of mRNA (66–68). MiRNA processing can mainly be categorized into two phases. The initial step is for DGCR8 and DROSHA to splice pri-miRNA generated by miRNA gene transcription to form pre-miRNA. After the pre-miRNA is exported from the nucleus to the cytoplasm via export protein 5 (XPO5), DICER further processes the pre-miRNA into mature miRNA (66, 69–71). Intriguingly, the maturation of miRNAs relies on m⁶A modification (6). It has been reported that the maturation of pri-miRNAs is m⁶A-dependent (72). METTL3 is a common m⁶A regulatory protein that associates with DGCR8 in bladder cancer cells to promote the maturation of miR-221/222 (73). METTL3 also promotes the cleavage of pri-miRNA-let-7e and pri-miR-17-92 to form let-7e-5p and miR-18a-5p (74). METTL14 positively regulates the maturation of the tumor suppressors miR-126, miR-375, and miR-19a, as well as inhibiting invasion and migration in several cancers (75–77). METTL3 and METTL14 cooperate to promote m⁶A RNA modification, but they have been reported to demonstrate opposite effects on the occurrence of the same tumor. On one hand, although METTL3 and METTL14 form a complex to jointly promote m⁶A RNA modification, METTL3 or METTL14 in this complex can have their direct targets. This target can act as a tumor suppressor or an oncogene, and the abnormal expression of METTL3 or METTL14 biases them toward the regulation of a certain target, thereby exhibiting the opposite effect. In addition, the complex formed by METTL3 and METTL14 promotes m⁶A RNA modification, and “Readers” further process RNAs in which m⁶A occurs to determine the state of these RNAs. Therefore, in different types of same cancer, owing to the differences in the expression of different “Readers”, METTL3 or METTL14 can have opposite effects in the same cancer type. In fact, the current understanding of this mechanism is unclear, warranting further research. Similarly, the demethylase FTO may influence miRNA stability by inhibiting the expression of miR-181b-3p and miR-576 by interacting with DGCR8 (78, 79). Various “readers” proteins promote miRNA maturation by recognizing the m⁶A-modified sites in addition to “writers” in m⁶A -modified regulatory proteins. During pancreatic cancer progression, NKAP leads to the overmaturation of miR-25 under overexpressed METTL3 m⁶A modification, thereby inhibiting PHLPP2 and activating the AKT-p70S6K signaling to promote cancer progression (80). Moreover, HNRNPA2B1 can interact with DGCR8 and DROSHA to promote their binding to miRNAs (81–83). HNRNPC binds miR-21 directly to promote miRNA maturation (84). m⁶A modification, however, can also inhibit miRNA processing (6). Through methylation, NSun2 inhibits the splicing of pri-miR125b2 to pre-miR125b2, which reduces the miR-125 expression (85, 86).

CircRNAs belongs to a class of circular non-coding RNAs with a closed circular structure. CircRNAs can be classified into the following 4 types: exonic circRNAs (ecircRNAs), intronic circRNAs (ciRNAs), exon-intron circRNAs (elciRNAs), and those formed by the circularization of viral RNA, tRNA, and shRNA (87). CircRNAs, as competitive endogenous RNAs (ceRNA), can sponge miRNA. CircRNA function can be influenced by m⁶A modification. Some circRNAs have protein-coding potential, while m⁶A modification can drive the translation process. Promoter eIF4G2 and YTHDF3, under the enhanced action of methyltransferase METTL3/14, start the translation of circRNA, while demethylase FTO inhibits the translation of circRNA (56, 88). CircCPSF6 is recognized and degraded by YTHDF2 during ALKBH5 demethylation (89). YTHDC1 has been reported to promote the nuclear output of circRNA NOP2/Sun RNA methyltransferase 2 (circNSUN2) (88).

Cancer progression is related to abnormal tumor proliferation, which is regulated by m⁶A modification. Ras is activated when a growth factor binds to the receptor, which in turn binds GTP to form Ras-GTP. The activated Ras affects multiple downstream proteins, including Ras-GTP, which then recruits and leads to Raf phosphorylation activation. The activated Raf kinase then binds to and activates the downstream MEK and ERK. After the phosphorylation-activated ERK is transferred from the cytoplasm to the nucleus, it can promote the phosphorylation activation of transcription factors such as c-Jun, MYC, and SP1, thereby inducing the expression of cell cycle and proliferation genes ( Figure 3 ) (120, 121). BATF2 is a tumor suppressor gene in GC that inhibits the ERK phosphorylation by binding with P53, while METTL3 methylation inhibits the BATF2 expression (98, 122, 123). In addition, KIAA1429 directly targets and promotes the expression of c-Jun, facilitates the transcription of downstream MMP3, MMP9, and cyclin D1, and causes GC abnormal proliferation ( Figure 3 ) (96, 98). Notably, METTL16 can promote the stability and high expression of cyclin D1 mRNA in this signaling pathway via m⁶A methylation, but whether cyclin D1 is the direct target of METTL16 requires further research (107). METTL3 promotes the MYC expression by m⁶A methylation and positively regulates the key components of the MYC pathway MCM5 and MCM6. Intriguingly, silencing HBXIP effectively suppressed the process, although further studies on the downstream targets of the HBXIP/METTL3/MYC axis are needed (1, 99). Similarly, the Wnt/β-catenin signaling pathway regulates abnormal GC proliferation ( Figure 3 ) (124). YTHDF1 can promote the translation of FZD7, a key molecule of the Wnt signaling pathway. FZD7 binds to the Wnt protein on the cell membrane, which promotes downstream β-catenin to enter the nucleus from the cytoplasm. β-catenin is a multifunctional protein that can not only act as an adaptor protein in the cytoplasm to promote cell adhesion but also bind to the transcription factor TCF/LEF in the nucleus to promote the transcription of downstream target genes HMGA2, cyclin D, CDC25A, COX2, Sox9, and VEGFA (101, 125–127). Therefore, YTHDF1 overexpression significantly promotes the activation of the Wnt pathway (101). Interestingly, YAP/TAZ regulates β-catenin in the Wnt pathway. YAP/TAZ can form the β-catenin destruction complex in the cytoplasm, resulting in β-catenin degradation. In the nucleus, YAP binds β-catenin to form a complex that promotes the expression of downstream target genes (128–130). An active Hippo signaling pathway results in MST1/2 and SAV1 phosphorylation activation of LATS1/2. Activated LATS1/2 further phosphorylates YAP/TAZ, preventing it from entering the nucleus, and therefore blocking the regulation of downstream target genes expression when the transcriptional co-activators YAP/TAZ link to TEAD ( Figure 3 ) (105, 110, 131–133). m⁶A methylation promotes abnormal proliferation in GC by regulating the expression of YAP in the Hippo signaling pathway. METTL3 induces methylation of YAP1 mRNA, and YTHDC2 recognizes m⁶A methylated YAP mRNA and promotes its translation (105, 110). The expression of METTL3 is positively correlated with the phosphorylation level of AKT. The downregulation of METTL3 resulted in low levels of p70S6K and cyclinD1, both downstream AKT (92). When the METTL3 is silenced, SOCS2 is unable to activate the JAK/STAT signal pathway, thereby inhibiting the proliferation, metastasis, and apoptosis in GC cells (97). Although YTHDF2 degradation of FOXC2 mRNA can inhibit the proliferation of abnormal GC, the upstream regulators in the YTHDF2/FOXC2 pathway have not been explored (111). m⁶A modification is implicated in several signaling pathways and mediates abnormal proliferation in GC. This finding suggests that m⁶A-modified mRNA in the regulation of abnormal proliferation is a potential therapeutic target with significant clinical implications.

The progression of cancer is marked by distant metastasis of tumor cells. The PI3K/AKT/mTOR signaling pathway is often activated during tumorigenesis. This pathway not only regulates cell proliferation, apoptosis, autophagy, and other physiological processes but is also closely related to cancer cell metastasis (134, 135). METTL14 overexpression has been reported in studies to inhibit the PI3K/AKT/mTOR pathway activation and metastasis in GC ( Figure 3 ) (100). METTL3 cooperates with YTHDF1 to promote SPHK2 mRNA translation, while m⁶A modification promotes GC metastasis via the METTL3/SPHK2/KLF2 axis, suggesting that the regulatory process is related to the poor prognosis in GC (106, 136). USP14 is a member of the USP ubiquitinase subfamily, and its expression in the GC tissues is higher than that in normal tissues. By knocking down USP14, it was found that the decreased expression of USP14 made GC cells sensitive to cisplatin by blocking the Akt or ERK signaling pathways. YTHDF1 could promote the translation of USP14 protein, which then promoted the tumorigenesis and metastasis of GC (109, 137). The PI3K/AKT/mTOR signaling pathway is activated in GC by HOXB13 overexpression, which is upregulated by FTO. In addition to “writers” and “readers”, the regulation of “erasers” in GC is equally significant. Past studies have shown that HOXB13 is highly expressed in GC cells due to increased demethylation of FTO and activation of the PI3K/AKT/mTOR signaling pathway ( Figure 3 ). Further research revealed that HOXB13 enhanced the expression of IGF-1R by acting on the promoter region of IGF-1R. The specific mechanism of HOXB13 controlling the IGF-1R expression warrants further analysis, although it is suggested that the FTO/HOXB13/IGF-1R/PI3K/Akt/mTOR axis should be explored as a potential therapeutic target for GC (104). FTO induces ITGB1 mRNA hypomethylation and promotes its expression in the FTO/ITGB1/p-FAK axis. Since ITGFB1 is a key signaling molecule in the signal transduction between cells and the extracellular matrix, increased ITGFB1 expression promotes metastasis in GC (108, 138, 139). Although METTL3 mediates PI3K/AKT/mTOR pathway activation, its direct target remains unknown. Therefore, positively searching for the downstream effectors of m⁶A regulatory proteins in the PI3K/AKT/mTOR signaling pathway can facilitate a better understanding of the role of this pathway in GC metastasis.

Abnormal glycolysis is one of the hallmarks of further deterioration in GC. Glycolysis is an important energy source for GC deterioration in the metabolic process (140–143). Only HDGF was always regulated by METTL3 in four GC cell lines with an overexpression or knockout of METTL3, while the expression of GLUT4 and ENO2 increased correspondingly in GC cells with a high expression of METTL3 ( Figure 3 ) (93). HDGF is essential in GC metastasis. The karyotype HDGF functions act as a transcriptional regulator to accelerate its expression in the promoter regions of GLUT4 and ENO2. The secreted HDGF promotes tumor angiogenesis by inducing VEGF (93, 144–146). The functional acetylation of H3K27me3 by p300 activates METTL3, which then promotes m⁶A methylation of HDGF mRNA. On the other hand, IGF2BP3 recognition of the m⁶A modification sites increases HDGF mRNA stability. HDGF promotes the expression of GLUT and ENO2 in the nucleus and initiates the glycolytic pathway in tumor cells (93). WTAP stabilized HK2 mRNA by binding to the 3’-UTR m⁶A site of HK2 mRNA and promoted its expression, making WTAP a novel target for GC therapy (21, 147). Moreover, ENO2 participates in the Warburg effect and together promotes aberrant glycolysis in GC under the regulation of m⁶A modification (93). We discovered that blocking tumorigenesis in the glycolysis process in GC can be employed as a new clinical treatment, and the proteins regulated by WTAP or those involved in m6A modification can be new drug targets.

In GC, EMT is the key step influencing distant metastasis (94). E-cadherin is a cell adhesion molecule that can promote adhesion between cells or between cells and the intercellular matrix. The emergence of EMT is marked by the deletion of the epithelial marker E-cadherin and the synthesis of the mesenchymal markers vimentin, N-cadherin, and smooth muscle actin (αSMA), which mark the occurrence of EMT (148). Past studies have demonstrated that METTL3 modifies ZMYM1 mRNA and promotes its high expression. ZMYM1 recruits the CtBP/LSD1/CoREST complex and inhibits the role of the E-cadherin promoter, resulting in EMT in GC ( Figure 3 ) (94). Snail is a zinc finger protein that inhibits the expression of E-cadherin (149). After modifying the snail protein via the HOXA10/TGF-β/Smad/METTL3/snail signaling pathway, METTL3 inhibited the E-cadherin expression, thereby facilitating GC metastasis ( Figure 3 ) (103). In another study, METTL14 and FTO knockdown activated the Wnt/PI3K-AKT signaling pathways ( Figure 3 ) (95). Wnt/PI3K-AKT pathways promote EMT in GC by inhibiting the expression of E-cadherin (95, 150, 151). However, it remains unclear whether m⁶A modification inhibits E-cadherin expression through pathways besides the snail pathway and Wnt/PI3K-AKT signaling pathway toward promoting the EMT generation.

Autophagy is a double-edged sword in tumorigenesis. On one hand, autophagy inhibits tumor growth by eliminating harmful cellular components and preventing DNA damage. Autophagy, on the other hand, can provide nutrition for tumor cells while also promoting cancer progression (152). The protective autophagy of tumor cells diminishes their drug sensitivity, which contributes to cancer progression. Since the PI3K/AKT/mTOR pathway is involved in the mechanism of autophagy, targeted modulation of this pathway has become an important approach for cancer therapy (153). The FTO expression was negatively correlated with disease-free survival in GC, and it not only regulated HOXB13 in the PI3K/AKT/mTOR signaling pathway but was also related to the mTORC1 expression ( Figure 3 ). When the ULK1 complex is activated, it directly phosphorylates Beclin-1 and ATG14L, thereby phosphorylating the downstream substrate VPS34 complex and promoting autophagy. mTORC1 functions in the general pathway by inhibiting the ULK1 complex, which suppresses cytoprotective autophagy and improves tumor drug sensitivity (102, 104, 152, 154–156). In the drug resistance study of GC cells, omeprazole inhibition of FTO could improve the inhibitory effect of 5-Fu, DDP, and TAX on GC cells as well as enhance the ability of mTORC1 to inhibit autophagy and improve the anti-tumor ability (102).

m⁶A modifications were involved in regulating other functions during GC progression. WTAP is an m⁶A-modified regulatory protein, and its high expression is closely related to m⁶A methylation. Interestingly, reduced WTAP expression has been linked to T lymphocyte infiltration (157). Apoptosis is a type of programmed cell death helpful to eliminate tumor cells. METTL3 reduces the expression of anti-apoptotic protein Bcl-2 through m⁶A methylation modification, while the pro-apoptotic protein Bax is relatively increased, which then activates Caspase-3 to promote the apoptotic pathway in GC ( Figure 3 ) (92).

During tumorigenesis, m⁶A methylation interacts with lncRNAs to regulate mRNA expression. Methyltransferase KIAA1429 inhibits the decay of lncRNA LINC00958, which can target GLUT1 to mediate aerobic glycolysis in GC. Therefore, KIAA1429 accelerates the glycolysis process by modifying LINC00958 with m⁶A methylation (116). ALKBH5 mediates lncNEAT1 demethylation, which affects the combination of lncNEAT1 and EZH2, thereby promoting GC development (63). METTL14 methylation enhanced LINC01320, which then absorbed miR-495-5p like a sponge. METTL3 promotes the RAB19 expression via the miR-495-5p/RAB19 axis, resulting in GC proliferation, migration, and invasion. However, the LINC01320/miR-495-5p/RAB19 axis in GC has not been thoroughly investigated, and this pathway warrants additional investigation (118). LncRNAs are not only modified by m⁶A, but they may also promote or inhibit m⁶A modification. As an oncogene, LINC00470 interacts with METTL3 and mediates the PTEN mRNA degradation in a YTHDF2-dependent manner ( Figure 4 ) (158). LncRNA NRON combined with ALKBH5 mediates Nanog mRNA demethylation of Nanog mRNA and promotes Nanog stability during tumorigenesis ( Figure 4 ) (115). LncRNA ARHGAP5-AS1 is also related to m⁶A regulatory protein. METTL3 combined with lncRNA ARHGAP5-AS1 to promote ARHGAP5 mRNA translation and induce protective autophagy in GC cells ( Figure 4 ) (112). In general, a high expression of m⁶A-modified lncRNA has been closely related to poor prognosis for GC, although the specific mechanism of lncRNA and m⁶A modification remains unclear. This observation emphasizes the need for additional research on m⁶A-modified lncRNAs.

The interaction between miRNA and m⁶A-regulatory protein promotes proliferation, metastasis, and invasion in GC ( Figure 4 ). Mature miR-17-92 inhibits the AKT/mTOR signaling pathway by affecting the expression of PTEN and TMEM127, which limits tumor proliferation and metastasis (114). METTL3 may promote pri-miR-17-92 processing through methylation, which is closely related to the poor prognosis in GC. The miR-660/E2F3 pathway is critical in the development of GC (119). He et al. reported that the m⁶A-modified tumor suppressor gene miR-660 could inhibit GC proliferation. Furthermore, miRNA is involved in the regulation of m⁶A-regulatory proteins. Both miR-338-5p and miR-4429 bind to METTL3 mRNA and inhibit its expression. SEC62 has a carcinogenic effect and is abundant in GC cells. It was discovered that METTL3 might interact with SEC62 mRNA via m⁶A modification, thereby allowing IGFBP1 to recognize and promote the stability of SEC62 mRNA. MiR-4429 inhibits the m⁶A methylation level of SEC62 mRNA by targeting METTL3, thus preventing GC progression (113). High EED expression in GC patients has been associated with poor prognosis. EED inhibited the splicing of METTL3 mRNA by downregulating the expression of miR-338-5p through H3K27me3. The increased METTL3 directly promoted the methylation modification and expression of CDCP1 mRNA, which, in turn, accelerated the occurrence in GC (117, 159). Some m⁶A regulatory proteins can precisely target specific miRNA sites, which can be employed as highly precise intervention targets. miRNAs bind to the mRNA of m⁶A regulatory proteins to regulate the m⁶A level of downstream signaling factors and provide a biomarker for future research.

It is well known that circRNA usually acts as ceRNA to competitively inhibit miRNA function. In GC cells, miR-30c-2-3p can inhibit the proliferation and metastasis of GC. Knocking down MTEEL14 reduced the m⁶A methylation level of circORC5, and increased the expression of circORC5, thereby reducing the content of miR-30c-2-3p (160). From this, we can know that METTL14 regulates miR-30c-2-3p/AKT1S1 axis through methylation modification of circORC5, thereby inhibiting the deterioration of GC. However, research on the connection of circRNA and m⁶A regulated is still needed.