Studies have shown that ARID1A is involved in cell proliferation, migration, and apoptosis and plays a significant role in tumor inhibition (23).ARID1A expression is correlated with cell cycle changes, and the expression of ARID1A is highest in the G0-G1 phase, significantly reduced in the S and G2-M phases, and almost entirely absent in actively dividing cells (33). It was found that ARID1A could act directly or indirectly to regulate p21/WAF1 gene expression and thus affect cell proliferation. Mechanistically, p21/WAF binds to and inhibits the activity of the cell cycle protein CDK2/CDK4 complex, thereby contributing to cell arrest in the G1 phase (25). In addition, long-chain non-coding RNAs (lncRNAs) DGCR5 interacts with ARID1A, promoting the transcription of p21 and thus affecting cell proliferation (34). In addition, it was found that down-regulation of the ARID1A gene in gastric and breast cancer cells significantly enhanced cell proliferation. In contrast, the proliferation of cells replying to ARID1A gene expression was inhibited (35). In addition, ARID1A is closely related to cell invasion. It has been shown that the activation of epithelial-mesenchymal transition (EMT) and stem cell recognition pathways partially depends on the deletion of ARID1A. Knockdown of ARID1A enhanced migration and invasion of gastric cancer cells, and exogenous overexpression of ARID1A inhibited cell migration. In addition, the knockdown of ARID1A and cell adhesion protein (E-cadherin) were down-regulated, and the overexpression of ARID1A resulted in the up-regulation of E-cadherin expression (36). This suggests that low expression or deletion of ARID1A enhances migration and invasion of gastric cancer cells by down-regulating E-cadherin. Furthermore, the knockdown of ARID1A in hepatocellular carcinoma and breast cancer cells significantly enhanced cell invasion and migration. ARID1A was also found to play a tumor suppressor role in nasopharyngeal carcinoma (NPC), and deletion of ARID1A activated the AKT signaling pathway to promote tumor cell migration and invasion (37). ARID1A was negatively correlated with the expression of MCL-1, an anti-apoptotic protein in colorectal cancer (38), and ARID1A could inhibit tumor growth by down-regulating MCL-1 (39). Knockdown of ARID1A in non-small cell lung cancer (NSCLC) resulted in up-regulation of the expression of cycle-related proteins cyclinD1 and Bcl-2 and inhibition of apoptosis (34). The ARID1A gene may exert tumor suppressive effects by affecting the above three cellular processes.

Changes in the external biochemical environment and cellular metabolites can lead to DNA damage, leading to aberrant DNA replication and transcription processes, and inactivation of DNA damage repair pathways under replication stress will lead to cell death or cancer (40, 41). Chromatin remodeling complexes are crucial in DNA damage repair (42, 43). Studies have shown that loss of ARID1A function, such as deletion or mutation (insertion, deletion, nonsense mutation, etc.), affects DNA damage repair (44). SWI/SNF complexes containing ARID1A allow DNA repair proteins to access DNA damage sites (45) efficiently. Inactivation of the SWI/SNF complex by deletion or knockdown of ARID1A impairs DNA DSB repair, increases sensitivity to DNA damaging agents, and impairs γ-H2AX induction (46). In addition, inhibition of ARID1A reduces the recruitment of NHEJ factors (e.g., KU70/KU80 and the ATPase subunit of the SWI/SNF complex) from the DSB locus, thereby affecting the NHEJ process (47). And ARID1A knockout cells were unable to perform effective NHEJ repair after irradiation. It was found that ARID1A is also involved in repairing damaged DNA by homologous recombination (HR). Mechanistically, ARID1A is recruited to the DNA DSB site by interacting with the upstream kinase ATR.ARID1A also helps to recruit the ATPase subunit of the SWI/SNF complex to the DNA damage site. Deletion of ARID1A also impairs the G2/M DNA damage checkpoint (48). In summary, ARID1A promotes DSB end resection and helps maintain checkpoint signaling. Overall, ARID1A protects the genome by interacting with mechanisms of different DNA repair mechanisms.

Activation of the acylinositol triphosphate kinase (PI3K/AKT) signaling pathway is an essential mechanism for the progression of many cancers, and studies have shown that the role of ARID1A in tumors can be associated with activation of the PI3K signaling pathway (49, 50). In ovarian clear cell carcinoma (OCCC), PIK3CA activating mutations occur concurrently with ARID1A deletion mutations (51), and mutations in PIK3CA are present in 46% of ARID1A deletion tumors (52). Mutations in the ARID1A gene in clear cell carcinoma of the ovary activate the PI3K/AKT signaling pathway, which activates the PIK3CA gene and inactivates PTEN, leading to tumorigenesis (51). In addition, the phosphorylation level of AKT was significantly higher in ARID1A-deficient gastric cancer cells than in non-deficient cells, as well as frequent mutations in the PI3K/AKT signaling pathway (deletion of PTEN or activation of PIK3CA) in endometrial carcinomas with deficient ARID1A expression (53). In ARID1A-mutated ovarian cancer cells, the expression of PIK3IP1 was reduced, which led to the activation of the PI3K/AKT/mTOR pathway (50). The above results indicate that ARID1A mutation deletion tends to occur concurrently with PIK3CA gene mutation and activation of the PI3K/AKT signaling pathway, suggesting that small molecule inhibitors of our PI3K/AKT/mTOR signaling pathway could be used as potential targeted therapeutic agents to inhibit ARID1A mutation-deficient tumors.

ARID1A is involved in DNA mismatch repair, and ARID1A mutations were enriched in MSI-H tumors. It has now been shown that in endometrial and colorectal cancers, ARID1A deficiency is associated with MLH1 silencing due to promoter hypermethylation. It was found that ARID1A interacts with the mismatch repair (MMR) protein MSH2, recruiting MSH2 to chromatin, and when ARID1A undergoes inactivating mutations, it impairs MMR function, leading to increased microsatellite instability and mutations (54). Specifically, ARID1A recruits MSH2 to chromatin during DNA replication and supports mismatch repair (54). ARID1A inactivation attenuates mismatch repair and enhances mutation. Thus, ARID1A deficiency may generate the microsatellite instability genomic signature MSH2 by interacting with perturbations in M. Finally, in homozygous mice, cancers arising from ARID1A-deficient ovarian cancer cell lines exhibited high mutation loads, elevated PD-L1 expression, and increased numbers of lymphocytes in tumor infiltration. Using an anti-PD-L1 antibody significantly reduced tumor burden in ARID1A-deficient mice and prolonged survival in ARID1A wild-type ovarian cancer mice (54).

Although the interaction of ARID1A with MSH2 and the resulting microsatellite instability may be associated with ARID1A-deficient cancer responsiveness to immune checkpoint blockade, other mechanisms may also play a role. Alterations in ARID1A are significantly associated with more prolonged progression-free survival (PFS) after immune checkpoint inhibition in various human cancers, independent of TMB or MSI status (55). Although the mechanistic mechanism by which ARID1A mutations are associated with immune checkpoint blockade benefits is unknown, the interaction of EZH2 methyltransferase with ARID1A may be relevant to the immune response. Indeed, ARID1A interacts with EZH2 and inhibits the EZH2-mediated interferon (IFN) response, which is essential for the immune response (56). It has been shown that alterations in ARID1A are associated with checkpoint gene expression and other markers of immunogenicity in MSS colorectal cancer (57).

These results indicate that ARID1A, as the subunit with the highest mutation rate in the SWI/SNF chromatin remodeling complex family, plays a role in tumor through the regulation of the above biological processes, and has broad application prospects in tumor development and targeted therapy.

Mutations, down-regulation of expression, and deletion of the ARID1A gene can predict drug resistance and recurrence in various cancers. Chemoresistance in epithelial ovarian cancer patients was associated with reduced expression of ARID1A (58). Bladder cancer patients had P153A mutations in the ARID1A/1B subunit before and after resistance to pazopanib (59). In addition, ARID1A deletion activates ANXA1 protein expression in HER2+ breast cancer cells, which consequently induces resistance to trastuzumab through activation of AKT protein (60). ARID1A knockdown was found to promote the lung adenocarcinoma cell cycle and accelerate cell division (61). Mechanistically, ARID1A knockdown increases the phosphorylation levels of the oncogenic proteins EGFR, ErbB2, and RAF1, inducing bypass activation of the ErbB pathway, activation of the VEGF pathway, and changes in the expression levels of biomarkers of epithelial-mesenchymal transition, leading to disease progression that is insensitive to treatment with epidermal growth factor receptor tyrosine kinase inhibitors (61). ARID1A deletion in ovarian cancer can transcriptionally activate MRP2 protein expression after chromatin remodeling, leading to multidrug resistance to carboplatin and paclitaxel in ovarian cancer (62). Immunohistochemical analysis of ovarian cancer samples also confirmed a strong negative correlation between ARID1A expression and MRP2 expression (62), which provides an opportunity to overcome ARID1A deletion-induced chemoresistance in ovarian cancer by targeting MRP2.ARID1A deletion leads to resistance to PARP inhibitors in ovarian cancer patients (63), significantly increases chemoresistance in squamous cell carcinoma (64), promoting cell proliferation, metastasis, and resistance to sunitinib in clear cell renal cell carcinoma (65).

The above results indicate that the ARID1A-induced tumor drug resistance mechanism study has made significant progress ( Table 1 , Figure 2 ). However, it is still at the protein level without in-depth exploration of other genes and metabolic pathway changes induced by ARID1A alteration. In addition, the above conclusions should be validated in more cancer patients.

Data from a preclinical study showed that concurrent ARID1A-PIK3CA mutations in endometrial and ovarian cancers promote ovarian clear cell tumorigenesis through the tumorigenic inflammatory cytokine signaling pathway, which can be blocked by IL-6 inhibitors (77). Whereas mTOR inhibitors such as everolimus inhibit the increased phosphorylation of Akt caused by alterations in ARID1A (50, 78), mechanistically, ARID1A and EZH2 associate with the PIK3IP1 promoter; PIK3IP1 inhibits PI3K/Akt/mTOR signaling.ARID1A, although activating the expression of PIK3IP1, overrides the expression of EZH2, thereby reducing PIK3IP1 expression, and when ARID1A is absent, EZH2 silences PIK3IP1 (50, 78, 79).

Some clinical data suggest that alterations in ARID1A are found in 73% of EBV-positive gastric cancers and correlate with high PD-L1 expression (80, 81). Data showed that gastric cancers with EBV were more sensitive to immune checkpoint blockade (82). Alterations in ARID1A are associated with a higher tumor mutation burden in various cancers (55, 57, 83), making such cancers or subtypes potentially more susceptible to immunotherapy.

Preclinical data suggest that ARID1A deficiency sensitizes cancer cells to PARP in both in vitro and in vivo models with inhibitors (47, 48, 54, 84, 85), a process mechanistically linked to DNA damage repair (48, 54). However, the activity of PARP inhibitors in patients with altered ARID1A is unknown. In several breast and ovarian cancer cell lines, ARID1A deletion resulted in resistance to PARP inhibitory therapy (63, 86).

Some preclinical data suggest that inhibition of EZH2 may be effective in ovarian cancer, and mechanistically, inhibition of EZH2 histone methyltransferase activity is synthetically lethal in ARID1A mutant cancers (87, 88). Some preclinical data suggest that HDAC inhibitors, such as SAHA, inhibiting ARID1A mutation-deficient tumors may be effective. Mechanistically, ARID1A mutations are sensitive to pan-HDAC inhibitors such as SAHA. This is associated with enhanced growth inhibition induced by inhibition of HDAC2 activity in ARID1A mutant cells. HDAC2 interacts with EZH2 in an ARID1A state-dependent manner (89–91).

Preclinical data suggest a synthetic lethal interaction between targeted ATR therapy and ARID1A deletion mutations in ovarian clear cell carcinoma and other cancer cell lines (84, 92–94). It was found that knockdown of ARID1A containing remodeling complexes may be active in kras mutant colorectal cancer, mechanistically due to the involvement of ARID1A in K-RAS downstream (nuclear) signaling (95). Furthermore, some exclusivity between wild-type TP53 and ARID1A mutations suggests that restoration of wild-type p53 could benefit a combinatorial therapeutic approach (96). Similarly, in ER-positive breast cancer, loss of ARID1A determines resistance to the ER-degrading agent fulvestrant (97).

Taken together, high-frequency mutations of ARID1A in tumors lead to loss of function of the protein product and affect many necessary signals in tumorigenesis. For example, the PI3K/Akt/mTOR pathway, the KRAS pathway, DNA damage repair, EZH2, and immune function (including interactions with the MSH2 mismatch repair gene product). Compounds targeting these pathways may be active based on preclinical mechanistic cancer models and/or based on cancer patient data, including immune checkpoint blockade (anti-PD-1/PD-L1) and inhibitors of mTOR, PARP, ATR, EZH2, and HDAC (47, 48, 54, 84, 85). Studying the function of ARID1A is essential for a more precise understanding of the function of ARID1A, as well as matching cancer patients carrying ARID1A mutations with homologous drugs to optimize efficacy.

In conclusion, ARID1A is an essential component of the SWI/SNF chromatin remodeling complex and a tumor suppressor mutated or absent in many tumors. However, the specific oncogenic mechanism of ARID1A is still unclear, and the variability of ARID1A mutations in different tumors and the impact of ARID1A mutations on tumor prognosis need to be investigated in depth. An in-depth study of the function of the ARID1A gene and its specific molecular mechanism of action will help to increase the understanding of the biological role of chromatin remodeling complexes and provide new ideas for cancer diagnosis and treatment.

A link between SMARCA4 and autophagy proteins was found, with autophagy significantly reduced in cells lacking SMARCA4 expression (105, 106). Furthermore, SMARCA4 deficiency was directly associated with reduced critical autophagy-regulated genes, mainly ATG16L1, ATG7, AMBRA1, and WIPI2.ATG16L1 and ATG7 are direct components of the ATG16L1-ATG5-ATG12 coupling system, and these proteins contribute to LC3 phosphatidylethanolamine coupling and autophagy cell formation (105, 106).Ambra1 stabilizes the VPS34/Beclin1 complex by binding to Beclin1 and aids in the maturation of autophagic vesicles. It has been hypothesized that AMBRA1 is required for autophagy and that AMBRA1 deficiency correlates with suppression of autophagic flux (105, 106).WIPI2 is one of four WIPI proteins that make up the support proteins. WIPI acts as an autophagic signaling modality that binds to PI3P.WIPI aids in LC3 through the recruitment and binding of ATG16L1-ATG5-ATG12 complexes lipolysis to aid in the lipidation of LC3 (107). Studies have shown that SMARCA4 and p53 can interact (108, 109). p53 is a regulatory protein involved in autophagy and apoptosis (110, 111). It has been found that in colorectal cancer, SMARCA4 binds to SIRT-1 and enhances SIRT-1-mediated deacetylation of p53 at the K382 locus. Knockdown of SMARCA4 reduces the efficiency of SIRT-1 deacetylation and increases p53 stability (108). Another study showed that in CRC, SMARCA4 knockdown resulted in increased p53 expression and found that the SMARCA4/CHD4/HDAC1 complex regulates p53 transcription and stability (109). Retinoblastoma protein (RB) was found to interact with SMARCA4 and inhibit autophagy. RB is an oncogenic protein that induces cell cycle block at the G1 checkpoint, and RB together with SMARCA4 inhibits E2F1 activation (100, 112).SMARCA4/R B inhibits E2F1 to initiate autophagy. In addition, SMARCA4 has been found to enhance the inhibitory effect of RB on E2F and cyclinA. This suggests that SMARCA4 may further interact with RB to initiate autophagy. However, it has also been found that overexpression of SMARCA4 inhibits autophagy and that the inhibitory process is achieved through activation of the WNT/β-cyclin signaling pathway. This suggests the need to investigate further the role of this pathway in autophagy in the context of SMARCA4 and autophagy SMARCA4 in different cancer types. In colorectal cancer, inhibition of SMARCA4 led to activation of the JNK pathway (113). Furthermore, in a study of melanoma cells exposed to UV light, SMARCA4 and MITF inhibited apoptosis and enhanced the trans-responsive effects of the melanoma apoptosis inhibitor ML-IAP (114). In addition, a reduction in SMARCA4-related target genes, including the protein KLK2, a known prostate cancer apoptosis inhibitor, was found in SMARCA4-deficient prostate cancer cells (115). Furthermore, in pancreatic ductal adenocarcinomas (PDAs) with intraepithelial neoplasia (panIN) originating from KRAS mutant cells, SMARCA4 was found to promote the initiation of panIN and its progression to pancreatic cancer (116). BRG1 inhibition has been observed to prevent panIN formation and block panIN-derived PDAs by inducing apoptosis.SMARCA4 binds to the SOX9 promoter and helps to initiate the transcription of SOX9.SOX9 has been shown to affect tumor proliferation, and SOX9 knockdown has a pro-apoptotic effect (117). Although the mechanism by which BRG1 inhibits apoptosis has not been further investigated, SOX9 may be another factor that inhibits apoptosis through the Wnt/β-linker protein signaling pathway (117).

The above study suggests that SMARCA4 is related to autophagy and affects apoptosis. It is possible that SMARCA4 promotes or inhibits tumor cell growth. However, SMARCA4 has different roles in different stages of autophagy. More research on the mechanism of SMARCA4’s role in different cancer types is needed in future studies.

Studies have shown that SMARCA4 deletion or low expression can induce drug resistance in some tumors. Currently, clinically significant resistance to the combination of ibrutinib and vinblastine is observed in patients with mantle cell lymphoma (MCL). Mechanistic studies have found that knockdown of SMARCA4 in MCL cells leads to deletion of the transcriptional repressor of the BCL2L1 gene, which transcriptionally upregulates the BCL2L1 gene and increases the BCL2L1 protein, resulting in significant resistance to the combination of ibrutinib and vinotec in MCL cells (71). In addition, it was found that the knockdown of SMARCA4 in triple-negative breast cancer (TNBC) resulted in hyperactivation of YAP1 and upregulation of EMT (73). This suggests that SAMRCA4 mediates EMT-induced cisplatin resistance by activating YAP1 in TNBC.

In addition to the separate study of SMARCA4 described above, studies have shown that high-grade serous carcinoma (HGSC) cells with a low SMARCA4/high SMARCA2 expression phenotype are highly resistant to carboplatin. Analysis of the mechanism showed that altered chromatin accessibility in HGSC cells resulted in activation of the FGFR1- pERK1/2 signaling pathway and overexpression of the anti-apoptotic gene BCL2, causing chemoresistance (72). In addition, SMARCA4/2 deletion reduces the expression of the Ca2+ channel IP3R3, which impairs the transfer of Ca2+ from the endoplasmic reticulum to the mitochondria required to induce apoptosis, leading to cisplatin resistance in ovarian and lung cancer (74). It was shown that switching SMARCA4 to SMARCA2 induced acquired resistance to EZH2 inhibitors in ARID1A-mutated ovarian clear cell carcinoma. In cell death/apoptosis signaling, the anti-apoptotic gene BCL2 is a direct target of SMARCA4, and SMARCA4 deletion upregulates the expression of the anti-apoptotic gene BCL2 in EZH2 inhibitor-resistant cells, resulting in the upregulation of BCL2 mRNA and protein levels. Thus, SMARCA4 deletion is associated with reduced death/apoptosis features in drug-resistant cells (75). Interestingly, the ATPase activity of the SMARCA4/SMARCA2 subunit can also promote resistance to oxitinib treatment in EGFR-mutated lung adenocarcinomas by altering chromatin openness (76).

The above studies systematically elucidated the changes of SMARCA4 in drug resistance in various tumors ( Table 1 , Figure 4 ). However, they did not delve into why the changes of the SMARCA4 gene are different in different cancer types, whether it is just due to the differences in tumor types or whether all the various abnormalities of SMARCA4 induce drug resistance in tumors through different mechanisms, which needs to be explored further.

Immunotherapy with ICIs has emerged as a promising treatment modality for SMARCA4-altered cancers. Studies have reported that ICI treatment is associated with a significantly improved prognosis in SMARCA4-abnormalized non-small cell lung cancer (119). Some ICI responses were seen in patients with high tumor mutation load (TMB) and/or high PD-L1 expression by IHC, both of which are markers of response to immunotherapy, but other patients with low TMB and PD-L1 negativity still responded (120–124).

Studies have shown that SWI/SNF deficiency leads to elevated PRC2, a potential therapeutic target for SMARCA4 abnormalities. Indeed, SMARCA4-deficient cancer cells showed sensitivity to inhibition of methyltransferase EZH2, the catalytic subunit of PRC2. In SCCOHT cells, EZH2 inhibitors effectively suppressed the growth of xenografts in SCCOHT cell lines (125). Targeting histone modification complexes in SCCOHT patients has also shown promising therapeutic promise. Re-expression of SMARCA2 due to HDAC inhibitors in SCCOHT inhibits SCCOHT cell proliferation, including in an in vivo xenograft model of SCCOHT cells, which respond to the HDAC inhibitor quintet (126).

SCCOHT cells are also sensitive to cell cycle protein-dependent kinase 4/6 (CDK4/6) inhibition. Deletion of SMARCA4 leads to down-regulation of cell cyclin D1, which restricts CDK4/6 kinase activity in SCCOHT cells and leads to sensitization to CDK4/6 inhibitors in vitro and in vivo (127). A similar synthetic lethal interaction between SMARCA4 deficiency and CDK4/6 inhibition was identified in SMARCA4-deficient NSCLC (128).

SMARCA4 binds to BRCA1, an essential gene product for DNA damage repair (129). Furthermore, preclinical data suggest that SMARCA4-deficient lung cancer cells exhibit enhanced replication stress. Exposure to p-ATR inhibitors (impairing DNA repair) would lead to replication catastrophe in these cells (130). Similarly, low expression of SMARCA4 significantly correlates with platinum-based chemoselectivity. The rapid reactivity of NSCLC may be due to platinum-induced extensive DNA damage (69).

In conclusion, targeted therapy using EZH2, HDAC, CDK4/6 inhibitors, DNA damage repair inhibitors, and immunotherapy with ICIs is mechanistically reliable and has some application prospects.