Ovarian cancer remains the most lethal gynecologic malignancy in the developed world. Despite the unprecedented progress in understanding the genetics of ovarian cancer, cures remain elusive due to a lack of insight into the mechanisms that can be targeted to develop new therapies. SWI/SNF chromatin remodeling complexes are genetically altered in approximately 20% of all human cancers. SWI/SNF alterations vary in different histologic subtypes of ovarian cancer, with ARID1A mutation occurring in approximately 50% of ovarian clear cell carcinomas. Given the complexity and prevalence of SWI/SNF alterations, ovarian cancer represents a paradigm for investigating the molecular basis and exploring therapeutic strategies for SWI/SNF alterations. This review discusses the recent progress in understanding SWI/SNF alterations in ovarian cancer and specifically focuses on: (i) ARID1A mutation in endometriosis-associated clear cell and endometrioid histologic subtypes of ovarian cancer; (ii) SMARCA4 mutation in small cell carcinoma of the ovary, hypercalcemic type; and (iii) amplification/upregulation of CARM1, a regulator of BAF155, in high-grade serous ovarian cancer. Understanding the molecular underpinning of SWI/SNF alterations in different histologic subtypes of ovarian cancer will provide mechanistic insight into how these alterations contribute to ovarian cancer. Finally, the review discusses how these newly gained insights can be leveraged to develop urgently needed therapeutic strategies in a personalized manner.

Ovarian cancer is the fifth leading cause of cancer-associated mortality among women in the United States and is the most lethal gynecologic cancer in the developed world. Ovarian cancer is histologically and genetically diverse and can be broadly classified into three groups: epithelial, germ cell, and stromal cell tumors. More than 90% of ovarian cancers are epithelial ovarian cancers (EOC). EOCs can be broadly divided into type I and type II tumors (1). Type I tumors include clear cell, endometrioid, mucinous, and low-grade serous carcinomas and are often associated with identifiable precursor lesions. Clear cell and endometrioid EOCs are typically associated with endometriosis. The most common high-grade serous ovarian cancer (HGSOC) is type II EOCs. In contrast to type I EOCs, at the time of diagnosis, HGSOCs have typically spread to other tissues within the peritoneal cavity. Small-cell carcinoma of the ovary, hypercalcemic type (SCCOHT) is a highly malignant and aggressive tumor that typically occurs in women younger than 40 years old. Although SCCOHT was speculated to be epithelial, sex cord, germ cells, and neuroendocrine, it is now categorized as miscellaneous ovarian neoplasms.

Despite the great recent progress in the genomic understanding of the diverse subtypes of ovarian cancer, therapeutic approaches for patients with ovarian cancer have not evolved in parallel. The standard care remains surgical debulking followed by chemotherapy. Thus, a personalized approach based on the genetic make-up of different subtypes of ovarian cancer is urgently needed to improve patient outcomes. The development of such personalized therapeutic approaches relies on a combination of genetic profiling and a deep understanding of how certain genetic alterations contribute to malignant phenotypes of tumors and their associated microenvironments. This review will focus on frequent genetic alterations in SWI/SNF chromatin remodeling complexes in a set of genetically well-defined subtypes of ovarian cancer (Fig. 1; Table 1), offering both mechanistic insights and therapeutic implications.

Figure 1.

SWI/SNF subunit alterations in different histologic subtypes of ovarian cancer. ARID1A mutation in endometriosis-associated ovarian clear cell and endometrioid cancer typically leads to loss of ARID1A protein expression. CARM1 amplification/upregulation in HGSOC causes methylation of the SWI/SNF core subunit BAF155. SMARCA4 mutation and epigenetic silencing of SMARCA2 in SCCOHT, often result in the loss of both of the two mutually exclusive SWI/SNF catalytic ATPases, BRG1 and BRM.

Figure 1.

SWI/SNF subunit alterations in different histologic subtypes of ovarian cancer. ARID1A mutation in endometriosis-associated ovarian clear cell and endometrioid cancer typically leads to loss of ARID1A protein expression. CARM1 amplification/upregulation in HGSOC causes methylation of the SWI/SNF core subunit BAF155. SMARCA4 mutation and epigenetic silencing of SMARCA2 in SCCOHT, often result in the loss of both of the two mutually exclusive SWI/SNF catalytic ATPases, BRG1 and BRM.

Close modal
Table 1.

Alterations in subunits and regulators of SWI/SNF complexes in ovarian cancer

Histologic subtypesGenetic alterationAssociated subunit alteration
OCCC, OEC ARID1A mutation ARID1A loss 
SCCOHT SMARCA4 mutation BRG1 loss 
HGSOC CARM1 amplification BAF155 methylation 
Histologic subtypesGenetic alterationAssociated subunit alteration
OCCC, OEC ARID1A mutation ARID1A loss 
SCCOHT SMARCA4 mutation BRG1 loss 
HGSOC CARM1 amplification BAF155 methylation 

DNA exists in the form of nucleosomes by wrapping around histone proteins in cells. Thus, accessibility of nucleosomes is a key determinant of gene transcription and expression. Chromatin-remodeling complexes are evolutionally conserved multi-subunit epigenetic regulators. They control gene accessibility by sliding, dissociating, and replacing nucleosomes using energy derived from ATP hydrolysis. The ATP-dependent chromatin-remodeling complexes include five different classes: SWI/SNF, ISWI, CHD, INO80, and SWR1. Relevant to this review, SWI/SNF genetic alterations occur in approximately 20% of all human cancers (2), which ranks the complex as the most frequently mutated epigenetic regulator in human cancers (3).

SWI/SNF complexes were initially discovered by yeast genetic screening based on mating-type switch (SWI) and nutrient switch (SNF). Subsequent characterization revealed multi-subunit protein complexes that regulate gene transcription in an ATP-dependent manner from yeast to human. Human SWI/SNF complexes include BAF complex that contains either BAF250a (ARID1A) or BAF250b (ARID1B) subunits, and PBAF complex that contains BAF180 (PBRM1) and BAF200 (ARID2). Both BAF and PBAF complexes contain one of the two mutually exclusive catalytic ATPases, either BRG1 (encoded by SMARCA4) or BRM (encoded by SMARCA2). All SWI/SNF complexes contain core subunits such as BAF155 (encoded by SMARCC1), BAF170 (encoded by SMARCC2), and BAF47 (encoded by SMARCB1). In addition to these core subunits, additional accessory subunits are associated with SWI/SNF complexes, which makes the SWI/SNF complexes highly variable in a cellular context–dependent manner.

Mammalian SWI/SNF chromatin-remodeling complexes are implicated in a number of important biological processes such as transcription, replication, repair, recombination, and chromatin structure. However, the most extensively studied role of the complexes is gene expression regulation through altering transcription. It is well recognized that SWI/SNF complexes may function as either transcription activators or repressors in a context-dependent manner. SWI/SNF complexes can interact with both histones and DNA. For example, ARID subunits of SWI/SNF complexes bind to AT-rich DNA sequences, whereas a number of the SWI/SNF subunits interact with histones through domains such as bromodomains (e.g., the catalytic subunits BRG1 and BRM). In addition, AP-1 transcription factors such as FOS/JUN have been shown to recruit the BAF chromatin-remodeling complex to establish accessible chromatin (4). This allows external signaling to regulate gene transcription. In addition to promoters, SWI/SNF complexes are also recruited to enhancers and clusters of active enhancers (the so-called super-enhancers; ref. 5, 6). Interestingly, disease-causing mutations in the subunits of SWI/SNF complexes cause disassembly of most SWI/SNF complexes at promoters and typical enhancers, while having minimal effects on super-enhancer-associated complexes (5). This raises the possibility that the distribution pattern of SWI/SNF complexes with disease-causing inactivating mutations may promote tumorigenesis while maintaining cell identity. This may also explain why SWI/SNF mutations exhibit a high degree of tissue specificity.

ARID1A mutation in endometriosis-associated clear cell and endometrioid EOCs

Endometriosis-associated ovarian clear cell carcinomas (OCCC) and ovarian endometrioid carcinomas (OEC) are the second and third most common histologic subtypes that together account for approximately 25% of the epithelial ovarian cancer cases. OCCC ranks second as the cause of death from epithelial ovarian cancer and is associated with a poorer prognosis compared with other ovarian cancer subtypes (7). OCCC typically has a low response rate to platinum-based standard care for ovarian cancer. Radiotherapy is potentially beneficial for a subset of OCCC. However, for advanced stage disease, there is currently no effective therapy. Notably, in Japan, its prevalence is higher than in western countries, with an estimated incidence of approximately 25% of epithelial ovarian cancer.

On the basis of saturation of mutational analysis of TCGA datasets, ARID1A is the most frequently mutated epigenetic regulator across human cancers (3). OCCC and OEC are among the types of cancer with highest ARID1A mutation rate. The discovery of somatic ARID1A mutations in over 50% OCCC and 30% OEC represents a major advance in these histologic subtypes (8, 9). Over 90% of the ARID1A mutations observed in EOC are frame-shift or nonsense mutations that result in loss of ARID1A protein expression (8–10). Notably, loss of ARID1A correlates with late-stage disease and predicts early recurrence (11). Thus, there is an even greater need for targeted therapies that are selective for ARID1A-mutated EOC.

Despite ARID1A mutation correlation with the loss of ARID1A protein expression, only approximately 30% of the OCCCs with ARID1A mutations have both alleles affected (9). This suggests that ARID1A mutation might be haploinsufficient. ARID1A and ARID1B are two mutually exclusive subunits of the BAF complex. Notably, ARID1B is upregulated when ARID1A is inactivated (12, 13). Inactivation of ARID1B is synthetically lethal with ARID1A mutation (12). However, simultaneous loss of ARID1A and ARID1B has been observed (14). This raises the possibility that residue BAF complexes containing either ARID1A or ARID1B may sustain the survival of these cancer cells. Alternatively, the PBAF complex may compensate for the loss of BAF complexes in these cells.

In addition to ARID1A mutation, OCCC and OEC are characterized by activation of the PI3K/AKT pathway through activating PIK3CA mutation or inactivating the tumor suppressor, PTEN. Notably, ARID1A mutation often coexists with PIK3CA mutation or PTEN inactivation. This suggests that ARID1A mutation synergizes with PIK3CA mutation or PTEN loss to promote OCCC or OEC. Indeed, genetic mouse modeling studies show that ARID1A loss, PIK3CA activation, or PTEN loss alone is not sufficient to drive tumorigenesis (15, 16). However, a combination of ARID1A loss with either PIK3CA activation or PTEN loss results in the development of OCCC or OEC, respectively (15, 16). Interestingly, PIK3IP1, a suppressor of the PI3K/AKT signaling, is a target of ARID1A-mediated transcription activation (17). Thus, ARID1A loss may synergize with activation of PI3K/AKT signaling through further enhancing the oncogenic signaling.

TP53 and ARID1A mutation are typically mutually exclusive in OCCC and OEC (18). This mutational pattern typically occurs for TP53 and ARID1A mutations in other cancer types (19). This suggests that ARID1A and TP53 may function in the same pathway. The current literature supports that ARID1A regulates p53′s function in both cell growth arrest and apoptosis. ARID1A is a coactivator of p53 for the expression of its target genes such as p21 in mediating cell growth arrest (18). However, ARID1A regulates p53′s role in apoptosis by affecting lysine 120 acetylation of p53 (p53K120Ac) through transcriptionally repressing HDAC6 (19). p53K120Ac's mitochondrial localization promotes apoptosis through decreasing mitochondrial membrane potential. p53K120Ac levels are regulated by acetyltransferase TIP60 and deacetylase HDAC6. ARID1A inactivation upregulates HDAC6 to decrease p53K120Ac. Thus, ARID1A regulates the p53 pathway through two independent mechanisms: (i) controlling cell growth arrest by acting as a coactivator of p53 for its target cell-cycle–regulating genes such as p21; and (ii) modulating apoptosis through transcription-independent, mitochondria-localizing p53K120Ac posttranslational modification by upregulating HDAC6.

The presence of ARID1A mutation in contiguous atypical endometriosis, the precursor lesions of OCCC and OEC, supports that ARID1A mutation is an early event during tumorigenesis (9). The driver role of ARID1A mutation has been validated using genetic mouse models with conditional ARID1A knockout (15, 16). Interestingly, recurring ARID1A mutations are present in cases of deep-infiltrating endometriosis without concurrent malignancy (20). In contrast to EOC-associated endometriosis, deep-infiltrating endometriosis rarely leads to cancer. This suggests that intrinsic and/or tissue microenvironmental factors cooperate with ARID1A mutation to drive tumorigenesis in EOC-associated endometriosis. Revelation of these mechanistic insights will be critical in evaluating cancer risk associated with different types of endometriosis and developing potential cancer preventive measures for women with endometriosis.

SMARCA4 mutation in SCCOHT

SCCOHT is a rare but highly aggressive ovarian malignancy that occurs mostly in young women. The traditional treatment of SCCOHT is surgical removal followed by adjuvant chemotherapy. Most adjuvant chemotherapy combinations include a platinum agent, but the prognosis of the disease remains dismal. In addition, radiotherapy is increasingly being incorporated, but the survival benefit remains inconclusive.

Notably, SCCOHT is a monogenic disease with SMARCA4 being the likely driver mutation (21–23). SMARCA4 encodes the BRG1 catalytic subunit of BAF complexes. BRG1 acts as a tumor suppressor in these tumors. In contrast to ARID1A mutation in OCCCs, both alleles of the SMARCA4 gene are inactivated in SCCOHT (21, 23). This biallele inactivation can be due to one germline and one somatic mutation, loss of heterozygosity of the wild-type allele with a germline or somatic genetic inactivation, or two separate somatic mutations. Despite no family history of the disease, up to 50% of the patients carry a germline mutation (22).

Similar to the mutual exclusivity between ARID1A and ARID1B, BRG1 and BRM are the mutually exclusive catalytic subunits of SWI/SNF complexes. Notably, cells with BRG1 loss are compensated by residual BRM-containing SWI/SNF complexes (24). However, BRM is also absent in SCCOHT due to epigenetic silencing (25). Both BRG1 and BRM are typically lost simultaneously in SCCOHT. Thus, SCCOHT cells are likely devoid of SWI/SNF complexes. It will be interesting to examine whether the survival of these cells depends on the other non–SWI/SNF chromatin-remodeling complexes.

Emerging evidence suggests that SCCOHT is highly immunogenic. The majority of SCCOHT are associated with expression of PD-L1, an immunosuppressive checkpoint protein, in both tumor and stromal cells such as macrophages (26). This correlates with strong T-cell infiltration. Thus, the PD-L1 pathway may act as an adaptive immune-resistant mechanism in SCCOHT. These findings suggest that although SCCOHT is a monogenic disease with low mutational burden, it is a highly immunogenic tumor type.

HGSOC is the most common subtype (>70% of EOC cases) and accounts for the majority of EOC-associated mortalities. Analysis of patients with HGSOC from TCGA revealed that CARM1 is amplified in >10% and overexpressed in an additional approximately 10% of the spontaneous HGSOC (27). In comparison, somatic BRCA1/2 mutations occur in approximately 3%–4% of these cases for each gene, which are among the most commonly mutated genes in HGSOC. HGSOC is genetically heterogeneous. Thus, it is imperative that therapeutic strategies must be personalized by targeting distinct molecular subsets of HGSOC.

CARM1, also known as PRMT4, is a type I protein arginine methyltransferase (PRMT) that asymmetrically dimethylates protein substrates on arginine residues. CARM1 is located at 19p13.2. Emerging evidence suggests that CARM1 functions as an oncogene in human cancers. High levels of CARM1 expression have been observed in several cancer types, including breast, colon, and prostate. Interestingly, CARM1 amplification does not typically occur in HGSOC with mutations in BRCA1/2. Thus, there is an even greater need for developing therapeutic approaches that correlate with CARM1 status. This is because platinum-based chemotherapy, the current standard of care, and emerging treatment with PARP inhibitors are typically more effective in patients with BRCA1/2 inactivation.

SWI/SNF subunits such as ARID1A are not mutated in HGSOC where CARM1 is amplified (27). Interestingly, BAF155, a core subunit of the SWI/SNF complex is a substrate of CARM1 (28). Indeed, BAF155 methylation (BAF155Me) is regulated by CARM1 in HGSOCs (27). Thus, SWI/SNF complexes are regulated by CARM1 in HGOSCs.

Since the discovery of ARID1A mutation in endometriosis-associated OCCC and OEC, and SMARCA4 mutation in SCCOHT, several targeted therapeutic strategies have been explored in preclinical settings. Some of these newly discovered approaches are entering clinical trials (Table 2). These targeted approaches can be broadly divided into the following categories: (i) epigenetic synthetic lethality; (ii) activation of proapoptotic role of wild-type p53; (iii) checkpoint blockaded immunotherapy; (iv) DNA damage signaling inhibitors; and (v) cellular kinase signaling pathways.

Table 2.

A list of immediately translatable therapeutic targets and their representative targeting agents in SWI/SNF–altered ovarian cancer

Histologic subtypesGenetic alterationTherapeutic targetRepresentative agent(s)Most advanced status
OCCC and OEC ARID1A mutation EZH2 Tazemetostat Phase II 
  PD-L1 Avelumab, atezolizumab, and durvalumab Approved 
  ATR AZD6738, VX-970 Phase II 
  BRD2 Apabetalone Phase III 
  HDAC2 Vorinostat Approved 
  HDAC6 Ricolinostat Phase II 
  PI3K Copanlisib, idelalisib Approved 
  AKT Ipatasertib, perifosine Phase III 
  Tyrosine kinases Dasatinib Approved 
SCCOHT SMARCA4 mutation EZH2 Tazemetostat Phase II 
  PD-1 Nivolumab, pembrolizumab Approved 
  Tyrosine kinases Ponatinib Approved 
HGSOC CARM1 amplification EZH2 Tazemetostat Phase II 
Histologic subtypesGenetic alterationTherapeutic targetRepresentative agent(s)Most advanced status
OCCC and OEC ARID1A mutation EZH2 Tazemetostat Phase II 
  PD-L1 Avelumab, atezolizumab, and durvalumab Approved 
  ATR AZD6738, VX-970 Phase II 
  BRD2 Apabetalone Phase III 
  HDAC2 Vorinostat Approved 
  HDAC6 Ricolinostat Phase II 
  PI3K Copanlisib, idelalisib Approved 
  AKT Ipatasertib, perifosine Phase III 
  Tyrosine kinases Dasatinib Approved 
SCCOHT SMARCA4 mutation EZH2 Tazemetostat Phase II 
  PD-1 Nivolumab, pembrolizumab Approved 
  Tyrosine kinases Ponatinib Approved 
HGSOC CARM1 amplification EZH2 Tazemetostat Phase II 

NOTE: The most advanced status of the representative agents is listed, regardless of the approved or clinical trial indications.

Epigenetic synthetic lethality

Inhibition of polycomb activity.

The antagonistic roles of SWI/SNF and polycomb proteins in gene transcription were initially suggested from genetic studies in Drosophila (29). In mammalian cells, recruitment of BAF complexes results in a rapid eviction of polycomb complexes in an ATP-dependent manner (30, 31). This suggests a synthetic lethality between inactivation of the SWI/SNF complex and inhibition of polycomb activity. EZH2, the catalytic subunit of polycomb repressive complex 2 (PRC2), silences gene expression by generating the lysine 27 trimethylation mark on histone H3 (H3K27Me3). Highly specific small-molecule EZH2 inhibitors have been developed (32–34). Inhibition of EZH2 activity by these newly developed small-molecule inhibitors suppresses the growth of ARID1A-mutated OCCC or OEC (17, 35, 36) and SMARCA4-mutated SCCOHT cells (37). In contrast, EZH2 inhibitors have minimal effects on the growth of ARID1A or SMARCA4 wild-type cells. In contrast to chemotherapy or existing targeted therapies, EZH2 inhibitors' growth-inhibitory effects take longer treatment to achieve. This feature of EZH2 inhibitors may reflect the requirement of cell division to dilute H3K27Me3 to a threshold to be effective. The best assay to achieve this effect is to perform the experiments either in three-dimensional cultures using Matrigel extracellular matrix or in anchorage-independent growth in soft agar.

Mechanistically, ARID1A activates, whereas EZH2 inhibits, ARID1A/EZH2 direct target genes (17). PIK3IP, an ARID1A/EZH2 target PIK3K inhibitor, plays a key role in the observed selectivity (17). There is evidence to suggest that noncatalytic activity of PRC2 may also contribute to the observed selectivity against SWI/SNF inactivation (35). However, catalytic activity–independent function of PRC2, such as genetic knockdown of EZH2, suppresses the growth of ovarian cancer cells regardless of the mutational status of SWI/SNF complexes. Thus, only the catalytic activity of PRC2 may constitute a therapy window in selectively targeting cancers with mutations in SWI/SNF complexes with precision.

Epigenetic gene silencing is often achieved by a coordinated action of histone-modifying corepressors. HDAC2 interacts with EZH2 in an ARID1A status–dependent manner (13). HDAC2 functions as a corepressor of EZH2 to suppress the expression of EZH2/ARID1A target tumor suppressor genes such as PIK3IP1 to inhibit proliferation and promote apoptosis. Indeed, growth inhibition and apoptosis induced by HDAC2 and EZH2 inhibition are comparable in ARID1A-mutated cells. Consistently, ARID1A mutation confers sensitivity to pan-HDAC inhibitors such as SAHA in OCCC. This is correlated with enhanced growth suppression induced by the inhibition of HDAC2 activity in ARID1A-mutated cells. These studies highlight the immediate translational potential by repurposing FDA-approved pan-HDAC inhibitors for ARID1A-mutated ovarian cancers.

Inhibition of BRD2 activity.

BRD2 is a member of the bromodomain and extra terminal domain (BET) family of proteins. In an unbiased screen, inhibition of BRD2 was predominantly lethal in ARID1A-mutated OCCC cells (38). This correlates with a suppression of the expression of multiple SWI/SNF subunits including the ARID1A mutually exclusive subunit ARID1B, supporting the notion that sensitization of ARID1A-mutated cell to BRD2 inhibition is due to synthetic lethality with ARID1A mutation and ARID1B suppression. Indeed, small-molecule BET inhibitors suppress proliferation of ARID1A-mutated OCCCs. However, the response to BET inhibitors is heterogeneous and does not have a perfect separation based on ARID1A mutational status. This is consistent with the complex nature of BET inhibitor–induced changes in transcriptional profiles.

Activation of the proapoptotic role of wild-type p53

Mutual exclusivity between ARID1A and TP53 mutations is observed across human cancer types (16, 19). This raises the possibility of targeting wild-type p53 in ARID1A-mutated cancers. ARID1A inactivation upregulates HDAC6, and HDAC6 directly deacetylates the apoptosis-promoting p53K120Ac posttranslational modification (19). p53K120Ac selectively regulates apoptosis but does not affect the expression of cell-cycle–regulatory p53 target genes such as CDKN1A (19). Thus, ARID1A mutation contributes to inactivation of p53′s apoptosis-promoting function by suppressing apoptosis-promoting p53K120Ac. HDAC6 activity is essential in ARID1A-mutated ovarian cancers. Inhibition of HDAC6 activity using a clinically applicable small-molecule inhibitor significantly improved the survival of mice bearing ARID1A-mutated tumors (19). This correlated with the suppression of growth and dissemination of ARID1A-mutated, but not wild-type, tumors. These findings also suggest that stabilizers of wild-type p53 (such as Nutlin) may have a similar therapeutic benefit as inhibition of HDAC6 in ARID1A-mutated cells. However, knockdown of other SWI/SNF subunits such as BRG1 does not affect HDAC6 inhibitor sensitivity (19). This is due to a compensation of BRG1 loss by the mutually exclusive catalytic subunit BRM on the HDAC6 gene promoter (19). This suggests that HDAC6 upregulation is unique to ARID1A inactivation, likely due to nonredundant roles played by ARID1A, ARID1B, and ARID2 in regulating gene transcription.

Checkpoint blockaded immunotherapy

In SMARCA4-mutated SCCOHT, PD-L1 is expressed in both tumor and stromal cells, and strong T-cell infiltration is observed in the majority of tumors (26). Consistently, emerging clinical evidence suggests that checkpoint blockades such as anti-PD-1 are effective in SCCOHTs (26). This suggests that inactivation of SWI/SNF complexes may sensitize tumors to checkpoint blockade immunotherapy. In clear cell renal cell cancer (ccRCC), patients who responded positively to anti-PD-1 or anti-PD-L1 therapy often carry loss-of-function mutation in the PBRM1 subunit of the PBAF complex (39). In an independent loss of function screen in melanoma cells, PBAF-specific BRD7, ARID2, and PBRM1 were identified as positive hits in conferring susceptibility to T-cell–mediated killing (40). The PBAF complex contributes to cancer cell immune resistance through control of IFNγ-stimulated gene expression. Notably, PBAF cooperates with EZH2 in silencing IFNγ target gene expression. Indeed, EZH2 inhibition enhances checkpoint blockage immunotherapy and the recruitment of cytotoxic T cells into the tumor microenvironment (41, 42).

Interestingly, ARID1A interacts with the mismatch repair (MMR) protein, MSH2, and ARID1A inactivation impairs MMR (43). Similar findings were also observed when the catalytic subunit SMARCA4 was depleted, suggesting that the SWI/SNF complex may be required for MMR. Consistently, ARID1A mutations are associated with a high mutation load in the TCGA datasets and ARID1A mutations are enriched in tumors with microsatellite instability. Notably, MMR deficiency is linked to the susceptibility to immune checkpoint blockade due to an increase in neoantigen load and tumor-infiltrating lymphocytes. Indeed, in an ovarian cancer syngeneic mouse model, ARID1A inactivation renders ovarian tumors more sensitive to an anti-PD-L1 treatment. However, anti-PD-L1 treatment only marginally improved the survival of ARID1A-deficient mice (43). This suggests that combinational approaches are necessary to improve the efficacy of checkpoint blockade therapy in ARID1A-mutated cancers.

DNA damage signaling inhibitors

ARID1A inactivation sensitizes cells to inhibit the DNA damage checkpoint kinase, ATR, by clinically applicable inhibitors (44). This correlates with defects in topoisomerase 2A and cell-cycle progression induced by ARID1A inactivation. Consequently, ATR inhibition triggers premature mitotic entry, genomic instability, and apoptosis. Consistently, ARID1A is recruited to DNA double-strand breaks through its interaction with ATR (45). In HCT116 colon cancer cell line with or without ARID1A knockout models, ARID1A inactivation leads to impaired checkpoint activation and sensitizes cells to DNA damage–inducing treatment such as PARP inhibitors. In addition to these targeted approaches, there is evidence to suggest that radiotherapy might be beneficial for a subset of patients with OCCC. However, it is unclear whether ARID1A mutation can predict response to radiotherapy in patients with OCCC.

Cellular kinase signaling pathways

Consistent with the finding that ARID1A mutation and PI3K/AKT signaling are epistatic in driving OCCC and OEC (15, 16), ARID1A positively regulates the expression of PIK3IP1, an inhibitor of PI3K (17). ARID1A mutational status in OCCCs correlates with an increase in phosphor-AKT levels. Several studies show that ARID1A-mutated cells are more sensitive to inhibition of PI3K/AKT signaling using small-molecule inhibitors (15). In addition, ARID1A cooperates with PI3K/AKT–regulated NFκB signaling to promote the expression of proinflammatory cytokines such as IL6 (46). Inhibition of NFκB activity or neutralization of the secreted IL6 is tumor suppressive in ARID1A-inactivated OCCCs (15, 46).

In a focused drug screen, tyrosine kinase inhibitor, dasatinib, was identified as a synthetic lethal drug in ARID1A-mutated OCCC cells (47). The sensitivity of ARID1A-inactivated cells to dasatinib is associated with G1–S cell-cycle checkpoint and is p2-1 and pRb-dependent. YES1 kinase plays a key role in the observed growth inhibition induced by dasatinib in ARID1A-inactivated cells. In addition, FDA-approved, tyrosine kinase inhibitor, ponatinib, was identified to show antitumor activity in SCCOHT with SMARCA4 mutations (48). This is due to the inhibition of multiple receptor tyrosine kinases by ponatinib.

Although HGSOCs typically lack genetic mutations in the subunits of SWI/SNF complexes, recent discovery suggests that CARM1 regulates the antagonism between SWI/SNF and PRC2 complexes (27, 28). CARM1-mediated methylation of the SWI/SNF subunit BAF155 leads to the switch from BAF155 to EZH2 at the promoters of the proapoptotic EZH2/BAF155 target genes. Indeed, CARM1-upregulated HGSOCs are selectively sensitive to EZH2 inhibitors (27).

Thus, in three different subtypes of ovarian cancer, the antagonism between SWI/SNF and PRC2 complexes are altered in three distinct manners, namely ARID1A mutation in OCCC and OEC, SMARC4 mutation in SCCOHT, and CARM1 upregulation in HGSOC. The molecular basis of the observed histologic subtype specificities of SWI/SNF alterations is intriguing. One possibility is that this reflects different roles played by these SWI/SNF subunits during tissue development and differentiation. However, for potential therapeutic strategies, these different flavors of antagonism between SWI/SNF and PRC2 complexes can all be leveraged using small-molecule EZH2 inhibitors.

Since the discoveries of specific SWI/SNF subunit mutations in various histologic subtypes of ovarian cancer, substantial progress has been made in both understanding the mechanistic basis of how these alterations contribute to tumorigenesis and the development of mechanism-guided precision medicine. Some of these newly discovered approaches are moving into clinical development (Table 2).

Previous clinical experience with targeted therapy and emerging evidence with checkpoint blockade immunotherapy show that one single agent will not likely deliver a cure for cancer due to intrinsic and/or acquired resistance. Thus, combinatory therapies offer the best hope for eventual defeat of ovarian cancer with SWI/SNF alterations, one subtype at a time. However, these combinatory strategies will be personalized on the basis of the genetic and immunologic makeup of a patient's tumor. Given the emerging potential of immunotherapy and the power of epigenetic modulation of immune response in tumor microenvironments, a combination of epigenetic and immunologic therapies will likely be most promising.

EZH2 inhibition will not only induce synthetic lethality in ovarian tumor cells with ARID1A mutation, SMARCA4 mutation, or CARM1 amplification, but also enhance the recruitment of cytotoxic T cells to the tumor microenvironment (42). Likewise, HDAC6 inhibitors induce apoptosis in ARID1A-mutated tumor cells (19), while enhancing T-cell activation and improving the function of antigen-presenting cells to boost antitumor immunity (49). Similarly, BET inhibitors suppress PD-L1 expression and enhance T-cell function (50). In contrast, targeted small-molecule inhibitors of cell signaling pathways or DNA damage agents may negatively impact tumor immunoenvironments. In summary, we propose that a combination of epigenetic inhibitors (e.g., EZH2, HDAC6, pan-HDAC, or BET inhibitors) and immunotherapy such as checkpoint blockades (e.g., anti-PD-L1 or anti-PD1) has significant potential in transforming the clinical management of subtypes of ovarian cancer with SWI/SNF alterations. These mechanistic and clinical insights could serve as a paradigm for developing therapeutics for other cancer types with SWI/SNF alterations, the most frequent epigenetic change in human cancers.

No potential conflicts of interest were disclosed.

We apologize to those whose important work we cannot cite in this review due to imposed limitation on the number of allowed references. This work was supported by US NIH grants (R01CA160331, R01CA163377, and R01CA202919 to R. Zhang) and US Department of Defense (OC140632P1 and OC150446 to R. Zhang).

1.
Kurman
RJ
,
Shih
IeM
. 
The dualistic model of ovarian carcinogenesis: Revisited, revised, and expanded
.
Am J Pathol
2016
;
186
:
733
47
.
2.
Kadoch
C
,
Hargreaves
DC
,
Hodges
C
,
Elias
L
,
Ho
L
,
Ranish
J
, et al
Proteomic and bioinformatic analysis of mammalian SWI/SNF complexes identifies extensive roles in human malignancy
.
Nat Genet
2013
;
45
:
592
601
.
3.
Lawrence
MS
,
Stojanov
P
,
Mermel
CH
,
Robinson
JT
,
Garraway
LA
,
Golub
TR
, et al
Discovery and saturation analysis of cancer genes across 21 tumour types
.
Nature
2014
;
505
:
495
501
.
4.
Vierbuchen
T
,
Ling
E
,
Cowley
CJ
,
Couch
CH
,
Wang
X
,
Harmin
DA
, et al
AP-1 transcription factors and the BAF complex mediate signal-dependent enhancer selection
.
Mol Cell
2017
;
68
:
1067
82
.
5.
Wang
X
,
Lee
RS
,
Alver
BH
,
Haswell
JR
,
Wang
S
,
Mieczkowski
J
, et al
SMARCB1-mediated SWI/SNF complex function is essential for enhancer regulation
.
Nat Genet
2017
;
49
:
289
95
.
6.
Mathur
R
,
Alver
BH
,
San Roman
AK
,
Wilson
BG
,
Wang
X
,
Agoston
AT
, et al
ARID1A loss impairs enhancer-mediated gene regulation and drives colon cancer in mice
.
Nat Genet
2017
;
49
:
296
302
.
7.
Chan
JK
,
Teoh
D
,
Hu
JM
,
Shin
JY
,
Osann
K
,
Kapp
DS
. 
Do clear cell ovarian carcinomas have poorer prognosis compared to other epithelial cell types? A study of 1411 clear cell ovarian cancers
.
Gynecol Oncol
2008
;
109
:
370
6
.
8.
Jones
S
,
Wang
TL
,
Shih
IeM
,
Mao
TL
,
Nakayama
K
,
Roden
R
, et al
Frequent mutations of chromatin remodeling gene ARID1A in ovarian clear cell carcinoma
.
Science
2010
;
330
:
228
31
.
9.
Wiegand
KC
,
Shah
SP
,
Al-Agha
OM
,
Zhao
Y
,
Tse
K
,
Zeng
T
, et al
ARID1A mutations in endometriosis-associated ovarian carcinomas
.
N Engl J Med
2010
;
363
:
1532
43
.
10.
Guan
B
,
Gao
M
,
Wu
CH
,
Wang
TL
,
Shih
IeM
. 
Functional analysis of in-frame indel ARID1A mutations reveals new regulatory mechanisms of its tumor suppressor functions
.
Neoplasia
2012
;
14
:
986
93
.
11.
Ye
S
,
Yang
J
,
You
Y
,
Cao
D
,
Huang
H
,
Wu
M
, et al
Clinicopathologic significance of HNF-1beta, AIRD1A, and PIK3CA expression in ovarian clear cell carcinoma: a tissue microarray study of 130 cases
.
Medicine
2016
;
95
:
e3003
.
12.
Helming
KC
,
Wang
X
,
Wilson
BG
,
Vazquez
F
,
Haswell
JR
,
Manchester
HE
, et al
ARID1B is a specific vulnerability in ARID1A-mutant cancers
.
Nat Med
2014
;
20
:
251
4
.
13.
Fukumoto
T
,
Park
PH
,
Wu
S
,
Fatkhutdinov
N
,
Karakashev
S
,
Nacarelli
T
, et al
Repurposing pan-HDAC inhibitors for ARID1A-mutated ovarian cancer
.
Cell Rep
2018
;
22
:
3393
400
.
14.
Coatham
M
,
Li
X
,
Karnezis
AN
,
Hoang
LN
,
Tessier-Cloutier
B
,
Meng
B
, et al
Concurrent ARID1A and ARID1B inactivation in endometrial and ovarian dedifferentiated carcinomas
.
Mod Pathol
2016
;
29
:
1586
93
.
15.
Chandler
RL
,
Damrauer
JS
,
Raab
JR
,
Schisler
JC
,
Wilkerson
MD
,
Didion
JP
, et al
Coexistent ARID1A-PIK3CA mutations promote ovarian clear-cell tumorigenesis through pro-tumorigenic inflammatory cytokine signalling
.
Nat Commun
2015
;
6
:
6118
.
16.
Guan
B
,
Rahmanto
YS
,
Wu
RC
,
Wang
Y
,
Wang
Z
,
Wang
TL
, et al
Roles of deletion of Arid1a, a tumor suppressor, in mouse ovarian tumorigenesis
.
J Natl Cancer Inst
2014
;
106
pii: dju146
.
17.
Bitler
BG
,
Aird
KM
,
Garipov
A
,
Li
H
,
Amatangelo
M
,
Kossenkov
AV
, et al
Synthetic lethality by targeting EZH2 methyltransferase activity in ARID1A-mutated cancers
.
Nat Med
2015
;
21
:
231
8
.
18.
Guan
B
,
Wang
TL
,
Shih Ie
M
. 
ARID1A, a factor that promotes formation of SWI/SNF-mediated chromatin remodeling, is a tumor suppressor in gynecologic cancers
.
Cancer Res
2011
;
71
:
6718
27
.
19.
Bitler
BG
,
Wu
S
,
Park
PH
,
Hai
Y
,
Aird
KM
,
Wang
Y
, et al
ARID1A-mutated ovarian cancers depend on HDAC6 activity
.
Nat Cell Biol
2017
;
19
:
962
73
.
20.
Anglesio
MS
,
Papadopoulos
N
,
Ayhan
A
,
Nazeran
TM
,
Noe
M
,
Horlings
HM
, et al
Cancer-associated mutations in endometriosis without cancer
.
N Engl J Med
2017
;
376
:
1835
48
.
21.
Jelinic
P
,
Mueller
JJ
,
Olvera
N
,
Dao
F
,
Scott
SN
,
Shah
R
, et al
Recurrent SMARCA4 mutations in small cell carcinoma of the ovary
.
Nat Genet
2014
;
46
:
424
6
.
22.
Witkowski
L
,
Carrot-Zhang
J
,
Albrecht
S
,
Fahiminiya
S
,
Hamel
N
,
Tomiak
E
, et al
Germline and somatic SMARCA4 mutations characterize small cell carcinoma of the ovary, hypercalcemic type
.
Nat Genet
2014
;
46
:
438
43
.
23.
Ramos
P
,
Karnezis
AN
,
Craig
DW
,
Sekulic
A
,
Russell
ML
,
Hendricks
WP
, et al
Small cell carcinoma of the ovary, hypercalcemic type, displays frequent inactivating germline and somatic mutations in SMARCA4
.
Nat Genet
2014
;
46
:
427
9
.
24.
Wilson
BG
,
Helming
KC
,
Wang
X
,
Kim
Y
,
Vazquez
F
,
Jagani
Z
, et al
Residual complexes containing SMARCA2 (BRM) underlie the oncogenic drive of SMARCA4 (BRG1) mutation
.
Mol Cell Biol
2014
;
34
:
1136
44
.
25.
Karnezis
AN
,
Wang
Y
,
Ramos
P
,
Hendricks
WP
,
Oliva
E
,
D'Angelo
E
, et al
Dual loss of the SWI/SNF complex ATPases SMARCA4/BRG1 and SMARCA2/BRM is highly sensitive and specific for small cell carcinoma of the ovary, hypercalcaemic type
.
J Pathol
2016
;
238
:
389
400
.
26.
Jelinic
P
,
Ricca
J
,
Van Oudenhove
E
,
Olvera
N
,
Merghoub
T
,
Levine
DA
, et al
Immune-active microenvironment in small cell carcinoma of the ovary, hypercalcemic type: rationale for immune checkpoint blockade
.
J Natl Cancer Inst
2018
;
110
:
787
90
.
27.
Karakashev
S
,
Zhu
H
,
Wu
S
,
Yokoyama
Y
,
Bitler
BG
,
Park
PH
, et al
CARM1-expressing ovarian cancer depends on the histone methyltransferase EZH2 activity
.
Nat Commun
2018
;
9
:
631
.
28.
Wang
L
,
Zhao
Z
,
Meyer
MB
,
Saha
S
,
Yu
M
,
Guo
A
, et al
CARM1 methylates chromatin remodeling factor BAF155 to enhance tumor progression and metastasis
.
Cancer Cell
2014
;
25
:
21
36
.
29.
Kennison
JA
,
Tamkun
JW
. 
Dosage-dependent modifiers of polycomb and antennapedia mutations in Drosophila
.
Proc Natl Acad Sci U S A
1988
;
85
:
8136
40
.
30.
Kadoch
C
,
Williams
RT
,
Calarco
JP
,
Miller
EL
,
Weber
CM
,
Braun
SM
, et al
Dynamics of BAF-Polycomb complex opposition on heterochromatin in normal and oncogenic states
.
Nat Genet
2017
;
49
:
213
22
.
31.
Stanton
BZ
,
Hodges
C
,
Calarco
JP
,
Braun
SM
,
Ku
WL
,
Kadoch
C
, et al
Smarca4 ATPase mutations disrupt direct eviction of PRC1 from chromatin
.
Nat Genet
2017
;
49
:
282
8
.
32.
McCabe
MT
,
Ott
HM
,
Ganji
G
,
Korenchuk
S
,
Thompson
C
,
Van Aller
GS
, et al
EZH2 inhibition as a therapeutic strategy for lymphoma with EZH2-activating mutations
.
Nature
2012
;
492
:
108
12
.
33.
Knutson
SK
,
Wigle
TJ
,
Warholic
NM
,
Sneeringer
CJ
,
Allain
CJ
,
Klaus
CR
, et al
A selective inhibitor of EZH2 blocks H3K27 methylation and kills mutant lymphoma cells
.
Nat Chem Biol
2012
;
8
:
890
6
.
34.
Qi
W
,
Chan
H
,
Teng
L
,
Li
L
,
Chuai
S
,
Zhang
R
, et al
Selective inhibition of Ezh2 by a small molecule inhibitor blocks tumor cells proliferation
.
Proc Natl Acad Sci U S A
2012
;
109
:
21360
5
.
35.
Kim
KH
,
Kim
W
,
Howard
TP
,
Vazquez
F
,
Tsherniak
A
,
Wu
JN
, et al
SWI/SNF-mutant cancers depend on catalytic and non-catalytic activity of EZH2
.
Nat Med
2015
;
21
:
1491
6
.
36.
Januario
T
,
Ye
X
,
Bainer
R
,
Alicke
B
,
Smith
T
,
Haley
B
, et al
PRC2-mediated repression of SMARCA2 predicts EZH2 inhibitor activity in SWI/SNF mutant tumors
.
Proc Natl Acad Sci U S A
2017
;
114
:
12249
54
.
37.
Wang
Y
,
Chen
SY
,
Karnezis
AN
,
Colborne
S
,
Santos
ND
,
Lang
JD
, et al
The histone methyltransferase EZH2 is a therapeutic target in small cell carcinoma of the ovary, hypercalcaemic type
.
J Pathol
2017
;
242
:
371
83
.
38.
Berns
K
,
Caumanns
JJ
,
Hijmans
EM
,
Gennissen
AMC
,
Severson
TM
,
Evers
B
, et al
ARID1A mutation sensitizes most ovarian clear cell carcinomas to BET inhibitors
.
Oncogene
2018
;
37
:
4611
25
.
39.
Miao
D
,
Margolis
CA
,
Gao
W
,
Voss
MH
,
Li
W
,
Martini
DJ
, et al
Genomic correlates of response to immune checkpoint therapies in clear cell renal cell carcinoma
.
Science
2018
;
359
:
801
6
.
40.
Pan
D
,
Kobayashi
A
,
Jiang
P
,
Ferrari de Andrade
L
,
Tay
RE
,
Luoma
AM
, et al
A major chromatin regulator determines resistance of tumor cells to T cell-mediated killing
.
Science
2018
;
359
:
770
5
.
41.
Zingg
D
,
Arenas-Ramirez
N
,
Sahin
D
,
Rosalia
RA
,
Antunes
AT
,
Haeusel
J
, et al
The histone methyltransferase Ezh2 controls mechanisms of adaptive resistance to tumor immunotherapy
.
Cell Rep
2017
;
20
:
854
67
.
42.
Peng
D
,
Kryczek
I
,
Nagarsheth
N
,
Zhao
L
,
Wei
S
,
Wang
W
, et al
Epigenetic silencing of TH1-type chemokines shapes tumour immunity and immunotherapy
.
Nature
2015
;
527
:
249
53
.
43.
Shen
J
,
Ju
Z
,
Zhao
W
,
Wang
L
,
Peng
Y
,
Ge
Z
, et al
ARID1A deficiency promotes mutability and potentiates therapeutic antitumor immunity unleashed by immune checkpoint blockade
.
Nat Med
2018
;
24
:
556
62
.
44.
Williamson
CT
,
Miller
R
,
Pemberton
HN
,
Jones
SE
,
Campbell
J
,
Konde
A
, et al
ATR inhibitors as a synthetic lethal therapy for tumours deficient in ARID1A
.
Nat Commun
2016
;
7
:
13837
.
45.
Shen
J
,
Peng
Y
,
Wei
L
,
Zhang
W
,
Yang
L
,
Lan
L
, et al
ARID1A deficiency impairs the DNA damage checkpoint and sensitizes cells to PARP inhibitors
.
Cancer Discov
2015
;
5
:
752
67
.
46.
Kim
M
,
Lu
F
,
Zhang
Y
. 
Loss of HDAC-mediated repression and gain of NF-kappaB activation underlie cytokine induction in ARID1A- and PIK3CA-mutation-driven ovarian cancer
.
Cell Rep
2016
;
17
:
275
88
.
47.
Miller
RE
,
Brough
R
,
Bajrami
I
,
Williamson
CT
,
McDade
S
,
Campbell
J
, et al
Synthetic lethal targeting of ARID1A-mutant ovarian clear cell tumors with dasatinib
.
Mol Cancer Ther
2016
;
15
:
1472
84
.
48.
Lang
JD
,
Hendricks
WPD
,
Orlando
KA
,
Yin
H
,
Kiefer
J
,
Ramos
P
, et al
Ponatinib shows potent antitumor activity in small cell carcinoma of the ovary hypercalcemic type (SCCOHT) through multikinase inhibition
.
Clin Cancer Res
2018
;
24
:
1932
43
.
49.
Adeegbe
DO
,
Liu
Y
,
Lizotte
PH
,
Kamihara
Y
,
Aref
AR
,
Almonte
C
, et al
Synergistic immunostimulatory effects and therapeutic benefit of combined histone deacetylase and bromodomain inhibition in non-small cell lung cancer
.
Cancer Discov
2017
;
7
:
852
67
.
50.
Zhu
H
,
Bengsch
F
,
Svoronos
N
,
Rutkowski
MR
,
Bitler
BG
,
Allegrezza
MJ
, et al
BET bromodomain inhibition promotes anti-tumor immunity by suppressing PD-L1 expression
.
Cell Rep
2016
;
16
:
2829
37
.