Introducton

Transcriptional enhancers are short regulatory DNA elements which bind RNA polymerase II (RNA Pol II), transcription factors and co-regulators, and are characterized by acetylated histone H3 lysine 27 (H3K27ac) and monomethylated H3K4 (H3K4me) signals in chromatin immunoprecipitation sequencing assays [1]. As enhancers can form loops with promoters over a long distance, enhancers augment the transcription of neighboring genes, irrespective of the sense or antisense direction of their target genes [2, 3].

Super-enhancers are large clusters of enhancers that are in close genomic proximity, are densely bound by the BET bromodomain protein BRD4 and master transcription factors, and are characterized by massive H3K27ac and H3K4me signals in ChIP sequencing [4,5,6].

Enhancers activate gene transcription and induce tumorigenesis

Transcriptional enhancers recruit BRD4, transcription factors and cofactors to activate RNA Pol II and gene transcription from gene promoters [7] (Fig. 1A). Transcriptional enhancers have been confirmed to play an important role in the activation and over-expression of oncogenes, such as MYC which is juxtaposed to the immunoglobulin heavy-chain gene enhancer in Burkitt’s lymphoma [8].

Fig. 1: Transcriptional enhancers and super-enhancers activate gene transcription.
figure 1

A, B The BET bromodomain protein BRD4 recognizes acetylated (Ac) histone H3 lysine 27, binds to and activates enhancers (A) and super-enhancers (B). BRD4 recruits the positive transcription elongation factor b (P-TEFb) and Mediator, leading to RNA Polymerase II (RNA Pol II) activation and binding to enhancer- and super-enhancer-associated gene promoter, transcriptional activation and target gene over-expression. As super-enhancers are bound by much larger clusters of BRD4 proteins, super-enhancer-associated oncogenes are transcribed at substantially higher levels than enhancer-associated genes.

The Hippo pathway transcription coactivators YAP/TAZ form a protein complex with TEAD and AP-1 at distal transcriptional enhancers rather than promoters, located >100,000 base pairs away from transcription start sties. Through chromatin looping, the YAP/TAZ/TEAD/AP-1 transcription cofactor and transcription factor complex activate the transcription of enhancer-associated genes including those controlling S-phase entry and mitosis of the cell cycle, resulting in cell proliferation and skin tumorigenesis [9]. The oncogenic transcription factor FOXA1 is hyperactive in metastatic endocrine-resistant breast cancer cells due to gene amplification or overexpression. FOXA1 induces enhancer reprogramming and transcriptional activation of pro-metastatic oncogenes in endocrine-resistant breast cancer cells [10].

The transcriptional activator NRF2 is frequently activated in non-small cell lung cancer, and NRF2 overexpression results in the accumulation of CCAAT Enhancer Binding Protein Beta (CEBPB) [11]. NRF2 and CEBPB co-operatively induce the establishment of transcriptional enhancers at the loci of oncogenes such as the NOTCH3 gene [11]. Importantly, in mouse models of non-small cell lung cancer, disruption of the NOTCH3 enhancer significantly suppresses tumor progression and augments the anticancer effects of cisplatin, demonstrating the important role of the NOTCH3 enhancer in tumorigenesis and drug resistance [11].

Recent transcriptome profiling has shown that squamous cell lineage markers are present in ~25% of pancreatic ductal adenocarcinoma tumors, and the squamous cell subtype is associated with poorer patient prognosis [12]. Aberrant enhancers have recently been found to be established in the squamous cell subtype of pancreatic ductal adenocarcinoma tumors. Enhancers at the loci of oncogenes, such as MYC and HRAS, play a critical role in pancreatic ductal adenocarcinoma cell transition into squamous cells, cell migration and invasion in vitro, and accelerated tumor growth and metastases in vivo [13].

Enhancers can also activate tumor suppressor gene transcription and suppress tumorigenesis

Enhancers can also activate tumor suppressor gene transcription and thereby suppress tumorigenesis. The N-terminal SNAG domain of the transcriptional repressor GFI1 binds to the CoREST transcriptional complex proteins LSD1 and RCOR1 at the enhancers of transcription factor genes, such as SPI1 (PU.1), CEBPA and IRF8 which are important for acute myeloid leukemia cell differentiation [14]. GFI1 inactivation or LSD1 inhibition with small molecule compound inhibitors disrupts the interaction between GFI1, LSD1 and RCOR1, leading to considerable increase in H3K27ac at enhancer regions of the transcription factor genes, transcriptional activation, acute myeloid leukemia cell differentiation, growth inhibition and clonogenicity reduction [14].

SWI/SNF (mSWI/SNF or BAF) chromatin remodeling complex inactivation contributes to >20% of human cancers. Forced over-expression of the core BAF complex subunit SMARCB1 in sarcoma cells results in the activation of distal typical enhancers and super-enhancers at the loci of genes such as CDKN1A [15]. The activated typical enhancers and super-enhancers play critical roles in sarcoma cell growth arrest, demonstrating a tumor suppression effect [15]. Therefore, enhancers can induce or suppress tumorigenesis, probably depending on cancer subtypes and cellular contexts.

Super-enhancers activate oncogene transcription and induce tumorigenesis

Super-enhancers consist of enhancer clusters, span large genetic regions, and are generally an order of magnitude larger than typical enhancers [4, 5]. Super-enhancers are bound by a large number of BRD4 which recruits the Mediator, a protein complex connecting the transcription factors at the super-enhancers and RNA pol II at the gene promoters (Fig. 1B) [4, 5].

Super-enhancers are emerging as critical regulators of oncogene transcription and tumorigenesis. In glioblastoma cells, super-enhancers have been found to be associated with a number of oncogenic genes, such as RUNX1, BCL3 and FOSL2 [6]. In glioblastoma stem cells isolated from PDX mouse models originally derived from human tumor samples, a subset of super-enhancers at the loci of critical genes, such as CDK6, SOX2, EGFR and BRD4, are shared by the majority of human glioblastoma stem cells [16]. Proximity of the super-enhancers to their associated genes correlates with gene over-expression in glioblastoma stem cells and human tumor samples, and the core glioblastoma stem cell super-enhancer-associated genes are essential for glioblastoma cell proliferation and tumorigenesis [16] (Table 1). In addition, patients with glioblastoma that is enriched of the core glioblastoma stem cell super-enhancer signature show more advanced tumor stage and poorer prognosis [16].

Table 1 Super-enhancers and BRD4 induce oncogene transcriptional activation and over-expression, cancer cell proliferation, survival, tumor initiation and progression.

Super-enhancers are extensively reprogrammed during liver cancer tumorigenesis [17]. Liver cancer cells acquire super-enhancers at the loci of critical oncogenic genes, such as SPHK1, MYC, MYCN, SHH and YAP1, to drive their substantial over-expression. The super-enhancer “writer” p300, super-enhancer “reader” BRD4, and super-enhancer activity regulators CDK7 and MED1 are often over-expressed in human liver cancer tissues, and their over-expression predicts poor patient prognosis [17]. Importantly, inhibition of p300, BRD4, CDK7 or MED1 reduces the expression of super-enhancer-associated oncogenes and exerts anticancer effects against liver cancer [17].

The histone demethylase KDM6A gene is often mutated in a variety of human malignancies. Loss of function of KDM6A causes squamous-like metastatic pancreatic cancer through aberrant activation of super-enhancers at the loci of MYC and RUNX3 oncogenes and consequent MYC and RUNX3 over-expression [18]. Treatment with BRD4 inhibitors results in KDM6A mutant pancreatic cancer cell differentiation and tumor growth inhibition in a mouse model [18] (Table 1).

The super-enhancer landscape of small cell lung cancer cells recapitulates embryonic, neural and tumorigenic signatures, as many super-enhancers are associated with lineage-specific transcription factor genes and oncogenes such as MYC, SOX2 and NFIB [19] (Table 1). In a high-throughput compound screening, small cell lung cancer cells have been found to be very sensitive to the CDK7 inhibitor THZ1 which selectively suppresses the expression of super-enhancer associated genes [19].

In chromosome 17q-gained neuroblastoma, the JMJD6 gene is over-expressed due to both gene gain and transcriptional super-enhancers, and suppression of super-enhancer activity reduces JMJD6 gene expression, neuroblastoma cell proliferation in vitro and tumor growth in a mouse model [20] (Table 1). Similarly, in diffuse intrinsic pontine glioma, the expression of critical oncogenic genes such as SOX2 and NOTCH1 is regulated by super-enhancers, and treatment with super-enhancer inhibitors reduces diffuse intrinsic pontine glioma cell proliferation in vitro and tumor progression in mouse models [21] (Table 1).

In clear cell renal cell carcinoma, super-enhancers are formed at the loci of CXC chemokine genes, such as CXCL1, CXCL5 and CXCL8, and induce CXC chemokine gene over-expression and renal cell carcinoma progression and metastasis [22]. Consistent with these findings, suppression of super-enhancer activity reduces CXC chemokine gene expression and renal cell cancer metastasis [22] (Table 1).

Compared with normal counterparts, colon cancer cells gain oncogenic super-enhancers, including super-enhancers associated with ASCL2, a transcription factor for intestinal stem cell fate, and the Wnt target gene MYC [23]. In addition, β-catenin and CTCF up-regulate MYC by connecting nucleoporins to oncogenic super-enhancers, leading to MYC mRNA export to the cytoplasm, stabilization and over-expression [24, 25]. Interestingly, inflammation in the tumor microenvironment results in the formation of super-enhancers at the PDZK1IP1 gene locus, resulting in colon cancer cell proliferation in vitro and tumor progression in a mouse model [23] (Table 1).

Medulloblastoma are divided into 4 distinct groups, WNT, SHH, Group 3, and Group 4 groups, and the 4 different groups show distinct super-enhancer profiles. Association of critical oncogenes with super-enhancers has been found at the ALK gene locus in WNT group, at SMO and NTRK3 gene loci in SHH group, at the LMO1, LMO2 and MYC gene loci in Group 3, and at the ETV4 and PAX5 gene loci in Group 4 [26] (Table 1).

In leukemic stem cells, the MYC gene locus is characterized by super-enhancers which recruit critical transcriptional factors including MYB, RUNX1 and GFI1b to drive MYC over-expression and leukemogenesis [27]. In chronic myelogenous leukemia stem cells, suppression of super-enhancer-driven gene transcription by a CDK7 inhibitor eradicates leukemia stem cells in a mouse model without effects in normal hematopoietic stem cells [28]. In human primary T cell acute lymphoblastic leukemia samples, a topologically associating domain ‘fusion’ event due to CTCF-mediated insulation absence results in the interaction between distal super-enhancers and the MYC gene promoter, leading to MYC over-expression and leukemogenesis [29] (Table 1).

Super-enhancers have also been shown to be important in epithelial-to-mesenchymal transition (EMT) and metastasis. ETS2, JUNB, EGFR and HNF4A genes are associated with super-enhancers in non-small cell lung cancer cells. Suppression of super-enhancer activity reduces the expression of these super-enhancer-associated genes, decreases non-small cell lung cancer cell migration and invasion, and abrogates TGF-β-induced EMT, demonstrating the role of super-enhancers in regulating EMT and tumor metastasis [30].

Super-enhancers can function as tumor suppressors

While generally proven to promote tumor initiation and progression, super-enhancers can also function as tumor suppressors. The histone methyltransferase KMT2D is often inactivated in human lung cancer tissues. Loss of KMT2D reduces the activity of super-enhancers at critical genes, such as the circadian rhythm repressor Per2, resulting in Per2 gene down-regulation, glycolysis and lung cancer tumorigenesis [31]. In breast cancer, loss of the tumor suppressor gene RCAN1.4 augments tumor metastasis. Unexpectedly, RCAN1.4 gene expression is driven by super-enhancers in breast cancer cells, and suppression of super-enhancer activity with BRD4 knockdown or BRD4 inhibitor treatment reduces RCAN1.4 tumor suppressor gene expression [32].

The super-enhancer “reader” BRD4 forms a protein complex with the repressive LSD1/NuRD transcription regulators at super-enhancers to suppress the expression of drug resistance genes in breast cancer cells [33]. Repression of super-enhancer activity with BRD4 inhibitors does not have an immediate effect on the expression of the drug resistance genes, however, long-time treatment with BRD4 inhibitors causes resistance to both BRD4 inhibitors and a broad spectrum of anticancer agents, demonstrating the role of super-enhancers and BRD4 in super-enhancer-mediated transcriptional repression of genes involved in tumorigenesis and chemoresistance [33]. Therefore, long-term treatment with BRD4 inhibitors might promote multidrug resistance and tumor progression, and close monitoring and prompt intervention are required in clinical trials.

The super-enhancer “reader” Brd4 promotes super-enhancer-associated oncogene transcription and tumorigenesis and Brd4 inhibitors exert anticancer effects

The BET bromodomain protein BRD4 recognizes, binds to and activates super-enhancers and substantially up-regulate the expression of super-enhancer-associated oncogenes (Fig. 1B), and BRD4 inhibitors blocks BRD4 binding and reduce oncogene expression (Fig. 2). In diffuse large B cell lymphoma, approximately one-third of BRD4 protein localizes to super-enhancers which occupy ~1.6% of genes [34]. Treatment with four different BRD4 inhibitors reduces the expression of super-enhancer-associated oncogenes, such as MYC, E2F1, BCL6 and PAX5, and reduces diffuse large B cell lymphoma cell proliferation. Treatment of mice xenografted with diffuse large B cell lymphoma with the BRD4 inhibitor JQ1 suppresses lymphoma progression [34] (Table 1).

Fig. 2: BRD4 inhibitors suppress oncogene transcription and expression.
figure 2

The BET bromodomain protein BRD4 recognizes acetylated (Ac) histone H3 lysine 27, binds to and activates super-enhancers, leading to RNA Polymerase II (RNA Pol II) binding to super-enhancer-associated oncogene promoter, gene transcriptional activation and over-expression. Treatment with BRD4 inhibitors displaces BRD4 at super-enhancers, leading to RNA Pol II disassociation from gene promoters and transcriptional suppression.

In estrogen receptor alpha (ERα)-positive breast cancer cells, BRD4 is a master activator of ERα-occupied super-enhancers and the transcription of ERα target genes, such as RET which in turn activates ERα phosphorylation and ERα target gene expression. BRD4 therefore induces breast cancer cell proliferation and tumor progression [35] (Table 1).

In human neck squamous cell carcinoma, BRD4 recruits Mediators and NF-κB at super-enhancers associated with cancer stemness genes such as MET, TP63 and FOSL1. Treatment with BRD4 inhibitors reduces stemness gene expression; suppresses cancer stem cell self-renewal, invasive growth and metastasis; and eliminates tumor cells and cancer stem cells in a mouse model of neck squamous cell carcinoma [36]. In nasopharyngeal carcinoma cells, super-enhancers are enriched of BRD4, NF-κB, IRF1 and IRF2 transcription factors at the loci of critical oncogenes such as ETV6, high expression of which in human nasopharyngeal carcinoma tissues is correlated with poor patient prognosis [37]. Treatment with the BRD4 inhibitor JQ1 significantly suppresses super-enhancer-associated ETV6 gene expression and induces nasopharyngeal carcinoma cell growth inhibition [37] (Table 1).

In diffuse intrinsic pontine glioma, super-enhancers are found at the loci of a number of genes indicating undifferentiation status such as SOX2 and NES as well as oncogenes such EGFR [21]. These super-enhancers are characterized by BRD4 binding, and BRD4 knockdown or inhibition reduces diffuse intrinsic pontine glioma cell proliferation in vitro and tumor progression in mouse models [21] (Table 1).

In rhabdomyosarcoma, super-enhancers are bound by core regulatory transcription factors and are characterized by the highest levels of histone acetylation [38]. Counterintuitively, the super-enhancers are also bound by the most histone deacetylases (HDACs), and HDAC inhibitors augment BRD4, but decreases RNA Pol II and core regulatory transcription factor, binding to the super-enhancers. The data demonstrate super-enhancer-specific requirement to balance histone acetylation and deacetylation for maintaining super-enhancer architecture and gene transcription [38]. In alveolar rhabdomyosarcoma, the chimeric transcription factor PAX3-FOXO1 interacts with the master transcription factors MYCN, MYOG and BRD4 at target gene super-enhancers, resulting in over-expression of SOX8, MYOD1, MYOG and MYCN, alveolar rhabdomyosarcoma tumorigenesis and dependence on BRD4 [39]. Inhibition of BRD4 with the BRD4 inhibitor JQ1 or OTX015 abolishes PAX3-FOXO1 function, suppresses alveolar rhabdomyosarcoma cell proliferation in vitro and induces tumor growth inhibition in mouse models [39] (Table 1).

Melanoma with PGC-1α over-expression is characterized by substantial BRD4 protein binding at the PGC- gene super-enhancer [40]. Treatment with the BRD4 inhibitor JQ1 or BAY 1238097 blocks BRD4 binding to the super-enhancer and PGC-1α expression, suppresses melanoma cell proliferation in vitro, and inhibits tumor growth in a mouse model [40] (Table 1).

In multiple myeloma, BRD4 and Mediator are enriched at super-enhancers associated with oncogenes including MYC, BCL-xL and IRF4. Treatment of multiple myeloma cells with the BRD4 inhibitor JQ1 results in BRD4 disassociation from super-enhancers, and reduction in MYC, BCL-xL and IRF4 gene expression and multiple myeloma cell proliferation [6]. In t(4;14)-positive multiple myeloma, BRD4 interacts with the histone lysine methyltransferase NSD2 at the HJURP gene super-enhancers, leading to HJURP gene over-expression, multiple myeloma cell proliferation and survival [41] (Table 1). Taken together, BRD4 promotes super-enhancer-associated oncogene transcription and tumorigenesis, and BRD4 inhibitors exert anticancer effects.

Small molecule compound Brd4 inhibitors and degraders exert promising anticancer effects in pre-clinical models

Small molecule compound BRD4 BD1 and BD2 bromodomain inhibitors in cancer therapy

In the past decade, a number of small molecule compound BRD4 inhibitors have been developed through chemical synthesis, structure-based in silico screen, and wet lab screen of small molecule compound libraries. The majority of the inhibitors, such as JQ1, OTX015, I-BET762, MK-8628, NHWD870, ABBV-744, PLX2853 and INCB054329, target both the BD1 and BD2 bromodomains of BRD4, reduce oncogene expression, and exert anticancer effects in pre-clinical models.

Pancreatic ductal adenocarcinoma, head and neck squamous cell carcinoma and leukemia are characterized by oncogene over-expression due to super-enhancers. Combination therapy with the BRD4 inhibitor JQ1 and the CDK7 inhibitor THZ1 synergistically reduces super-enhancer-associated oncogene expression and exerts synergistical anticancer effects against pancreatic ductal adenocarcinoma and head and neck squamous cell carcinoma in vitro and in mouse models [42, 43]. Interestingly, combination therapy with BRD4 inhibitors and CDK7 inhibitors overcomes resistance to BRD4 inhibitor therapy in leukemia cells and mouse models of leukemia [44]; and nanoparticle-mediated delivery of JQ1 and THZ1, compared with free drug formulation, considerably reduces cytotoxicity to liver cells but synergistically suppresses tumor progression in a mouse model of drug-resistant pancreatic ductal adenocarcinoma [42].

CDK4/CDK6 inhibitors have also been shown to exert synergistic anticancer effects with BRD4 inhibitors in castration-resistant prostate cancer and NUT midline carcinoma [45, 46]. Castration-resistant prostate cancer cells with high levels of the deubiquitinase DUB3 and NUT midline carcinoma cells with high levels of KLF4 are resistant to BRD4 inhibitors, because DUB3 binds to BRD4 and augments its deubiquitination and stabilization and KLF4 up-regulates E2F and MYC gene expression [45, 46]. As DUB3 is activated after phosphorylation by CDK4 and CDK6 and E2F and MYC expression are activated after Rb phosphorylation by CDK4 and CDK6, treatment with the CDK4/CDK6 inhibitor Palbociclib sensitizes prostate cancer and NUT midline carcinoma cells to the BRD4 inhibitor JQ1, and exerts synergistic anticancer effects with JQ1 in vitro and in mouse models of castration-resistant prostate cancer and NUT midline carcinoma [45, 46].

In a high-throughput drug screen, BRD4 inhibitors have been found to be one of the two classes of compounds exerting the best synergistic anticancer effects with the CDK4/CDK6 inhibitor Ribociclib in medulloblastoma cells [47]. A reverse combination drug screen identifies CDK4/CDK6 inhibitors as the compounds exerting the best synergy with the BRD4 inhibitor JQ1 against medulloblastoma cells [47]. Treatment with the orally bioavailable BRD4 inhibitor MK-8628 suppresses medulloblastoma cell proliferation and induces apoptosis by reducing MYC expression, and MK-8628 suppresses medulloblastoma tumor progression in preclinical models [48]. Co-treatment with MK-8628 and the PLK1 inhibitor Volasertib, which targets MYC protein for degradation, shows synergistic anti-medulloblastoma effects in vitro and in preclinical models [48].

Another well-studied anticancer agent for BRD4 inhibitor combination therapy is HDAC inhibitors, particularly the pan-HDAC inhibitor Panobinostat. Combination therapy with the BRD4 inhibitor JQ1 or OTX015 and Panobinostat synergistically reduces the expression of oncogenes, such as MYC, MYCN and LIN28B; suppresses proliferation and induces apoptosis in MYCN gene-amplified neuroblastoma, medulloblastoma and diffuse intrinsic pontine glioma cells; and significantly suppresses neuroblastoma and medulloblastoma tumor progression in mouse models [21, 49, 50]. In neuroblastoma due to TERT gene rearrangement with super-enhancers, BRD4 is required for TERT gene transcription and neuroblastoma cell proliferation [51]. In an unbiased screen of approved oncology drugs, the BRD4 inhibitors I-BET762 and OTX015 exert the best synergistic anticancer effects with the proteasome inhibitor Carfilzomib; and OTX015 and carfilzomib synergistically reduce TERT expression, induces TERT gene-rearranged neuroblastoma cell apoptosis, blocks tumor progression and improves survival in multiple mouse models of TERT gene-rearranged neuroblastoma [51].

Unbiased high-throughput drug combination screens reveal that PI3K-AKT-mTOR pathway inhibitors exert synergistic anticancer effects with BRD4 inhibitors against small cell lung cancer cells, and mTOR inhibitors exhibit the best synergy [52]. Mechanistically, while BRD4 inhibitors up-regulate RSK3 to activate the mTOR pathway, mTOR inhibitors block this cell survival signaling and enhance BRD4 inhibitor-mediated cancer cell apoptosis [52]. In multiple patient-derived xenograft models of small cell lung cancer, combination therapy with the mTOR inhibitor Everolimus and the BRD4 inhibitor NHWD870 synergistically induce cancer cell apoptosis and blocks tumor progression without significantly increasing toxicity to normal tissues in mice [52]. In Ewing sarcoma cell lines and patient-derived xenograft (PDX) lines, AKT pathway activation protects Ewing sarcoma cells against BRD4 inhibitors, and IGF1R inhibitors and mTOR inhibitors suppress AKT pathway activation and synergistically enhance cancer cell sensitivity to BRD4 inhibitors [53]. In PDX models of Ewing sarcoma, treatment with the BRD4 inhibitor NHWD870 and the IGF1R inhibitor BMS754807 results in substantial and durable anticancer effects, while monotherapy was much less effective [53].

Genome-wide loss-of-function clustered regularly interspaced short palindromic repeats (CRISPR) screens identify SPOP gene deficiency as a resistance factor to BRD4 inhibitor therapy in KMT2A gene-rearranged leukemia cells [54]. Kinase vulnerability CRISPR screens identify GSK3 inhibitors as effective agents to overcome SPOP deficiency-induced BRD4 inhibitor resistance. Combination therapy with the BRD4 inhibitor ABBV-744 and the GSK3 inhibitor CHIR-98014 considerably suppresses KMT2A-rearranged leukemia progression in patient-derived xenograft models in mice, confirming ABBV-744 and CHIR-98014 combination therapy as an effective therapeutic strategy [54]. Since it is now clear that targeted therapies need to be combined with other anticancer agents in the clinic to exert better anticancer effects and to reduce toxicity, the other anticancer agents should be identified by unbiased screening of anticancer drug libraries for each cancer subtype.

Small molecule compound BRD4 BD2 selective bromodomain inhibitors in cancer therapy

While the majority of BRD4 inhibitors bind to the BD1 and BD2 bromodomains of BRD4 with similar affinities, the small molecule compound ABBV-744 selectively binds to the BD2 bromodomain [55]. By selectively suppressing the BD2 bromodomain, ABBV-744 induces acute myeloid leukemia and prostate cancer cell growth inhibition, and exhibits significant anticancer effects against acute myeloid leukemia and prostate cancer in mouse models with better toxicity profile and therapeutic index than BRD4 BD1 and BD2 bromodomain inhibitors [55, 56]. In addition, while the BRD4 inhibitors PLX2853 and INCB054329 show synergistic anticancer effects when combined with the BCL2 inhibitor Venetoclax in mouse models of diffuse large B-cell lymphoma and acute myeloid leukemia, ABBV-744 also exerts synergistic anticancer effects with Venetoclax in mouse models of acute myeloid leukemia (Table 2) [56, 57]. Interestingly, GSK620, another small molecule compound BRD4 BD2 bromodomain selective inhibitor, suppresses inflammatory disease in pre-clinical models (Table 2) [58].

Table 2 BRD4 BD2 bromodomain selective inhibitors, PROTAC BRD4 protein degraders, and dual BRD4 and CBP/p300 bromodomain co-inhibitors.

Small molecule compound proteolysis-targeting chimera (PROTAC) BRD4 protein degraders in cancer therapy

PROTAC protein degraders are emerging as novel anticancer agents. ARV-771, a small molecule compound PROTAC BRD4 protein degrader, down-regulates the expression of oncogenes such as MYC [59]. ARV-771 reduces cell proliferation and induces apoptosis substantially more effectively than the BRD4 inhibitor JQ1 and OTX015 in castration-resistant prostate cancer and diffuse large B cell lymphoma cells [59, 60]. Importantly, while OTX015 suppresses castration-resistant prostate cancer progression, treatment with ARV-771 results in tumor regression in mice xenografted with castration-resistant prostate cancer cell tumors [60] and growth inhibition in mice xenografted with diffuse large B cell lymphoma cells [59] (Table 2).

A1874 is a nutlin-based small molecule compound PROTAC BRD4 protein degrader. A1874 combines the activities of the BRD4 inhibitor JQ1 and the MDM2 antagonist idasanutlin, degrades BRD4 protein by 98% at nanomolar concentrations and stabilizes p53 protein [61]. Treatment with A1874 more significantly reduces cell proliferation and induces cell death in a variety of cancer cell lines with wild type p53 than PROTAC BRD4 protein degraders [61] (Table 2). PROTAC BRD4 protein degraders are therefore likely to be more effective anticancer agents than BRD4 bromodomain inhibitors.

Small molecule compound dual BRD4 and CBP/p300 bromodomain co-inhibitors in cancer therapy

Another effective approach is to target the bromodomains of the super-enhancer “reader” BRD4 and the “writers” CBP/p300 simultaneously. The dual BRD4 and CBP/p300 bromodomain co-inhibitor XP-524 exhibits higher potency and superior tumoricidal activity than the BRD4 inhibitor JQ-1, and shows anticancer efficacy comparable to combination therapy with high-dose JQ-1 and the CBP/p300 inhibitor SGC-CBP30 in pancreatic ductal adenocarcinoma cells [62]. XP-524 suppresses KRAS activity, blocks KRAS-induced malignant transformation in vivo and improves mouse survival in transgenic mouse models of aggressive pancreatic ductal adenocarcinoma. In addition, XP-524 and an anti-PD-1 antibody exert synergistic anticancer effects and improve survival in two transgenic mouse models of pancreatic ductal adenocarcinoma cells [62] (Table 2).

The other dual BRD4 and CBP/p300 bromodomain co-inhibitor NEO2734 up-regulates the expression of p53 and its target PUMA and induces colorectal cancer cell apoptosis through the intrinsic and extrinsic apoptosis pathways, suppression of the intrinsic or extrinsic apoptosis pathway partly rescues colorectal cancer cells, and NEO2734 represses colon cancer progression by inducing colorectal cancer cell apoptosis in a mouse model [63] (Table 2). In addition, NEO2734 shows more potent anticancer effects than single-agent BRD4 or CBP/p300 inhibitors in lymphoma and acute myeloid leukemia cell lines, and exerts substantial anticancer effects in mouse models of lymphoma and acute myeloid leukemia [64] (Table 2). Dual BRD4 and CBP/p300 bromodomain co-inhibitors are therefore likely to be more effective anticancer agents than BRD4 bromodomain inhibitors.

BRD4 inhibitors show anticancer effects in clinical trials

More than a dozen BRD4 BD1 and BD2 bromodomain inhibitors, including ABBV-075, AZD5153, BAY 1238097, BMS-986158, BMS-986378, CC-90010, CPI-0610, FT-1101, GSK525762 (Molibresib), INCB054329, INCB057643, ODM-207, OTX015 and PLX51107 have been or are currently in clinical trials in patients with cancer from various organ origins. The BRD4 inhibitors show anticancer effects in clinical trials as monotherapy, but it is now clear that BRD4 inhibitors need to be combined with other anticancer agents to effectively treat cancer patients (Table 3).

Table 3 BRD4 BD1 and BD2 bromodomain inhibitors in clinical trials.

In a dose-escalation, phase I clinical study in acute myeloid leukemia, lymphoma and myeloma patients, plasma OTX015 concentration increases proportionally up to 120 mg/day [65, 66]. A minority of patients achieve complete remission or partial remission [65, 66]. While minor side effects, including thrombocytopenia, diarrhea, vomiting, fatigue and hyponatraemia occur, OTX015 is well-tolerated and is currently undergoing phase II clinical trials in patients with acute leukemia, lymphoma or myeloma on a 14 days on and 7 days off schedule (Table 3).

In a Phase I clinical trial of the BRD4 inhibitor CC-90010 in 67 solid tumor and 2 lymphoma patients, one patient each with astrocytoma or endometrial carcinoma achieves a complete response or a partial response, and six additional patients experience prolonged stable disease [67]. Side effects including thrombocytopenia anemia and fatigue are well-tolerated, and CC-90010 at 45 mg on a 4 days on and 24 days off schedule has been proposed for Phase II clinical trials [67] (Table 3). In addition, in a Phase Ib clinical trial in glioblastoma patients, CC-90010 in combination with Temozolomide is safe and well tolerated with encouraging anticancer efficacy [68] (Table 3).

ABBV-075 has been tested in 12 patients with prostate cancer, 72 patients with other solid tumors such as melanoma, colorectal, breast and pancreatic cancers, and 44 patients with acute myeloid leukemia [69, 70]. While ABBV-075 monotherapy shows limited anticancer effects in both solid tumor and leukemia patients, combination therapy with ABBV-075 and the BLC2 inhibitor Venetoclax is considerably more effective. Despite adverse events including dysgeusia, loss of appetite, diarrhea, thrombocytopenia, fatigue, nausea and anemia, ABBV-075 has a good safety profile for Phase II studies at the dose of 1.5 mg daily [69, 70] (Table 3).

In two independent Phase I/II dose-escalation, safety and tolerability studies of the BRD4 inhibitors INCB054329 and INCB057643 in patients with solid tumors or lymphoma, 69 and 134 patients have been recruited to INCB054329 (completed) and INCB057643 (ongoing) studies respectively [71]. Two complete responses and four partial responses have been observed in INCB057643 treatment group; INCB057643 shows a more favorable pharmacokinetic profile than INCB054329; and side effects, including thrombocytopenia, nausea, fatigue and decreased appetite, can be safely managed in both INCB054329 and INCB057643 treated patients [71] (Table 3).

The BRD4 inhibitor Pelabresib (CPI-0610) has shown synergistic anticancer effects, when combined with Ruxolitinib, the current standard of care treatment in myelofibrosis patients, in 84 myelofibrosis patients in a Phase II clinical trial [72]. At 24 weeks, 68% patients reached a reduction in spleen volume of ≥35%, and 56% acquired a reduction in total symptom score of ≥50%. Side effects including thrombocytopenia and anemia are not common and are manageable. Importantly, a double-blinded placebo-controlled Phase III clinical trial is currently ongoing to examine the synergistic anticancer effects of Ruxolitinib and CPI-0610 combination therapy in myelofibrosis patients [73] (Table 3).

The BRD4 inhibitor GSK525762 (Molibresib) has shown promising anticancer effects in a Phase I clinical trial in patients with NUT carcinoma [74, 75]. However, in a dose-escalation Phase I clinical trial of GSK525762 in 87 patients with acute myeloid leukemia, non-Hodgkin lymphoma or multiple myeloma and in a Phase II clinical trial in 24 patients with relapsed/refractory myelodysplastic syndrome or cutaneous T-cell lymphoma, only 6 patients achieved complete response and 7 patients partial responses [76]. Adverse effects such as thrombocytopenia, anemia and neutropenia limit dose escalation and anticancer effects [76] (Table 3).

Two other BRD4 BD1 and BD2 bromodomain inhibitors also show significant toxicity to normal tissues. In the first phase I, open-label, non-randomized clinical trial of the BRD4 inhibitor BAY 1238097 in 8 patients with solid tumors, BAY 1238097 shows on-target effects on BRD4-inhibition biomarkers, such as reduction in MYC expression, but results in dose-limiting toxicities including nausea, vomiting, headache, back pain and fatigue, and the study has been terminated [77] (Table 3). In an open-label Phase I clinical trial of the BRD4 inhibitor ODM-207 in 35 patients with solid tumors including castrate-resistant prostate cancer, no complete or partial responses were observed, and side effects such as thrombocytopenia, anorexia, nausea, diarrhea and fatigue were common, indicating that ODM-207 is not efficacious and has a narrow therapeutic window [78] (Table 3).

Importantly, the BRD4 BD2 domain inhibitor ABBV-744, which shows much less toxicity to normal tissues in preclinical models, has also entered a Phase I clinical trial in relapsed or refractory acute myeloid leukemia patients. However, clinical data have not been published.

Conclusions and future perspective

Characterized by massive histone H3K27 acetylation signal at the loci of cell identity genes and critical oncogenes, super-enhancers are recognized by the BET bromodomain protein BRD4; and super-enhancers and BRD4 play critical roles in oncogene transcriptional activation, over-expression, malignant transformation, cancer cell proliferation, survival, tumor initiation, progression and metastasis in a number cancer types. However, it is important to note that super-enhancers and BRD4 can also activate tumor suppressor gene transcription and suppress drug resistance gene expression. While super-enhancers and BRD4 generally promote tumorigenesis, it is imperative to comprehensively investigate the specific scenarios, such as certain sub-types of cancer cells under particular cellular context, in which super-enhancers and BRD4 exert tumor suppressive, rather than tumorigenic, functions.

BRD4 bromodomain BD1 and BD2 inhibitors have been discovered through small molecule compound library screen, in silico compound screen and chemical synthesis. By blocking BRD4 binding to super-enhancers, BRD4 inhibitors suppress oncogene transcription and expression, reduce cancer cell proliferation and survival, and suppress tumor progress in cancers of a variety of organ origins. However, BRD4 inhibitors, like other targeted therapies, show moderate anticancer effects when employed as a monotherapy. Pre-clinical studies have shown that BRD4 inhibitors exert synergistic anticancer effects when combined with other anticancer agents, such as CDK7 inhibitors, CDK4/CDK6 inhibitors, HDAC inhibitors and BCL-2 inhibitors in vitro and in mouse models of various cancers.

More than a dozen BRD4 BD1 and BD2 bromodomain inhibitors, such as OTX015, have been or are currently in clinical trials in patients with cancer of various organ origins. It is now clear that BRD4 BD1 and BD2 bromodomain inhibitors induce weak to moderate anti-cancer effects in patients as a monotherapy and some of the inhibitors cause significant side effects, such as thrombocytopenia, dysgeusia, diarrhea, fatigue, nausea and anemia. More recently, BRD4 BD2 bromodomain selective inhibitor ABBV-744, PROTAC BRD4 protein degraders such as ARV-771 and A1874, and dual BRD4 and CBP/p300 bromodomain co-inhibitors NEO2734 and XP-524 have been developed and have shown better anticancer effects and/or better safety profile in pre-clinical models. In addition, data from clinical trials of ABBV-744 and NEO2734 are expected to be released, and will further shed lights on the utility of the novel BRD4 inhibitors in the clinical setting.

Future endeavors can focus on developing more potent and selective small molecule compound BRD4 BD2 bromodomain inhibitors to reduce cytotoxicity to normal cells, PROTAC BRD4 protein degraders, and dual BRD4 and CBP/p300 bromodomain co-inhibitors through chemical synthesis, structure-based in silico screen, and wet lab screen of small molecule compound libraries. Their safety profile in normal cells and tissues, pharmacokinetics and anticancer effects can be examined both in vitro and in multiple mouse models. Nevertheless, it should be noted that treatment with BRD4 inhibitors can reduce tumor suppressor gene expression under specific conditions, and that long-term treatment with BRD4 inhibitors can result in cancer cell resistance to a broad spectrum of anticancer agents. It is therefore important to investigate the specific scenarios, such as certain sub-types of cancer cells under particular context and chemotherapy-naïve or -exposed cancer cells and mouse models, in which BRD4 inhibitors reduce tumor suppressor gene expression, augment drug resistance gene expression and render cancer cell resistance to anticancer agents.

As all targeted therapies are expected to be employed in the clinic in combination therapies, the other anticancer agents which exert the best synergistic anticancer effects with BRD4 inhibitors should be identified by unbiased screening of approved anticancer drug libraries against each cancer type. Ultimately, the best combination therapies with BRD4 inhibitors and other anticancer drugs are expected to be tested in clinical trials in patients.