Search

Our previous finding that the BET inhibitor (BETi) JQ1 increases levels of the DNA damage marker γH2AX suggested that JQ1 might enhance the sensitivity of tumor cells to PARP inhibitors (PARPi), which are selectively toxic to cells that harbor relatively high levels of DNA damage. To address this hypothesis, we evaluated the effect of a BETi (JQ1 or I-BET762) combined with a PARPi (olaparib or veliparib) in KKU-055 and KKU-100 cholangiocarcinoma (CCA) cell lines and of JQ1 with olaparib in a xenograft model of CCA. Each combination was more effective than any of the four drugs as single agents. Combination indices ranged from 0.1 to 0.8 at the ED50 for all combinations, indicating synergy and demonstrating that synergy was not limited to a specific combination. Mechanistically, downregulation of BETi molecular targets BRD2 or BRD4 by shRNA sensitized CCA cells to BETi as single agents as well as to the combination of a BETi + a PARPi. Our data indicate that combinations of a BETi with a PARPi merit further evaluation as a promising strategy for CCA.

Gemcitabine is used to treat pancreatic cancer (PC), but is not curative. We sought to determine whether gemcitabine + a BET bromodomain inhibitor was superior to gemcitabine, and identify proteins that may contribute to the efficacy of this combination. This study was based on observations that cell cycle dysregulation and DNA damage augment the efficacy of gemcitabine. BET inhibitors arrest cells in G1 and allow increases in DNA damage, likely due to inhibition of expression of DNA repair proteins Ku80 and RAD51. BET inhibitors (JQ1 or I-BET762) + gemcitabine were synergistic in vitro, in Panc1, MiaPaCa2 and Su86 PC cell lines. JQ1 + gemcitabine was more effective in vivo than either drug alone in patient-derived xenograft models (P &lt, 0.01). Increases in the apoptosis marker cleaved caspase 3 and DNA damage marker γH2AX paralleled antitumor efficacy. Notably, RNA-seq data showed that JQ1 + gemcitabine selectively inhibited HMGCS2 and APOC1 ~6-fold, compared to controls. These proteins contribute to cholesterol biosynthesis and lipid metabolism, and their overexpression supports tumor cell proliferation. IPA data indicated that JQ1 + gemcitabine selectively inhibited the LXR/RXR activation pathway, suggesting the hypothesis that this inhibition may contribute to the observed in vivo efficacy of JQ1 + gemcitabine.

Pancreatic ductal adenocarcinoma (PDAC), the most common type of pancreatic cancer (PC), is now the third leading cause of cancer related deaths in the US. This is a highly aggressive disease in that 80% of patients present with locally advanced or metastatic disease and their only treatment option is systemic chemotherapy. Ultimately patient tumors develop resistance to therapy and resulting in a median survival of ~6 months. Therefore, there is an imperative need to identify therapies that provide a more durable response for this patient population. The family of bromodomain and extraterminal domain (BET) proteins has recently become a target of interest for the treatment of PDAC. The BET proteins (BRD2, BRD3, BRD4, and BRDT) function to regulate transcription by recruiting positive transcriptional activators to the promoters of genes. Our lab has shown that inhibition of BET bromodomain function using the BET inhibitor JQ1 suppresses PDAC tumor growth in vivo and inhibits cell viability in vitro. Importantly, PDAC cells and tumors exposed to JQ1 show evidence of DNA damage. We hypothesized that because of the role BET proteins are known to play in regulating gene expression, that the observed JQ1-induced DNA damage may result from JQ1 inhibiting the expression of DNA repair genes whose expression are dependent on BET protein transcriptional complexes. qRT-PCR and immunoblot analysis demonstrated that treatment of the pancreatic cancer cell line BxPC3 with JQ1 inhibited the expression of two double strand break repair proteins, Ku80 and RAD51, both of which have been shown to be overexpressed in cancer including PDAC. Although it has been established that JQ1 co-occupies 99% of the genomic loci that BRD4 is known to bind to, specific gene products whose expression is dependent on BRD4 in PDAC have not been thoroughly investigated. The goal of the current study was to determine if the expression of DNA repair genes Ku80 and RAD51 were specific gene targets of BRD4 transcriptional complexes. We exposed BxPC3 cells to JQ1 for 48 hours and assessed changes in the association of BRD4 with the promoter loci of Ku80 and RAD51 using chromatin immunoprecipitation (ChIP) assays. qRT-PCR of the ChIP demonstrated that treatment with JQ1 significantly (p Citation Format: Aubrey Lynn Miller, Samuel C. Fehling, Patrick L. Garcia, Karina J. Yoon. BET inhibition decreases the expression of DNA repair enzymes in pancreatic cancer [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2019; 2019 Mar 29-Apr 3; Atlanta, GA. Philadelphia (PA): AACR; Cancer Res 2019;79(13 Suppl):Abstract nr 5244.

Pancreatic ductal adenocarcinoma (PDAC), the most common type of pancreatic cancer, is a highly aggressive tumor with a 5-year survival of We and others document that inhibiting the function of BET bromodomain proteins suppresses PDAC tumor growth in preclinical models. The BET family of proteins is characterized by two N-terminal bromodomains, BD1 and BD2, which bind to acetylated lysine residues on specific histones. The binding of BET proteins to acetylated lysines promotes the formation of transcriptional complexes, and impacts gene transcription and chromatin structure. The literature suggests the potential specific roles for BD1 vs BD2 using pharmacological inhibitors and dominant negative mutational analyses. However, the relative importance of each bromodomain with respect to transcriptional regulation compared to chromatin structure is not well understood. Further, there is suggestion in the literature that each bromodomain function may differ in specific tumor cell types. Functional differences between BD1 and BD2 in PDAC cells have not been reported. The goal of the current study was to compare the efficacy of the BD2-specific BET inhibitor (BETi) SJ018 with the efficacy of the pan-BET inhibitor I-BET762 in PDAC cells. We exposed four PDAC cell lines (Panc1, BxPC3, MiaPaCa2, and CFPAC) to a range of concentrations of SJ018 or I-BET762 for 96 hours, and assessed cell viability using alamarBlue. Data, analyzed using GraphPad Prism 7, demonstrated that SJ018 exerted up to ~60-fold lower IC50 values across all cell lines tested compared to the pan-BET inhibitor I-BET762. Further, we recently made a novel observation that the pan-BETi JQ1 induced DNA damage and inhibits expression of the DNA repair protein Rad51 in PDAC cells. Immunoblot analysis for phosphorylated Ser-139 H2AX (γH2AX), a marker of DNA damage, demonstrated that SJ018 and I-BET762 also increased DNA damage, and that the BD2-specific SJ018 induces DNA damage to a greater extent and at a lower dose than the BD1/BD2 inhibitor I-BET762 in BxPC3 cells. The data also showed that both SJ018 and I-BET762 inhibited expression of the homologous recombination protein Rad51 in vitro. We conclude that SJ018 is more potent than I-BET762 in PDAC cells as a single agent, and that degree of potency is reflected by higher levels of DNA damage. Citation Format: Aubrey Lynn Miller, Peter J. Slavish, Brandon M. Young, Philip M. Potter, Anang A. Shelat, Karina J. Yoon. Targeting the BD2 BET bromodomain to inhibit proliferation of pancreatic ductal adenocarcinoma cells [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2018; 2018 Apr 14-18; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2018;78(13 Suppl):Abstract nr 5794.

Cholangiocarcinoma (CCA) is an aggressive bile duct neoplasm, and the second most common primary hepatic malignancy. CCA is usually diagnosed at a late stage where the current standard of care, resection followed by gemcitabine with or without cisplatin, is not effective. Up to 90% of CCA patients are ineligible for resection and, of those eligible, postoperative chemotherapy does not significantly prolong overall survival. The limited number of CCA models hinders drug development and discovery of new therapeutic targets. Mutations have been identified in KRAS (17% of CCA tumors), p53 (44%) and SMAD4 (17%), but are not recognized as essential to CCA progression. c-Myc protein is highly expressed in 94% of CCA cases while low to undetectable in normal adult liver, suggesting c-Myc overexpression may contribute to CCA tumor progression but this hypothesis has not been tested. Toward developing novel effective therapies for CCA, we evaluated the effect of the bromodomain and extraterminal domain (BET) inhibitor JQ1 on c-Myc expression and on the proliferation of CCA cells. JQ1 inhibits c-Myc transcription by binding to the acetylated lysine (K-Ac) binding pockets of BET protein family members (BRD2, BRD3, BRD4 and BRDT), preventing their interaction with K-Ac residues on nuclear proteins, disrupting the formation of functional transcriptional complexes and inhibiting c-Myc transcription. We have shown that JQ1 inhibited tumor growth in a patient-derived xenograft (PDX) model of CCA and downregulated expression of c-Myc and its transcriptional targets BRCA2 and Chk1, both involved in DNA damage response. Our current data show that JQ1 decreased c-Myc protein expression, CCA cell viability, and clonogenic potential in KKU-055 CCA cell line. Of note, the literature suggests that tumors deficient in BRCA2 or Chk1 are relatively sensitive to PARP inhibitors (PARPi). Therefore, we hypothesize that JQ1-induced downregulation of c-Myc and its transcriptional targets BRCA2 and Chk1 would potentiate the efficacy of PARPi. We exposed CCA cell line KKU-055 with a range of JQ1 and PARPi (olaparib and veliparib) concentrations as single agent or combination therapy. Each combination of JQ1 with either PARPi was more effective than any of the three drugs as single agents. Combination indices (CI), calculated using CompuSyn 1.0, ranged from 0.05 to 0.09 which indicated strong synergy for JQ1 with olaparib or veliparib. Further, synergy with JQ1 was not dependent on a single PARPi. JQ1 with olaparib (1:1) also decreased the clonogenic potential of KKU-055 cells, compared to each drug as a single agent. These data suggest that inhibition of BET protein function sensitizes CCA cells to PARPi, coincident with downregulation of c-Myc and its transcriptional targets BRCA2 and Chk1. Research funded by NIH/NCI R21 CA205501 Citation Format: Samuel C. Fehling, Aubrey L. Miller, James E. Bradner, Karina J. Yoon. The BET inhibitor JQ1 sensitizes cholangiocarcinoma cells to PARP inhibitors [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2018; 2018 Apr 14-18; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2018;78(13 Suppl):Abstract nr 5824.

Cholangiocarcinoma (CCA) is a lethal malignancy of the biliary epithelium that can arise in any part of the biliary tree. Surgery is the only curative treatment for CCA, but only ∼30% of patients present with resectable disease. The remaining 70% of patients present with advanced or metastatic disease and, if eligible, undergo systemic chemotherapy with the first-line combination of gemcitabine and cisplatin. This combination was shown in a phase II clinical trial to significantly increase median survival from 8.1 to 11.7 months, compared to gemcitabine alone. In order to improve on current treatment, pre-clinical evaluation of novel therapeutics is essential to improving outcome. Unfortunately, the paucity of data describing characteristics common to CCA make development of targeted therapy difficult. However, as is true for other types of solid tumors, c-Myc expression likely contributes to CCA phenotype: c-Myc expression has been observed in 95% of CCA tumors, and experimental down-regulation of c-Myc decreases the invasive potential of CCA cells in vitro. Recently it has become possible to inhibit expression of c-Myc using BET inhibitors. Therefore, we evaluated the efficacy of the bromodomain (BET) inhibitor JQ1 using in vivo models of CCA. The five patient-derived xenograft (PDX) models of CCA that we developed are the first such models to be reported. These models retain the heterogeneity, architecture and specific genetic characteristics of the primary tumors from which they were derived. We used three of these models to examine whether the BET inhibitor JQ1 inhibited CCA tumor growth and generated expression profiles of vehicle- and drug-treated tumors. We administered 50 mg/kg of JQ1 i.p. daily for 20 days and monitored tumor growth. This treatment regimen was well tolerated by tumor-bearing mice, without apparent toxicity. Our data demonstrate that JQ1 suppressed tumor growth in two of the three models, compared to vehicle control treated mice. The data also showed that JQ1-treated tumors had lower levels of c-Myc RNA (↓5-fold) and protein and of RNA encoding multiple transcriptional targets downstream of this oncogenic transcription factor. We conclude that BET inhibitors such as JQ1 warrant further investigation as potentially effective drugs for the treatment of CCA. Citation Format: Patrick L. Garcia, Aubrey L. Miller, Kelly Kreitzburg, Tracy L. Gamblin, Leona N. Council, John D. Christein, Pablo Arnoletti, Marty Heslin, Sushanth Reddy, Joseph H. Richardson, Eddy S. Yang, Jun Qi, James E. Bradner, Karina J. Yoon. Bromodomain inhibitor JQ1 inhibits cholangiocarcinoma tumor growth in patient-derived xenograft models. [abstract]. In: Proceedings of the 106th Annual Meeting of the American Association for Cancer Research; 2015 Apr 18-22; Philadelphia, PA. Philadelphia (PA): AACR; Cancer Res 2015;75(15 Suppl):Abstract nr 1779. doi:10.1158/1538-7445.AM2015-1779