Pleiotropic Impact of DNA-PK in Cancer and Implications for Therapeutic Strategies Emanuela Dylgjeri1,2, Christopher Mcnair1,2, Jonathan F

Published OnlineFirst July 2, 2019; DOI: 10.1158/1078-0432.CCR-18-2207

Translational Cancer Mechanisms and Therapy Clinical Cancer Research Pleiotropic Impact of DNA-PK in Cancer and Implications for Therapeutic Strategies Emanuela Dylgjeri1,2, Christopher McNair1,2, Jonathan F. Goodwin1,2, Heather K. Raymon3, Peter A. McCue4, Ayesha A. Shaﬁ1,2, Benjamin E. Leiby2,5, Renee de Leeuw1,2, Vishal Kothari4, Jennifer J. McCann1,2, Amy C. Mandigo1,2, Saswati N. Chand1,2, Matthew J. Schiewer1,2, Lucas J. Brand1,2, Irina Vasilevskaya1,2, Nicolas Gordon1,2, Talya S. Laufer1,2, Leonard G. Gomella4, Costas D. Lallas4, Edouard J. Trabulsi4, Felix Y. Feng6,7,8, Ellen H. Filvaroff3, Kristin Hege3, Dana Rathkopf9, and Karen E. Knudsen1,2,4,10

Abstract

Purpose: DNA-dependent protein kinase catalytic subunit DNA-PK suppresses tumor growth both in vitro, in vivo, and (DNA-PK) is a pleiotropic kinase involved in DNA repair and ex vivo; (iii) DNA-PK transcriptionally regulates the known transcriptional regulation. DNA-PK is deregulated in selected DNA-PK–mediated functions as well as novel cancer-related cancer types and is strongly associated with poor outcome. The pathways that promote tumor growth; (iv) dual targeting of underlying mechanisms by which DNA-PK promotes aggres- DNA-PK/TOR kinase (TORK) transcriptionally upregulates sive tumor phenotypes are not well understood. Here, unbi- androgen signaling, which can be mitigated using the andro- ased molecular investigation in clinically relevant tumor mod- gen receptor (AR) antagonist enzalutamide; (v) cotargeting AR els reveals novel functions of DNA-PK in cancer. and DNA-PK/TORK leads to the expansion of antitumor Experimental Design: DNA-PK function was modulated effects, uncovering the modulation of novel, highly relevant using both genetic and pharmacologic methods in a series of protumorigenic cancer pathways; and (viii) cotargeting in vitro models, in vivo xenografts, and patient-derived explants DNA-PK/TORK and AR has cooperative growth inhibitory (PDE), and the impact on the downstream signaling and effects in vitro and in vivo. cellular cancer phenotypes was discerned. Data obtained were Conclusions: These ﬁndings uncovered novel DNA-PK used to develop novel strategies for combinatorial targeting of transcriptional regulatory functions and led to the develop- DNA-PK and hormone signaling pathways. ment of a combinatorial therapeutic strategy for patients with Results: Key ﬁndings reveal that (i) DNA-PK regulates advanced prostate cancer, currently being tested in the clinical tumor cell proliferation; (ii) pharmacologic targeting of setting.

Introduction Multiple DNA damage repair (DDR) mechanisms have been 1 Department of Cancer Biology at Thomas Jefferson University, Philadelphia, selected for through evolution to preserve genomic integrity. DNA 2 Pennsylvania. Sidney Kimmel Cancer Center at Thomas Jefferson University, double-strand breaks (DSB) are the most deleterious and toxic Philadelphia, Pennsylvania. 3Celgene Corporation, San Francisco, California. 4Department of Urology, Sidney Kimmel Cancer Center Thomas Jefferson forms of damage that, if left unrepaired, lead to cell-cycle arrest and University, Philadelphia, Pennsylvania. 5Department of Pharmacology and cell death (1, 2). Two main pathways are employed to repair DSB: Experimental Therapeutics, Thomas Jefferson University, Philadelphia, Penn- homologous recombination (HR), which utilizes a sister chroma- sylvania. 6Department of Radiation Oncology, University of California, San tid in close proximity as a template resulting in high-ﬁdelity DSB 7 Francisco, San Francisco, California. Department of Urology, University of repair (3, 4); and nonhomologous end-joining (NHEJ), which 8 California, San Francisco, San Francisco, California. Department of Medicine, does not require a sister chromatid template resulting in a more University of California, San Francisco, San Francisco, California. 9Memorial Sloan Kettering Cancer Center, New York, New York. 10Departments of Medical error-prone form of repair that can occur throughout the cell Oncology and Radiation Oncology, Thomas Jefferson University, Philadelphia, cycle (5, 6). Although both the processes aid in maintaining Pennsylvania. genomic integrity in normal cells, cancer cells utilize these pro- Note: Supplementary data for this article are available at Clinical Cancer cesses, including upregulation of key DDR proteins, to acquire Research Online (http://clincancerres.aacrjournals.org/). more aggressive phenotypes, and develop resistance to DNA- damaging agents (7). Therefore, targeting the DNA repair machin- Corresponding Author: Karen E. Knudsen, Thomas Jefferson University, 233 South 10th Street, Bluemle (BLSB) 1050, Philadelphia, PA 19107. Phone: 215-503- ery and/or its components that are deregulated in cancer has the 5692; Fax: 215-923-4498; E-mail: [email protected] potential to be employed as anticancer therapeutic strategies. Among many DDR proteins deregulated in cancer, DNA- Clin Cancer Res 2019;25:5623–37 dependent protein kinase catalytic subunit (DNA-PKcs, referred doi: 10.1158/1078-0432.CCR-18-2207 to as DNA-PK herein), a key DNA repair protein involved in NHEJ, 2019 American Association for Cancer Research. is known to play a protumorigenic role in many cancers including

www.aacrjournals.org 5623

Dylgjeri et al.

processes. Genetic and pharmacologic inhibition, using a specific Translational Relevance laboratory grade DNA-PK inhibitor (NU7441) and a clinical DNA-dependent protein kinase catalytic subunit (DNA-PK) grade dual DNA-PK/TOR kinase (DNAPK/TORK) inhibitor is a driver of aggressive disease and has been nominated as a (CC-115; refs. 22–26), led to the inhibition of proliferation in therapeutic target in multiple cancer types. Targeting DNA-PK castration-resistant prostate cancer (CRPC) models. Unbiased is an attractive therapeutic strategy that can lead to significant transcriptomic analyses demonstrated the modulation of path- anticancer effects. However, further understanding of DNA-PK ways known to be regulated by DNA-PK, including androgen functions, especially transcriptional regulation, is essential for response, estrogen signaling, cell cycle, and proliferation path- the development of effective treatments. This study demon- ways. Novel processes of cancer relevance modulated by DNA-PK, strates that DNA-PK transcriptionally modulates gene net- including oxidative phosphorylation, epithelial–mesenchymal works beyond its known function in DNA repair, hormone transition, TNFa signaling via NF-kB, TGFb signaling, and KRAS signaling, and metastatic pathways. Data herein identified signaling were also uncovered. Inhibition of DNA-PK/TORK via novel DNA-PK–mediated functions including the regulation CC-115 led to the transcriptional upregulation of androgen of epithelial–mesenchymal transition, immune response, and signaling due to TORK inhibition, which was expected on the metabolic processes. Moreover, the unbiased transcriptomic basis of previous studies. The observed upregulation of androgen data in this study informed the investigation of a combina- response upon dual DNA-PK/TORK inhibition served as a torial strategy targeting DNA-PK/TOR kinase (TORK), and rationale to test the combination with androgen receptor (AR) androgen receptor (AR), which is currently being evaluated antagonist, enzalutamide. Combinatorial treatment of DNA-PK in the clinical setting in castration-resistant prostate cancer targeting agents with enzalutamide resulted in an expansion of (CRPC). The data presented in this study have led to bench-to- the transcriptional changes and uncovered distinct downstream bed discoveries that have the potential to affect the manage- transcriptional alterations as compared with single-agent tar- ment and treatment of CRPC in the clinical setting. geting including Wnt–b-catenin signaling, Hedgehog signaling, inflammatory response, and immune response signaling. Fur- thermore, several cancer-relevant pathways regulated exclusively by DNA-PK were identified by comparing the transcriptional prostate, breast, colon, cervix, and chronic leukemias (8–11). effects caused by a specific TORK inhibitor (CC-223), with the DNA-PK is upregulated as a function of disease progression in dual DNA-PK/TORK inhibitor (CC-115), both in combination prostate cancer, among other, and both DNA-PK overexpression with enzalutamide. Finally, cotargeting of DNA-PK/TORK and and hyperactivation are associated with aggressive disease (12). AR led to cooperative antiproliferative effects in vitro, in vivo, Increased DNA-PK expression and activity correlate with resis- and ex vivo in PDEs. In sum, the data herein demonstrate that in tance to both chemotherapy and radiation therapy and overall the absence of exogenous DNA damage, DNA-PK regulates poor outcomes, thus nominating DNA-PK as a potential thera- protumorigenic pathways that can be effectively targeted using peutic target in the management of cancer (8, 13). Previous clinically relevant pharmacologic agents and that DNA-PK studies in multiple tumor types, including prostate cancer, have inhibitors can act in concert with AR antagonists in advanced shown that downregulation of DNA-PK via genetic perturbation prostate cancer. or pharmacologic inhibition leads to sensitization to radiation, decreased tumor size, decreased metastasis, and increased survival, thus providing the rationale for investigating DNA-PK inhibi- Materials and Methods tors in the clinic for prostate cancer (12, 14–16). Proliferation assay Although the DNA repair functions of DNA-PK are well estab- Cell lines. Cells lines, C4-2, 22Rv1, LNCaP, VCaP, and LN95 were lished, the kinase has also been shown to regulate multiple authenticated by ATCC and checked for Mycoplasma upon thawing cellular processes of cancer relevance, including genomic stability, and at termination of maintenance after <20 passages. cell cycle, metabolism, metastasis, and transcriptional regulation (8, 17–19). Previous studies have shown that DNA-PK plays Inhibitors. Cell lines LNCaP, VCaP, C4-2, 22Rv1, and LN95, were a protective role against drug-induced apoptosis, and promotes plated in 96-well plates at 1,500, 1,500, 500, 1,000, and 1,000 proliferation and radiation resistance (18, 20). Moreover, DNA- cells/well concentration, respectively. All cell lines were grown in PK is involved in transcriptional regulation of prometastatic gene full serum culture media, with the exception of LN95, which are networks in prostate cancer and secretion of metastasis-associated normally cultured in charcoal-stripped serum. Cells were treated factors that influence the tumor microenvironment leading to the next day with 0–25 mmol/L concentrations of NU7441, migration and invasion in melanoma (12, 21). Given these roles CC-115, and CC-223 for 6 days and compared with vehicle of DNA-PK in cancer, therapeutic agents to target DNA-PK kinase control, DMSO. IC25,IC50, and IC75 (when possible) were deter- activity have been developed and are being tested in numerous mined for CC-115 in each cell line. Combination treatments with clinical trials (NCT02516813, NCT02316197, NCT01353625, enzalutamide and CC-115 þ enzalutamide were conducted by and NCT02833883). treating the cells with CC-115 IC25,IC50, and IC75 dose and To further understand the functions of DNA-PK, unbiased titrating the concentration of enzalutamide (0–25 mmol/L). After molecular investigation was performed in clinically relevant 6 days, Quant-iT PicoGreen dsDNA Assay Kit (Abcam) was used tumor models with the further goal of suppressing pleiotropic according to protocol and data were recorded using a BioTek DNA-PK functions. Using a series of in vitro models, in vivo Synergy HT plate reader. Results were analyzed and graphed xenografts, and ex vivo patient-derived explants (PDE), it was using GraphPad Prism7 to generate dose-response curves for the demonstrated that DNA-PK regulates tumor cell proliferation as single-agent treatments. The combination treatment data were well as modulates the known and novel DNA-PK transcriptional analyzed using CompuSyn Software analysis, to determine the

5624 Clin Cancer Res; 25(18) September 15, 2019 Clinical Cancer Research

Combined DNA-PK/TORK/AR Targeting in Prostate Cancer

combination index of CC-115 and enzalutamide using noncon- pestle and mortar (Potter–Elvehjem tissue grinder). For multiday stant ratio parameters. studies, the compound was freshly formulated every third day. Between doses, the formulated compound was stored under siRNA. The CRPC cell lines, C4-2 and 22Rv1, cells were seeded at a constant stirring using magnetic stirrer at 4C in the dark. The 5 – 1 10 density on poly L-lysine coated plates in culture media for treatment's article and vehicle were administered by oral gavage. 24 hours. Cells were then transfected for 8 hours in serum-free PRKDC media conditions with either control or siRNA pools Development of CRPC tumor model. LNCaP tumor cells (6 106/ fi fi (Thermo Scienti c Scienti c) according to manufacturer's proto- mouse) were injected subcutaneously into the hind flanks of male col as previously described (12). Cells were then maintained in CB17 SCID mice. When the tumor volumes reached approxi- complete media for 96 hours post transfection and processed for mately 200 mm3, the mice were surgically castrated, and the either RNA, protein, or growth assays. tumor growth was closely monitored. Relapsed tumors that reached 300 to 500 mm3 in volume were used as donor tissue Bromodeoxyuridine incorporation assay for transplanting to the next cohort of castrated mice. For trans- Bromodeoxyuridine (BrdU) labeling and detection were per- plantation, tumors from several donor mice were pooled, minced, formed as previously described (12, 14). Samples were acquired and mixed with Matrigel (BD Biosciences) before implanting using a GUAVA easyCyte flow cytometer and analyzed using into the hind flank of the recipient mice. Approximately 25 to InCyte software for BrdU incorporation. 50 mg of tumor tissue per mouse was transplanted for each Western blotting mouse. Several cycles (2–5) of tumor transplantation were per- Cells were treated as specified and cell lysates were generated as formed until enhanced tumorigenicity (80%–90% tumor take- previously described (12). AR (N-20, directed against amino acids rate) was observed. 1-20), DNA-PK (Thermo Fisher Scientific, #MS-423-P), vinculin (Sigma-Aldrich, #V9264-200UL), VAV3 (EMD Millipore, #07-464), Efficacy studies with CRPC model. For efficacy studies, tumor Prex1 (EMD Millipore, #MABC178), lamin B (Santa Cruz Bio- fragments from passage 6 were used for inoculations. Twenty technology, #6217), pAKT (Cell Signaling Technology, 9271S), days after tumor fragment inoculation, mice bearing HR-LNCaP 3 AKT (Cell Signaling Technology, 9272S), pS6 (Cell Signaling tumors of 200–400 mm were randomly assigned to receive oral Technology, 2211S), S6 (Cell Signaling Technology, 2217S), doses of vehicle, CC-115, enzalutamide, or combination of PARP/cleaved PARP (Cell Signaling Technology, 9542S) antibo- CC-115 and enzalutamide once a day for the duration of the dies were used for immunoblotting. study. Tumor volumes were determined before the initiation of treatment and were considered as the starting volumes. Tumors Gene expression were measured twice a week for the duration of the study. The long Cells were treated as specified above. RNA was isolated using and short axes of each tumor were measured using a digital caliper TRIzol (Life Technologies) and quantitative PCR was conducted in millimeters. The tumor volumes were calculated using the using primers as shown in Supplementary Table S1. formula: width2 length/2 and expressed in cubic millimeters Mouse gene expression analysis was performed by extracting (mm3). total RNA from tumor tissues using TRIzol Reagent (Sigma). RT-PCR was performed using One-Step RT-PCR Kit, SuperScript In vivo target validation One-Step RT-PCR Systems (Life Technologies), following manu- pS6RP in tumor samples was determined using Meso Scale facturer's instructions. Probes for human FKBP5 (Hs00188025) Discovery Kit [pS6RP: MA6000 p-S6RP (Ser 235/236)], Whole was purchased from Life Technologies. Cell Lysate Kit (MSD, #K110DFD-2), and expressed as the mean Meso Scale Counts SEM per manufacturer's instructions. PDE pAKT (S473) and the total AKT in tumor samples were mea- PDE experiments were conducted as described previously sured simultaneously using the Meso Scale Discovery Kit [pAKT (12, 27, 28). Tissues samples were treated as specified. IHC (S473)]/Total AKT: MS6000 Phospho(S473)/Total AKT Whole staining for Ki67 was performed by Thomas Jefferson Pathology Cell Lysate (MSD, #K11100D-2) multiplex assay. The amount of Core Facilities and scored by a board-certified pathologist who phosphorylated AKT was calculated per manufacturer's instruc- reported percent Ki67 positivity after counting all cancer cells in tions and reported as a percentage of total AKT. The results were the slide provided. PDEs are Institutional Review Board exempt expressed as the mean percentage phosphorylated AKT SEM for due to deidentification of specimens. The Thomas Jefferson each group. University Institutional Review Board has reviewed this procure- pDNA-PK (S2056) and total DNA-PK in tumor samples were ment protocol and determined this research to be in compliance measured simultaneously using IHC. Five- to 10-mm thick cryostat with federal regulations governing research on deidentified speci- sections were used. Shortly, frozen sections were fixed in 4% mens and/or clinical data [45 CFR 46.102(f)]. paraformaldehyde for 10 minutes at room temperature, washed Mouse models in PBS, blocked, and permeabilized with 5% normal goat serum Animals. Male 6- to 8-weeks-old CB17 SCID mice were obtained and 0.3% Triton X-100. Sections were then incubated with a from Charles River Laboratories. All animal studies were per- cocktail of primary antibodies (1 mg/mL of anti-mouse anti- fi formed under the protocols approved by Institutional Animal human DNA-PKcs mAb; Thermo Fisher Scienti c, #MS-369-P0) – Care and Use Committees. and anti pDNA-PK [rabbit anti-human pDNA-PK (S2056) poly- clonal antibody (Abcam, ab18192)] for 2 hours followed by Formulation. Suspensions of CC-115 and enzalutamide were incubation with a cocktail of secondary antibodies (60 minutes). prepared in aqueous 0.5% carboxymethyl cellulose and 0.25% The sections were washed, counterstained with Hoechst dye Tween-80. The formulations were homogenized using a Teflon (0.4 mg/mL), and mounted with antifade reagent. The sections

www.aacrjournals.org Clin Cancer Res; 25(18) September 15, 2019 5625

Dylgjeri et al.

were visualized with a Nikon E800 microscope and data were tion significantly reduced cells in active S-phase in C4-2 and quantitated using Metamorph software. Using a 20 objective, 22Rv1 by 20.7% and 4.5%, respectively (Fig. 1B, Supplementary five different fields from each section and four tumors from each Fig. S1B). In summary, DNA-PK suppression inhibits cell prolif- treatment or control group were used for quantitation. The data eration and cell-cycle progression. are expressed as the percentage threshold area of pDNA-PK Although these and previous findings link DNA-PK to the staining over the threshold area of DNA-PKcs (total DNAPK regulation of cell proliferation, coupled with previous data dem- staining). The data from each individual animal was used to onstrating that tumor-associated DNA-PK acutely promotes calculate the mean SEM for each group. metastasis (12, 21, 33), the overall mechanisms by which DNA-PK promotes disease progression are not completely under- Statistical analysis stood. An RNA-seq approach was utilized to delineate the molec- In vitro data are presented as mean SD, xenograft and in vivo ular functions of DNA-PK in CRPC. DNA-PKi resulted in alter- biomarker data are presented as mean SEM. Statistical analyses, ation of transcriptional networks, with 980 upregulated and P including P values, performed using GraphPad Prism7. 1,346 downregulated transcripts (Fig. 1C, adj < 0.05). GSEA was performed using the Hallmark pathway analysis from the Molec- ular Signatures Database (MSigDB) to identify the pathways RNA sequencing modulated by DNA-PK (Fig. 1C, right); significantly over-repre- fi RNA sequencing. C4-2 cells were plated in hormone-pro cient sented pathways (FDR < 0.25) are shown using a circos plot where conditions at 50,000 cells per plate overnight, followed by significantly enriched and deenriched pathways are represented 1-mmol/L single-agent drug treatments (NU7441, CC-115, by blue and green ribbons, respectively (FDR < 0.25). Consistent CC-223, and enzalutamide) in biological triplicate. 22Rv1 cells with previous reports, GSEA revealed the deenrichment in path- fi were plated in hormone-pro cient conditions at 100,000 cells per ways known to be modulated by DNA-PK, including androgen plate overnight, followed by 1 mmol/L treatment with CC-115 and response, DNA repair, cell cycle/proliferation, and protumori- vehicle control for 24 hours, with experiments collected in trip- genic processes, thus recapitulating the highly selective nature of fi licate. RNA was extracted and puri ed using TRIzol and RNAeasy DNA-PK's transcriptional regulatory functions (8, 14). In addi- Mini Kit (Qiagen) according to the manufacturer's instructions. tion, this study of DNA-PK inhibition revealed novel pathways RNA sequencing (RNA-seq) libraries were subsequently con- affected by DNA-PK including TGFb signaling, KRAS signaling, structed using the TruSeq Stranded Total RNA Library Prep Gold TNFa signaling via NF-kB, oxidative phosphorylation, and Kit (protocol # 15031048 Rev E) and sequenced on Illumina's unfolded protein response (Fig. 1C, right). These clinically rele- NextSeq 500 sequencer at the Sidney Kimmel Cancer Sequencing vant pathways are known to play important roles in tumor core facility using paired-end 75 bp reads. For combination progression, thus highlighting the importance of delineating the treatments all drugs were used at 1 mmol/L concentration. roles of DNA-PK with respect to transcriptional regulation in cancer to reveal novel mechanisms of action. Together, these data RNA-seq analyses. RNA-seq was aligned against the hg19 human demonstrate that the pro-proliferative functions of DNA-PK in genome using STAR v2.5.2a (29). Differential gene expression was cancer cells are associated with distinct protumorigenic transcrip- generated using DESeq2 v1.12.4 (30). Gene set enrichment anal- tional regulatory events. ysis (GSEA) was performed using gene sets from the Molecular Signature Database (31). Circos plots were created using Circos Targeting DNA-PK using a therapeutically active compound v0.69-3 (32). RNA-seq data have been deposited in the Gene inhibits tumor cell proliferation and regulates known DNA-PK Expression Omnibus (GEO) repository under the accession num- transcriptional processes ber GSE116765. Although NU7441 is a highly specific DNA-PK inhibitor, this agent is not suited for clinical use. However, the dual DNA-PK and TORK inhibitor (CC-115; Supplementary Fig. S1C) has been Results recently developed and is in multiple clinical trials. Assessment DNA-PK regulates tumor cell proliferation in CRPC of the dual kinase targeting in hormone-sensitive prostate cancer DNA-PK is a multifunctional kinase that plays pleiotropic roles (HSPC) and CRPC models revealed that targeting DNA-PK in biological processes, including DNA-repair, transcriptional using both CC-115 and NU7441 reduced cell viability in a regulation, and genomic instability (8, 18). In prostate cancer, dose-dependent manner, with CC-115 inducing apoptosis DNA-PK is associated with the development of metastatic disease, (Fig. 2A, Supplementary Fig. S1D). Similarly, FACS analysis dem- and functional data support the concept that this is due, in large onstrated that targeting of DNA-PK with CC-115 significantly part, to the transcriptional functions of DNA-PK (12, 18, 21). To reduced cells in active S-phase compared with vehicle control and better understand the roles of DNA-PK in advanced prostate NU7441-treated cells (Supplementary Fig. S1B and S1E). These data cancer and investigate DNA-PK as a therapeutic target, CRPC suggest that targeting DNA-PK using CC-115 more potently inhibits models were utilized to interrogate the effects of targeting tumor cell proliferation compared with NU7441 in both HSPC and DNA-PK expression through genetic and pharmacologic pertur- CRPC models. Considering the important role of DNA-PK in bation. Depletion of DNA-PK using RNAi (Fig. 1A, left) and the transcriptional regulation, it was imperative to identify the tran- inhibition of DNA-PK activity using a highly specific pharmaco- scriptomic alterations caused by this therapeutically active agent. logic inhibitor (NU7441; Fig. 1A, right; Supplementary Fig. S1A) To investigate the impact of targeting DNA-PK using the dual in CRPC cell models resulted in decreased tumor cell proliferation kinase inhibitor on the transcriptome, RNA-seq analysis was (by 45.8% and 50.0% in C4-2 and 26% and 43.6% in 22Rv1, performed in CRPC cells treated with CC-115. Principal com- respectively). In addition, the assessment of 5- BrdU incorpo- ponent analysis (PCA) and sample clustering provided a high ration by actively proliferating cells shows that DNA-PK inhibi- level of confidence in the effects observed for each treatment as

5626 Clin Cancer Res; 25(18) September 15, 2019 Clinical Cancer Research

Combined DNA-PK/TORK/AR Targeting in Prostate Cancer

demonstrated by tight sample clustering between their respec- scripts in 22Rv1 (Fig. 2B; Supplementary Fig. S2A; Padj < 0.05). tive treatments (Supplementary Fig. S1C). As CC-115 is a dual Through the analyses of both transcript proﬁles and GSEA, kinase inhibitor targeting both DNA-PK and TORK, enhanced pathways modulated by DNA-PK/TORK targeting via CC-115 effects on the transcriptome were anticipated as compared with were highly conserved in C4-2 and 22Rv1, with 3149 and 2187 DNA-PK–exclusive targeting agents. Indeed, CC-115 treatment transcript changes in common but down- and upregulated by resulted in the alteration of transcriptional networks, with CC-115, respectively. Furthermore, consistently modulated tar- 4,896 upregulated and 5,878 downregulated transcripts in gets of CC-115 in both models systems were validated in C4-2, and 3,555 upregulated and 4,358 downregulated tran- multiple CRPC models (Supplementary Fig. S2C).

Figure 1. DNA-PK regulates tumor cell proliferation in CRPC. A, Proliferation assays using DNA-PK knockdown (siDNA-PK) and pharmacologic inhibition of DNA-PK (NU7441,1 mmol/L) in CRPC cell lines using Trypan Blue (left) and Pico Green (right) assay, respectively. Cell counts were obtained at 1, 4, and 6 days post transfection for knockdown experiments, and at day 0 (prior to treatment), 4, and 6 for inhibitor experiments. Relative cell number was calculated for each day normalizing to their appropriate controls. Data are represented as mean SD of biological triplicate. Student t test statistical analyses were used where , P < 0.05; , P < 0.001 compared with control. B, Representative FACS plots of BrdU incorporation after 24-hour (hr) treatment with 1 mmol/L NU7441 and vehicle control (DMSO). Quantiﬁcation of active S-phase FACS data are presented as mean SD of biological triplicate. C, RNA-seq schematic of C4-2 cell line treated with vehicle control (DMSO) and DNA-PK inhibitor (NU7441, 1 mmol/L) in triplicate for 24 hours before RNA was harvested. MA plot was generated by comparing NU7441-treated cells to vehicle control showing gene expression modulation with the number of transcripts upregulated (top) and downregulated (bottom). GSEA using Hallmark geneset from MSigDB analysis was used to identify enriched and deenriched pathways for DNA-PKi–treated cells compared with vehicle control using FDR < 0.25. Each pathway is depicted by a ribbon in the circos plot, where blue ribbons identify pathways enriched exclusively by DNA-PKi and green ribbons identify pathways deenriched by NU7441 treatment.

www.aacrjournals.org Clin Cancer Res; 25(18) September 15, 2019 5627

Dylgjeri et al.

Moreover, the results of two GSEAs in C4-2 (CC-115 and target genes induced by the inhibition of TORK in CRPC models NU7441 treatments) were overlaid to uncover high-confidence (Fig. 3B; Supplementary Figs. S2E and S3A). In summary, the DNA-PK–exclusive pathways (Fig. 2C, left). The combined tran- combination treatments of DNA-PK inhibition with enzaluta- scriptionally altered pathways are illustrated using a circos plot mide reduce the upregulation of AR signaling in CRPC. where significantly enriched and deenriched pathways exclusive To further understand the impact of DNA-PK suppression in to NU7441 (light and dark blue ribbons, respectively), versus combination with AR targeting on the transcriptome, RNA-seq pathways exclusive to CC-115 (orange and red ribbons respec- was performed on C4-2 (Supplementary Fig. S4A) and validated tively; FDR < 0.25) are shown. As expected, the dual kinase in two other models for specific targets (Supplementary Fig. S4B). inhibitor CC-115 altered the TORK-regulated pathways of cancer Enzalutamide alone resulted in 638 upregulated and 791 down- relevance, including fatty acid metabolism, glycolysis, and apo- regulated transcripts. As expected, a larger number of transcripts ptosis pathways (34, 35). By contrast, a robust core set of path- were altered in enzalutamide þ NU7441 (2,284 upregulated, ways commonly regulated by both the single DNA-PKi (NU7441) 3,644 downregulated) and enzalutamide þ CC-115 (3,632 upre- and the dual DNA-PKi/TORi (CC-115) were identified (purple gulated, 534 downregulated) than enzalutamide alone (Fig. 3C, and green ribbons). The number of commonly deenriched path- top; Padj < 0.05). Up- and downregulated genes by 1.5-fold change ways (purple) exceeded the number of commonly enriched path- for single-agent exposure (enzalutamide alone, NU7441 alone, ways (green) by 14 and two respectively, suggesting that DNA-PK and CC-115 alone) were compared with their respective combi- primarily upregulates the transcriptional processes that play nation exposure (enzalutamide þ Nu7441 and enzalutamide þ important roles in tumor progression. Among these processes CC-115) as represented by Venn diagrams (Fig. 3C, middle). An were known DNA-PK effectors, such as DNA-repair and cell cycle/ expansion of transcriptomic alterations was observed for both proliferation, but also novel pathways where the role of DNA-PK NU7441 and CC-115 in combination with enzalutamide. GSEA transcriptional regulation is less understood, including oxidative pathway analysis was performed to identify up- and downregu- phosphorylation, epithelial–mesenchymal transition (EMT), lated pathways, represented as heatmaps, based on the gained TNFa signaling via NF-kB, IL6/Jak/Stat3 signaling, and TGFb genes exclusive to each combination treatment (Fig. 3C, bottom; signaling (Fig. 2C, left). In summary, these data show that target- FDR < 0.25). Transcriptional downregulated events associated ing DNA-PK utilizing a clinical grade inhibitor, CC-115, potently with NU7441 þ enzalutamide combination were enriched for inhibits the proliferation of CRPC cells and impacts DNA-PK– 18 pathways, whereas upregulated gene expression events were regulated transcriptional events enriched for cancer-relevant enriched for only two pathways. The downregulated gene pathways. set exhibited enrichment of known DNA-PK and AR-modulated processes such as androgen response, cell cycle, and hypoxia, Cooperative effects of cotargeting AR and DNA-PK pathways as expected. Similarly, CC-115 þ enzalutamide down- and upre- Among the common regulated pathways, GSEA identified gulated gene sets were enriched for known AR, TORK, and androgen signaling, a known driver of prostate cancer at all DNA-PK known processes. However, novel pathways including stages of disease (36). As DNA-PK is a known modulator of AR Wnt b–catenin signaling, Hedgehog signaling, inflammatory signaling (14), it was not surprising that treatment with a DNA- response, and immune response signaling were gained upon PK–specific inhibitor (NU7441) resulted in deenrichment of both combinatorial treatments, which were not previously androgen signaling (Fig. 3A, left; FDR < 0.25). In contrast, seen in the single-agent targeting approach with either kinase androgen signaling was enriched when targeting DNA-PK/ inhibitor or enzalutamide (Supplementary Fig. S4B). In sum- TORK axis using CC-115 (Fig. 3A, middle; FDR < 0.25). mary, targeting DNA-PK and AR signaling in concert leads to an Although androgen signaling downregulation was expected expanded transcriptomic profile that modulates novel protu- upon DNA-PK suppression, it has been previously described morigenic signaling pathways when compared with each treat- that AR signaling upregulation can occur upon TORK inhibi- ment alone. tion in prostate cancer (37, 38). Upregulation of androgen To uncover DNA-PK–specific effects in the context of AR signaling was associated with elevated AR protein levels in C4-2 inhibition with enzalutamide, RNA-seq was performed using (Fig. 3A, right; Supplementary Fig. S2D) and was confirmed via the TORK-specific inhibitor, CC-223, in combination with enza- RT-PCR demonstrating the induction of AR target gene expres- lutamide to mitigate the effects of targeting AR and TORK, thus sion upon CC-115 treatment (Fig. 3B, Supplementary Fig. S2E). uncovering putative DNA-PK–specific effects (Supplementary Similar upregulation of AR target gene transcripts was also seen Fig. S3B–SD). Utilizing the genes modulated by 1.5-fold change using a specific TORK inhibitor (CC-223, also in clinical trials for CC-115 þ enzalutamide and CC-223 þ enzalutamide com- NCT02031419, NCT01177397, and NCT01545947), which binations, GSEA analysis was performed (Supplementary further indicates that the inhibition of TORK is likely causing Fig. S3D). Although there was significant overlap between the AR upregulation (Supplementary Fig. S3A; ref. 39). In summa- two conditions signifying genes/pathways modulated by TORK ry, these data suggest that TORK suppression leads to upregula- and AR inhibition, CC-115 þ enzalutamide and CC-223 þ tion of AR signaling in CRPC. enzalutamide had distinct effects, which can be attributed to Because AR targeting agents are used as first-line therapy for DNA-PK and TORK, respectively. Processes that can be attributed metastatic disease (40, 41), an FDA-approved AR antagonist, to DNA-PK (CC-115 exclusive) consisted of pathways involved in enzalutamide, was used in combination with NU7441 and metabolism, inflammatory response, and protumorigenic signal- CC-115. As expected, the combination with enzalutamide further ing similar to the DNA-PK pathways previously uncovered downregulated the AR target gene transcript levels when used with in Fig. 3C. In summary, the upregulation of AR signaling upon NU7441 due to a negative feedback loop between DNA-PK and DNA-PK/TORK dual targeting with CC-115 can be mitigated AR, which has been described previously (14). The combination using enzalutamide and this combination strategy led to an of enzalutamide with CC-115 mitigated the upregulation of AR expansion of transcriptomic alterations that are distinct from

5628 Clin Cancer Res; 25(18) September 15, 2019 Clinical Cancer Research

Combined DNA-PK/TORK/AR Targeting in Prostate Cancer

CC-115 vs. Ctrl A HSPC B n = 4896 VCaP LNCaP 150 150

100 100

50 50 Percent cell viabilty cell Percent Relative survival at 6 day at survival Relative 202 4

0 0 Log fold-change 0.0001 0.001 0.01 0.1 1 10 100 0.0001 0.001 0.01 0.1 1 10 100 − Inhibitor concentration (µmol/L) Inhibitor concentration (µmol/L) 4 − CRPC n = 5878 C4-2 22Rv1 1e-01 1e+01 1e+03 1e+05 150 150 Mean of normalized counts

100 100

50 50 Relative survival at day 6 day at survival Relative Relative survival at day 6 day at survival Relative 0 0 0.0001 0.001 0.01 0.1 1 10 100 0.0001 0.001 0.01 0.1 1 10 100 Inhibitor concentration (µmol/L) Inhibitor concentration (µmol/L)

NU7441 (DNA-PKi) CC-115 (DNA-PKi/TORKi)

NU7441 (DNA-Pki) NU7441 + CC-115 CC-115 (DNA-Pki/TORKi) C Exclusively deenriched Commonly deenriched Exclusively deenriched Exclusively enriched Commonly enriched Exclusively enriched

GSEA (Hallmark Geneset) GSEA (Hallmark Geneset) Deenriched NU7441 CC-115 MTORC1 SIGNALING MYC TARGETS V1 MYC TARGETS V2 E2F TARGETS G2M CHECKPOINT OXIDATIVE PHOSPHORYLATION PI3K AKT MTOR SIGNALING DNA REPAIR ESTROGEN RESPONSE LATE IL6 JAK STAT3 SIGNALING MITOTIC SPINDLE UV RESPONSE UP EPITHELIAL MESENCHYMAL TRANSITION TNFA SIGNALING VIA NFKB

Deenrichment FDR q-Value

00.25

Enriched NU7441 CC-115 COAGULATION TGF BETA SIGNALING

Enrichment FDR q-Value

00.25

Figure 2. Targeting DNA-PK with CC-115 potently inhibits tumor cell proliferation and regulates known DNA-PK transcriptional processes. A, Dose–response proliferation assays using NU7441 (DNA-PK inhibitor) and CC-115 (dual DNA-PK/TORK inhibitor) in HSPC and CRPC cell lines using Pico Green assay at 6 days following treatment compared with vehicle control. B, RNA-seq schematic of C4-2 cell line treated with 1 mmol/L of CC-115 in triplicate for 24 hours before RNA was harvested. MA plots were generated for CC-115 treatment compared with control showing gene expression modulation with the number of transcripts upregulated (top) and downregulated (bottom). C, GSEA using Hallmark geneset from MSigDB analysis was used to identify enriched and deenriched pathways for NU7441- and CC-115–treated cells compared with control using FDR < 0.25. Each pathway is depicted by a ribbon in the circos plot (blue, pathways deenriched exclusively by NU7441; dark blue, exclusive enrichment by NU7441; orange, exclusive deenrichment by CC-115; red, enrichment by CC-115; purple, commonly deenriched; and green, commonly enriched). Commonly enriched and deenriched pathways regulated by DNA-PK are represented using heatmaps to the right.

www.aacrjournals.org Clin Cancer Res; 25(18) September 15, 2019 5629

Dylgjeri et al.

1 4 5 Androgen 4 1 A l r 7 1 Deenriched response Enriched t U - C N C (DNA-PKi NU7441) (DNA-PK/TORKi CC-115) C FDR q-Value = 0.196 AR NES = 1.39 1.00 1.09 1.63 FDR q-Value = 0.244 ± 0.09 ± 0.23 NES = -1.19 Vinculin

NU7441 Ctrl CC-115 Ctrl 123

B C4-2 KLK3 PSAmRNA TMPRSS2TMPRSS2 mRNA FKBP5FKBP5 mRNA 8 8 4

* ** * 6 6 3

4 4 2

* * 2 2 1 * ** **** ns normalized to GAPDH normalized to GAPDH normalizedGAPDH to Relative gene expression gene expression Relative Relative gene expression gene expression Relative Relative gene expression expression gene Relative 0 0 0 Enzalutamide - + - + - + Enzalutamide - + - + - + Enzalutamide - + - + - +

Ctrl NU7441 CC-115 Ctrl NU7441 CC-115 Ctrl NU7441 CC-115 Enza vs. Ctrl C 24 hr NU744+ 24 hr CC-115+ Enza n = 638 Enza C4-2 1 μmol/L each Enza C4-2 1μmol/Leach Enza 101 2 3

NU7441 + Enza vs. Ctrl − CC-115 + Enza vs. Ctrl 2 Log fold-change n = 2284 − n = n = 3632

3 791 − 1e-01 1e+01 1e+03 1e+05 Mean of normalized counts 10 1 2 3 101 2 3 − − 2 2 Log fold-change Log fold-change − n = 3644 − n 3 = 3 534 − − 1e-01 1e+01 1e+03 1e+05 1e-01 1e+01 1e+03 1e+05 Mean of normalized counts Mean of normalized counts Downregulated Upregulated Downregulated Upregulated

Enza/Ctrl Enza/Ctrl Enza/Ctrl Enza/Ctrl CC-115/Ctrl NU7441/Ctrl NU7441/Ctrl CC-115/Ctrl 78 158 257 0 0 107 1 3 17 22 500 815 19 28 331 311 128 118 6 49 192 117 2,984 2,171 168 295 1,310 749

NU7441 + Enza/Ctrl NU7441 + Enza/Ctrl CC-115 + Enza/Ctrl CC-115 + Enza/Ctrl GSEA (Hallmark Geneset) GSEA (Hallmark Geneset)

MYC TARGETS V1 TNFA SIGNALING VIANFKB IL2 STAT5 SIGNALING KRAS SIGNALING DN OXIDATIVE PHOSPHORYLATION CHOLESTEROL HOMEOSTASIS E2F TARGETS ESTROGEN RESPONSE EARLY UNFOLDED PROTEIN RESPONSE EPITHELIAL MESENCHYMAL TRANSITION G2M CHECKPOINT ADIPOGENESIS IL2 STAT5 SIGNALING INFLAMMATORY RESPONSE UV RESPONSE UP HYPOXIA NOTCH SIGNALING PROTEIN SECRETION INTERFERON GAMMA RESPONSE GLYCOLYSIS GLYCOLYSIS HEDGEHOG SIGNALING TNFA SIGNALING VIA NFKB PI3K AKT MTOR SIGNALING APICAL SURFACE MYC TARGETS V2 G2M CHECKPOINT ALLOGRAFT REJECTION COAGULATION Downregulated DNA REPAIR APICAL JUNCTION ESTROGEN RESPONSE EARLY FDR q-Value NOTCH SIGNALING UV RESPONSE DN ESTROGEN RESPONSE LATE FATTY ACID METABOLISM COMPLEMENT HYPOXIA 00.25 MYC TARGETS V2 INFLAMMATORY RESPONSE ANDROGEN RESPONSE Upregulated XENOBIOTIC METABOLISM MYOGENESIS HEDGEHOG SIGNALING FDR q-Value ANDROGEN RESPONSE XENOBIOTIC METABOLISM WNT BETA CATENIN SIGNALING TNFA SIGNALING VIA NFKB PANCREAS BETA CELLS UV RESPONSE DN 00.25 APOPTOSIS FATTY ACID METABOLISM HEME METABOLISM Downregulated Upregulated UV RESPONSE UP FDR q-Value FDR q-Value INTERFERON GAMMA RESPONSE MITOTIC SPINDLE 00.2500.25

5630 Clin Cancer Res; 25(18) September 15, 2019 Clinical Cancer Research

Combined DNA-PK/TORK/AR Targeting in Prostate Cancer

those observed with single-agent targeting. Furthermore, novel respectively (Fig. 4D). Tumor volume was monitored throughout putative DNA-PK–exclusive transcriptional regulatory events the duration of the study (39 days total, 19 days post initial were identified, further implicating DNA-PK as a transcriptional treatment). At the end of the study, CC-115 alone showed regulator of cancer-relevant pathways. marginal activity leading to approximately 22% decrease in tumor volume, enzalutamide alone showed dose-dependent activity Feasibility of cotargeting DNA-PK/TORK and AR axis in vivo (35% and 48% tumor volume decrease for 5 mg/kg and 10 To investigate whether the combination of DNA-PK–targeting mg/kg, respectively), and the combination treatments of CC- agents with enzalutamide demonstrate superior efficacy in 115 þ low- and high-dose enzalutamide showed additive effects comparison with CC-115 and enzalutamide alone, combina- with a 61% and 62% decrease in tumor volume with the com- tion index analyses were performed after treatments with bination at high enzalutamide dose demonstrated a 29% decrease CC-115 at IC25,IC50,orIC75 concentrations combined with of tumor volume as compared with enzalutamide (10 mg/kg) and varied doses of enzalutamide in both HSPC and CRPC models. 64% decrease when compared with vehicle (Fig. 4E). Moreover, Combination of CC-115 (IC25,IC50,orIC75) when combined similar studies performed in another CRPC model, LNCaP–AR, with enzalutamide (all doses) showed synergism in both HSPC recapitulate the results presented herein where the combination and CRPC models (Fig. 4A). To extend these observations into a treatment leads to greater tumor growth inhibition compared preclinical model of CRPC, the combination of CC-115 þ with enzalutamide or CC-115 alone (unpublished data). In enzalutamide was tested in a xenograft model generated as summary, these data demonstrate that targeting the DNA-PK/ showninFig.4Bontheleft.Followingtumorengraftment, TORK and AR with CC-115 in combination with enzalutamide mice were treated with vehicle, CC-115 alone (2 mg/kg), results in decreased tumor proliferation both in vitro and in vivo enzalutamide alone (low and high dose, 5 mg/kg and when compared with control or either agent treatment alone. 10 mg/kg, respectively), and the combination of CC-115 þ enzalutamide at low and high enzalutamide doses (2 mg/kg Dual DNA-PK/TORK inhibition and cotargeting AR elicits CC-115 þ 5 mg/kg enzalutamide and 2 mg/kg CC-115 þ 10 cooperative antitumor effects in human prostate cancer mg/kg enzalutamide). In addition, mRNA expression of the AR explants target gene, FKBP5, was interrogated to validate the efficacy of Because targeting of DNA-PK-TOR-AR axis showed synergis- targeting the AR axis in this model. As expected, FKBP5 mRNA, tic and additive antiproliferative effects in vitro and in vivo as well as TORK and DNA-PK activity (Supplementary Fig. S5A respectively, the impact of this combination treatment was and S5B), were elevated upon treatment with CC-115 alone, studied in a PDE model using human tumor samples from but were attenuated by the combination of CC-115 þ enzalu- high-volume disease (Supplementary Table S2) that were tamide (low and high dose), thus corroborating the in vitro obtained after radical prostatectomy as described previous- finding as shown in Fig. 3B (Fig. 4B, right). Moreover, CC-115 ly (28) and summarized in Fig. 5A. Importantly, these tissues and CC-115 þ enzalutamide were shown to reduce DNA-PK retain histoarchitecture, AR expression, proliferation rate, and activity and TORK target expression in an in vivo HSPC model the microenvironment features of the original tumor (27, 28, study treated with CC-115 alone, enzalutamide, and CC-115 þ 42). Upon resection, the tumor samples were subdivided and enzalutamide (unpublished data). In sum, these data recapit- treated with vehicle control, CC-115 (0.1, 0.5, and 1 mmol/L) ulate androgen signaling upregulation upon DNA-PKi/TORKi alone, enzalutamide (1 mmol/L) alone, and combination of CC-115 inhibition observed in vitro and mitigation of this effect with (0.1, 0.5, and 1 mmol/L) þ enzalutamide (1 mmol/L). The levels of the enzalutamide þ CC-115 combination in vivo. Ki67, an indicator of tumor cell proliferation, were measured using Tumor doubling time was increased upon treatment with IHC. Seventy-five percent of PDEs (6/8) responded to enzalutamide enzalutamide alone and CC-115 þ enzalutamide but not CC-115 treatment (lower Ki67 positivity) as compared with control. CC-115 alone; however, combination of CC-115 þ enzalutamide (low at high dose (1 mmol/L) showed strong antiproliferative effects when and high dose) led to significantly increased tumor doubling time used as a single agent. The antiproliferative effects were enhanced compared with CC-115 alone, demonstrating that the combina- upon combining CC-115 with enzalutamide especially with 0.1 tion treatment is more effective than single-agent treatment in vivo and 0.5 mmol/LCC-115(PDEs#1,3,7,8).Whentheconcentration (Fig. 4C). Similarly, combination treatments led to higher survival of CC-115 was increased to 1 mmol/L, robust antiproliferative (using a tumor volume of 1,500 mm3 as an endpoint) compared effects were observed but no further benefit was seen with the with either single treatment alone, with CC-115 þ enzalutamide addition of enzalutamide (1 mmol/L) (PDEs #2, 4, 5, 6; Fig. 5C). low and high dose having a 75% and 87.5% survival rate, These data demonstrate that targeting DNA-PK/TORK alone and

Figure 3. Dual targeting of DNA-PK and TORKs in combination with enzalutamide leads to distinct downstream transcriptional alterations compared with single-agent targeting. A, The Androgen Response Hallmark is oppositely regulated by targeting DNA-PK (NU7441) and DNA-PK/TOR (CC-115). GSEA enrichment plots for androgen response upon NU7441 and CC-115 are shown. Immunoblot analyses of AR expression in C4-2 cells is depicted to the right, with quantification of AR in three independent experiments represented as average as mean SD. AR is significantly upregulated upon DNA-PK/TOR targeting (P < 0.01). B, Interrogation of AR target gene expression via qPCR using single and combination drug treatment (1 mmol/L) with enzalutamide (1 mmol/L) after 24 hours before RNA was harvested. Data represented as mean SD of biological triplicates. Student t test statistical analyses were used where , P < 0.05; , P < 0.01; , P < 0.001; , P < 0.0001 compared with single-agent treatments. ns, not significant. C, RNA-seq schematic of C42 cell line treated with enzalutamide (1 mmol/L) and NU7441 (1 mmol/L) þ enzalutamide (1 mmol/L) and enzalutamide (1 mmol/L) and CC-115 (1 mmol/L) þ enzalutamide (1 mmol/L) for 24 hours (hr) before RNA was harvested. MA plots were generated for enzalutamide alone and NU7441 þ enzalutamide combination compared with control and CC-115 þ enzalutamide combination compared with control. Venn diagrams were generated using genes up- or downregulated by at least 1.5-fold change and Padj < 0.05. Genes that were gained in the combination treatment were utilized for GSEA hallmark geneset analysis. Pathways shown in the heatmaps above were obtained using FDR < 0.25. Enza, enzalutamide.

www.aacrjournals.org Clin Cancer Res; 25(18) September 15, 2019 5631

Dylgjeri et al.

VCaP LNCaP A LNCaP 2 VCaP 2

Antagonism Antagonism Addive IC25 CC-115 eﬀect Addive Synergism eﬀect Synergism IC50 CC-115

Combinaon index IC75 CC-115 Combinaon index

0 0 0 50 100 0 100 50 % Reduced VCaP proliferaon % Reduced LNCaP proliferaon

C4-2 C4-2 22Rv122Rv1 LN95LN95 2 2 2

Antagonism Antagonism Antagonism Addive Addive Addive effect Synergism effect effect Synergism Synergism Combinaon index Combinaon index Combinaon index

0 0 0 0 50 100 0 50 100 0 50 100 % Reduced C4-2 proliferaon %Reduced 22Rv1 proliferaon %ReducedLN95proliferaon

B C Tumor regrowth Surgical Hormone refractory Develop tumor Monitor tumor castration tumors (CRPC) xenografts growth ~200 mm3 Transplant to +/- Treatment ** castrated 15 LNCaP male SCIDs 2-5X ** FKBP5 mRNA *** 12 **** * * 2.0 * 9 *

1.5

6 1.0 *

3 normalized to HuPO 0.5 Relative gene expression Average tumor doubling time (days)

0.0 0 Vehicle CC-115 Enza Enza CC-115 CC-115 (2 mg/kg) (5 mg/kg) (10 mg/kg) (2 mg/kg) (2 mg/kg) Vehicle CC-115 Enza Enza CC-115 CC-115 ++ (2 mg/kg) (5 mg/kg) (10 mg/kg) (2 mg/kg) (2 mg/kg) Enza Enza (5 mg/kg) (10 mg/kg) ++ Enza Enza (5 mg/kg) (10 mg/kg) D E 12 Control 100 CC-115 2 mg/kg )

3 Enza 5 mg/kg 9 Enza 10 mg/kg Control Enza 5 mg/kg + CC-115 2 mg/kg CC-115 2 mg/kg * Enza 10 mg/kg + CC-115 2 mg/kg ** *** Enza 5 mg/kg 6 *** -64% 50 Enza 10 mg/kg -29% Enza 5 mg/kg + CC-115 2 mg/kg Percent survival Relative tumor volume volume tumor Relative 3 Enza 10 mg/kg + CC-115 2 mg/kg (compared to start of treatment) of start to (compared (tumor volume 1,500 mm 1,500 volume (tumor

0 0 0 5 10 15 20 25 05101520 Days post treatment Days post treatment

Figure 4. Cotargeting DNA-PK/ TORK and AR axis has additive growth inhibitory effects in vivo. A, Combination index determination in prostate cancer cell lines when

using CC-115 at IC25,IC50,andIC75 in combination with varied concentrations of enzalutamide. Experiments were performed in biological triplicates. B, Schematic of CRPC mouse model developed by Celgene. Upon establishment of CRPC tumors, the AR axis was interrogated by measuring FKBP5 mRNA expression levels in tumors after treatments via qPCR (right). Data represented as mean SEM of biological triplicates. Student t test statistical analyses were used where , P < 0.05 compared with control (red) or as otherwise indicated by the brackets. C, Average tumor doubling time was calculated for each treatment cohort. D, The percentage of tumors reaching 1,500 mm3 was calculated for each cohort. E, Tumor growth was monitored for 19 days after single-agent or combination drug treatment. Relative tumor volume is shown for each treatment normalized to tumor volume at the start of treatment. Mouse data are presented as mean SEM, and one-way ANOVA was used for statistical analysis where , P < 0.05; , P < 0.01; , P < 0.001; , P < 0.0001 compared with control. Enza, enzalutamide.

5632 Clin Cancer Res; 25(18) September 15, 2019 Clinical Cancer Research

Combined DNA-PK/TORK/AR Targeting in Prostate Cancer

in combination with enzalutamide has antiproliferative effects these previous findings, unbiased profiling of transcriptional in primary prostate human tumors, and provide the impetus for networks sensitive to DNA-PK inhibition further supports the clinical evaluation of DNA-PK–targeting agents in combination concept that DNA-PK positively regulates gene networks with enzalutamide. involved in DNA repair and hormone signaling as observed Based, in part, on the findings presented here, a first-in-man in Figs. 1C and 2C. Previously published data have also linked phase IB/II clinical trial (NCT02833883) is being conducted DNA-PK to play a prometastatic role in cancer (12, 21). DNA- in prostate cancer, with endpoints assessing the safety and PK promotes metastasis, in part, through transcriptional mod- pharmacokinetics of escalating doses of CC-115 in combina- ulation of prometastatic gene networks in the Rac-Rho pathway tion with enzalutamide (160 mg twice a day) to establish the in prostate cancer (12). Another study in melanoma has shown MTD. These data will be used in a phase II trial to assess that DNA-PK modifies the tumor microenvironment by mod- endpoints of safety, biochemical recurrence by looking at PSA ulating the secretion of promigratory molecules and promotes levels, and progression-free survival (Fig. 5D). Combined the metastasis (21). Data herein identified EMT and TNFa via NF- studies presented herein have further defined the transcription- kB pathways as also responsive to DNA-PK function. These al-regulatory roles of DNA-PK, and have identified the process- findings are impactful as EMT is an important step toward es that are affected by combinatorial targeting strategies of development of metastases, and TNFa plays a critical role in DNA-PKi/TORKi/enzalutamide (Fig. 5E). Furthermore, the tumor microenvironment and EMT plasticity. Previous studies combination of DNA-PKi/TORi and enzalutamide has dem- have described DNA-PK to directly interact with the EMT onstrated to have superior anticancer effects to single-agent protein Snail and drive metastasis (33); however, it is of high treatment in vitro, in vivo,andex vivo. These studies serve as the importance to understand how DNA-PK modulates EMT/TNFa rationale for clinical investigation of the combination of DNA- gene networks transcriptionally. Understanding the mechan- PK/TOR-targeting agents and enzalutamide in the manage- isms by which DNA-PK transcriptionally modulates prometa- ment of CRPC. static signaling would allow for the development of new therapeutics to target these processes with the goal of suppressing metastasis. Discussion In addition to regulation of metastatic pathways, the data DNA-PK is a multifunctional kinase that is deregulated in presented here uncovered a novel DNA-PK–mediated influence multiple human malignancies. In aggressive prostate cancer, on immune and inflammation response pathways including: DNA-PK is strongly associated with poor outcome (12). IFNa response, IFNg response, inflammatory response, and Despite being studied for its role in DNA repair and transcrip- TNFa pathway. These data complement previous findings that tional regulation, much remains to be uncovered about the identified DNA-PK as a modulator of V(D)J recombination, mechanisms by which DNA-PK promotes cancer phenotypes. activator of the innate immune response, and subsequent inflam- The study presented here identifies DNA-PK as a transcriptional matory response in the presence of foreign DNA and patho- regulator of multiple known and novel cancer-relevant path- gens (46–49). Uncovering the crucial impact of DNA-PK on ways in the absence of exogenous DNA damage in prostate both the innate and adaptive immune response is the focus of cancer. Furthermore, data herein demonstrate that DNA-PK ongoing investigations. Moreover, understanding how the innate targetingusingaclinicalgrade inhibitor, in concert with a and immune responses are modulated in patients treated with standard-of-care AR antagonist, has cooperative antitumor DNA-PK–directed therapeutics will likely give insight into DNA- effects in prostate cancer. Key findings reveal that: (i) DNA-PK PK–dependent immune-related mechanisms of response and/or regulates tumor cell proliferation; (ii) pharmacologic targeting resistance. A recent study identified DNA-PK inhibition as a of DNA-PK suppresses tumor growth both in vitro, in vivo, and therapeutic approach that modulates immunity in melanoma ex vivo; (iii) DNA-PK transcriptionally regulates known and may increase the efficacy of immunotherapies (50), thus it DNA-PK–mediated functions as well as novel cancer-related would be of interest to study immunotherapy and DNA-PK pathways that promote tumor growth; (iv) dual targeting of inhibitors in prostate cancer tumors that overexpress DNA-PK DNA-PK/TORK transcriptionally upregulates androgen signal- and assess their efficacy as novel therapeutic strategies. In addition ing, which can be mitigated using the AR antagonist, enzalu- to transcriptional regulation of immune response pathways, tamide; (v) cotargeting AR and DNA-PK/TORK leads to the data herein suggest that DNA-PK modulates cancer cell metabo- expansion of antitumor effects, uncovering modulation of lism including fatty acid metabolism, cholesterol homeostasis, novel, highly relevant protumorigenic cancer pathways; and and oxidative phosphorylation (Figs. 1C and 2C; Supplementary (viii) cotargeting DNA-PK/TORK and AR has cooperative Fig. S3D). Although these findings support previously published growth-inhibitory effects in vitro and in vivo. In sum, this study data describing the role of DNA-PK in lipogenesis and mitochon- uncovered multiple novel cancer-relevant processes that are drial biogenesis and function (51–53), this study is the first to transcriptionally modulated by DNA-PK in models of advanced describe DNA-PK directly affecting the transcriptional regulation disease, and demonstrated that DNA-PK/TORK can be effec- of metabolic gene networks rather than modulating metabolic tively targeted in combination with enzalutamide to prevent processes through phosphorylation of cofactors or protein–pro- tumor growth in prostate cancer in the absence of exogenous tein interactions. These findings support the rationale to assess the DNA damage. This study provides strong rationale for using critical impact of DNA-PK in tumor-associated metabolic and this three-pronged attack by targeting DNA-PK, TORK, and AR immunomodulating processes. axes in the management of lethal CRPC. On the basis of the protumorigenic role of DNA-PK in cancer DNA-PK has been previously shown to modulate cancer and the data linking it to metastatic potential, DNA-PK has been phenotypes through DNA repair via NHEJ and transcriptional nominated as a therapeutic target. Recently developed DNA-PK regulatory mechanisms (8, 12, 14, 43–45). Consistent with inhibitors have entered the clinical trial space in combination

www.aacrjournals.org Clin Cancer Res; 25(18) September 15, 2019 5633

Dylgjeri et al.

A Ki67 Day 0 Nonneoplas cTumor

Prostatectomy Subdivide IHC High volume and culture +/- disease as explant Treatment

B Day 6 Control CC-115 C 60 PDE #1 PDE #2 40 PDE #3 PDE #4 Control PDE #5 PDE #6 20 PDE #7 PDE #8 40X %Ki67 Positivity

l tr C mol/L) mol/L) mol/L) mol/L) mol/L) mol/L) µ µ µ µ µ µ Enza (1 .1 .5 (1 (1 (1 a (0 (0 5 a a (1 µmol/L) z 1 z za z n 5 5 1 n n n E 1 1 - E E E -1 -1 C + C C C + + C C

mol/L) mol/L) mol/L) µ µ µ 1 .1 .5 ( 0 0 5 ( ( 1 5 5 1 1 1 - -1 -1 C C C C C C D Clinical Trial : NCT02833883 Phase IB Phase II

CC-115 + Enza Dose Escalaon CC-115 + Enza Expansion CC-115 + Enza Expansion n = 12 Interim n = 16 CC-115 Phase IB Outcomes: Phase II Outcomes:

5mg 10 mg 20 mg CC-115 analysis CC-115 BID BID BID RP2D/MTD RP2D/MTD n = 3-6 n = 3-6 n = 3-6

Enzalutamide 160 mg BID Enzalutamide 160 mg QD Enzalutamide 160 mg QD

NU441 CC-115 Enzalutamide + CC-115

DNA-PK DNA-PK DNA-PK Proliferaon

AR In vitro Ex-vivo AR

• Androgen signaling Putative DNA-PK targets • DNA Repair • Upregulated AR signaling • Downregulated AR signaling • Cell cycle/proliferation • DNA Repair • DNA Repair • Cell cycle/proliferation • Cell cycle/proliferation Survival • Metabolism • Metabolism • Inflammatory response • Inflammatory response • Immune response In vivo • Prometastatic networks • Apoptosis

Figure 5. Dual DNA-PK/TORK inhibition and cotargeting AR elicits cooperative antitumor effects in human prostate cancer. A, Schematic of human PDE model generation from human prostatectomy samples that are treated with single-agent CC-115 (0.1, 0.5, and 1 mmol/L DNA-PK/TOR inhibitor) and enzalutamide (1 mmol/L) and combination treatments (0.1, 0.5, and 1 mmol/L CC-115 þ 1 mmol/L enzalutamide) for 6 days. B, Representative Ki67 IHC tissue staining after each treatment is shown at 40 magnification. Enza, enzalutamide. C, Each PDE sample (each patient is represented by a different color) was scored by by Digital Imaging Analysis (Aperio System) and confirmed by a pathologist. Data is represented as Ki67 mean SD of each cohort of patient tissues treated with the same treatment. D, Schematic of the CC-115 and enzalutamide clinical trial currently conducted in patients with CRPC. BID, twice a day; Enza, enzalutamide; QD, every day; RP2D, recommended phase II dose. E, Model summarizing findings in this article.

5634 Clin Cancer Res; 25(18) September 15, 2019 Clinical Cancer Research

Combined DNA-PK/TORK/AR Targeting in Prostate Cancer

with irradiation in multiple liquid and solid tumors. In the study identified, and was fast-tracked into a clinical study. CRPC presented here, two DNA-PK inhibitors were used: a specific remains universally lethal disease; however cotargeting DNA-PK, laboratory-grade DNA-PK inhibitor (NU7441) and a clinical- TORK, and AR could provide an effective anticancer therapeutic grade dual DNA-PK/TORK inhibitor (CC-115). The advantage strategy for the management of CRPC. of using the dual DNA-PK inhibitor is multi-fold: (i) CC-115 inhibits not only DNA-PK but also TORK, which is commonly deregulated in cancer and is involved in cancer metabolism, Disclosure of Potential Conflicts of Interest tumor microenvironment, proliferation, and metastasis; (ii) H.K. Raymon has ownership interests (including patents) in Celgene. TORK targeting is already under investigation in multiple cancers B.E. Leiby is a consultant/advisory board member for Bayer Healthcare. including prostate, and already FDA approved in renal cell car- L.G. Gomella is a consultant/advisory board member for Janssen, Astellas, cinoma (54, 55), however, patients with CRPC have had negli- Bayer, Merck, and Celgene. F.Y. Feng is a consultant/advisory board member for gible results, suggesting that targeting DNA-PK in combination Janssen, Sanofi, Astellas, Bayer, Dendreon, Ferring, Celgene, EMD Serono, fl with TORK may lead to better results in CRPC (38, 56); (iii) CC- Re exion, Clovis, and Blue Earth Diagnostics. E.H. Filvaroff has ownership interests (including patents) in Celgene. K. Hege has ownership interests 115 can potently inhibit proliferation as a single agent in models (including patents) in Celgene. D. Rathkopf is a consultant/advisory board of HSPC and CRPC to a greater magnitude than either DNA-PK or member for Janssen. K.E. Knudson reports receiving commercial research TORK inhibition alone; and (iv) CC-115 is under current clinical support from Celgene. No potential conflicts of interest were disclosed by the investigation. One of the challenges that was anticipated and other authors. observed in this study with the use of two TORK-targeting agents was the upregulation of the androgen response due to a feedback Authors' Contributions regulatory mechanism between TOR and AR (37, 38). Utilization Conception and design: E. Dylgjeri, C. McNair, J.F. Goodwin, A.A. Shafi, of TORK inhibitors as single agents has demonstrated minimal J.J. McCann, A.C. Mandigo, K. Hege, K.E. Knudsen efficacy in clinical trials in CRPC, which is mainly attributed to the Development of methodology: E. Dylgjeri, C. McNair, J.F. Goodwin, L.J. Brand, feedback response with AR (57, 58). Combinations of TORK- K.E. Knudsen targeting agents with a standard-of-care androgen antagonist, Acquisition of data (provided animals, acquired and managed patients, provided facilities, etc.): E. Dylgjeri, H.K. Raymon, P.A. McCue, A.A. Shafi, enzalutamide, are being investigated in clinical trials in CRPC R. de Leeuw, S.N. Chand, N. Gordon, E.J. Trabulsi, D. Rathkopf (NCT02125084 and NCT02407054). The data herein show that Analysis and interpretation of data (e.g., statistical analysis, biostatistics, DNA-PKi/TORKi/ enzalutamide combination performed better computational analysis): E. Dylgjeri, C. McNair, H.K. Raymon, B.E. Leiby, than enzalutamide alone and DNA-PK/TORKi alone in inhibiting J.J. McCann, S.N. Chand, I. Vasilevskaya, N. Gordon, E.J. Trabulsi, F.Y. Feng, proliferation in prostate cancer models in vitro, in vivo and ex vivo. E.H. Filvaroff, K.E. Knudsen Thus, cotargeting of DNA-PK, TORK, and AR would be a multi- Writing, review, and/or revision of the manuscript: E. Dylgjeri, C. McNair, J.F. Goodwin, H.K. Raymon, P.A. McCue, A.A. Shafi, B.E. Leiby, R. de Leeuw, factorial treatment strategy that impresses upon multiple protu- V. Kothari, J.J. McCann, A.C. Mandigo, S.N. Chand, M.J. Schiewer, L.J. Brand, morigenic pathways, with the potential to improve therapeutic I. Vasilevskaya, T.S. Laufer, L.G. Gomella, C.D. Lallas, E.J. Trabulsi, F.Y. Feng, response. Despite the likely benefits of the DNA-PK/TORK/AR E.H. Filvaroff, K. Hege, D. Rathkopf, K.E. Knudsen combination treatment, there are potential concerns that can be Administrative, technical, or material support (i.e., reporting or organizing hypothesized, such as increased toxicity, drug–drug interaction data, constructing databases): E. Dylgjeri, N. Gordon, T.S. Laufer, E.H. Filvaroff effects, and development of resistance mechanisms. On the basis Study supervision: K.E. Knudsen of the findings presented here, a first-in-man DNA-PK–targeted therapeutic clinical trial was opened to study the combination Acknowledgments treatment of the dual inhibitor CC-115 with enzalutamide in men The authors gratefully thank all the members of the Knudsen laboratory for their intellectual and technical support. Additionally, the authors thank the with CRPC (NCT02833883). Clinical assessment of CC-115/ following institutions that supported this work: the NIH/NCI grants to enzalutamide combination will shed more light on whether the K.E. Knudsen (RO1 CA176401, R01 CA182569, R01 CA217329, and P30 toxicities are manageable and whether they outweigh the benefits. CA056036), the Prostate Cancer Foundation (to K.E. Knudsen, A.A. Shafi, and In summary, DNA-PK is known to be overexpressed, hyper- R. de Leeuw), the Sidney Kimmel Cancer Center (SKCC) Support Grant activated, and a driver of aggressive phenotypes in advanced (5P30CA056036), and the Translational Pathology, MetaOmics, and Biosta- prostate cancer (12), however the underpinning mechanisms are tistics shared resource core facilities at SKCC. not well understood. Data presented herein defined the global The costs of publication of this article were defrayed in part by the payment of transcriptomic functions of DNA-PK in the absence of exogenous page charges. This article must therefore be hereby marked advertisement in DNA damage, and uncovered novel processes that are modulated accordance with 18 U.S.C. Section 1734 solely to indicate this fact. by DNA-PK including transcriptional regulation of EMT, immune response, and metabolism. Moreover, a combination therapy Received July 11, 2018; revised August 28, 2018; accepted March 5, 2019; targeting multiple critical nodes involved in prostate cancer was published first July 2, 2019.

References 1. O'Connor MJ. Targeting the DNA damage response in cancer. Mol Cell 4. Ceccaldi R, Rondinelli B, D'Andrea AD. Repair pathway choices and 2015;60:547–60. consequences at the double-strand break. Trends Cell Biol 2016;26: 2. Jackson SP, Bartek J. The DNA-damage response in human biology and 52–64. disease. Nature 2009;461:1071–8. 5. Radhakrishnan SK, Jette N, Lees-Miller SP. Non-homologous end joining: 3. Moynahan ME, Jasin M. Mitotic homologous recombination maintains emerging themes and unanswered questions. DNA Repair 2014;17:2–8. genomic stability and suppresses tumorigenesis. Nat Rev Mol Cell Biol 6. Davis AJ, Chen DJ. DNA double strand break repair via non-homologous 2010;11:196–207. end-joining. Transl Cancer Res 2013;2:130–43.

www.aacrjournals.org Clin Cancer Res; 25(18) September 15, 2019 5635

Dylgjeri et al.

7. Jeggo PA, Pearl LH, Carr AM. DNA repair, genome stability and cancer: a 29. Dobin A, Davis CA, Schlesinger F, Drenkow J, Zaleski C, Jha S, et al. historical perspective. Nat Rev Cancer 2016;16:35–42. STAR: ultrafast universal RNA-seq aligner. Bioinformatics 2013;29: 8. Goodwin JF, Knudsen KE. Beyond DNA repair: DNA-PK function in cancer. 15–21. Cancer Discov 2014;4:1126–39. 30. Robinson MD, McCarthy DJ, Smyth GK. edgeR: a Bioconductor package for 9. Beskow C, Skikuniene J, Holgersson A, Nilsson B, Lewensohn R, differential expression analysis of digital gene expression data. Bioinfor- Kanter L, et al. Radioresistant cervical cancer shows upregulation of matics 2010;26:139–40. the NHEJ proteins DNA-PKcs, Ku70 and Ku86. Br J Cancer 2009;101: 31. Subramanian A, Tamayo P, Mootha VK, Mukherjee S, Ebert BL, Gillette MA, 816–21. et al. Gene set enrichment analysis: a knowledge-based approach for 10. Soderlund Leifler K, Queseth S, Fornander T, Askmalm MS. Low expres- interpreting genome-wide expression profiles. Proc Natl Acad Sci U S A sion of Ku70/80, but high expression of DNA-PKcs, predict good 2005;102:15545–50. response to radiotherapy in early breast cancer. Int J Oncol 2010;37: 32. Krzywinski M, Schein J, Birol I, Connors J, Gascoyne R, Horsman D, et al. 1547–54. Circos: an information aesthetic for comparative genomics. Genome Res 11. Willmore E, Elliott SL, Mainou-Fowler T, Summerfield GP, Jackson GH, 2009;19:1639–45. O'Neill F, et al. DNA-dependent protein kinase is a therapeutic target and 33. Pyun BJ, Seo HR, Lee HJ, Jin YB, Kim EJ, Kim NH, et al. Mutual regulation an indicator of poor prognosis in B-cell chronic lymphocytic leukemia. between DNA-PKcs and Snail1 leads to increased genomic instability and Clin Cancer Res 2008;14:3984–92. aggressive tumor characteristics. Cell Death Dis 2013;4:e517. 12. Goodwin JF, Kothari V, Drake JM, Zhao S, Dylgjeri E, Dean JL, et al. DNA- 34. Loewith R, Hall MN. Target of rapamycin (TOR) in nutrient signaling and PKcs-mediated transcriptional regulation drives prostate cancer progres- growth control. Genetics 2011;189:1177–201. sion and metastasis. Cancer Cell 2015;28:97–113. 35. Laplante M, Sabatini DM. mTOR signaling at a glance. J Cell Sci 2009;122: 13. Hsu FM, Zhang S, Chen BP. Role of DNA-dependent protein kinase 3589–94. catalytic subunit in cancer development and treatment. Transl Cancer Res 36. Knudsen KE, Penning TM. Partners in crime: deregulation of AR activity 2012;1:22–34. and androgen synthesis in prostate cancer. Trends Endocrinol Metab 2010; 14. Goodwin JF, Schiewer MJ, Dean JL, Schrecengost RS, de Leeuw R, Han S, 21:315–24. et al. A hormone-DNA repair circuit governs the response to genotoxic 37. Carver BS, Chapinski C, Wongvipat J, Hieronymus H, Chen Y, Chan- insult. Cancer Discov 2013;3:1254–71. darlapaty S, et al. Reciprocal feedback regulation of PI3K and androgen 15. O'Connor MJ, Martin NM, Smith GC. Targeted cancer therapies based receptor signaling in PTEN-deficient prostate cancer. Cancer Cell 2011; on the inhibition of DNA strand break repair. Oncogene 2007;26: 19:575–86. 7816–24. 38. Bitting RL, Armstrong AJ. Targeting the PI3K/Akt/mTOR pathway in 16. Shinohara ET, Geng L, Tan J, Chen H, Shir Y, Edwards E, et al. DNA- castration-resistant prostate cancer. Endocr Relat Cancer 2013;20: dependent protein kinase is a molecular target for the development R83–99. of noncytotoxic radiation-sensitizing drugs. Cancer Res 2005;65: 39. Mortensen DS, Fultz KE, Xu S, Xu W, Packard G, Khambatta G, et al. CC- 4987–92. 223, a potent and selective inhibitor of mTOR kinase: in vitro and in vivo 17. Lee KJ, Lin YF, Chou HY, Yajima H, Fattah KR, Lee SC, et al. Involvement of characterization. Mol Cancer Ther 2015;14:1295–305. DNA-dependent protein kinase in normal cell cycle progression through 40. Scher HI, Fizazi K, Saad F, Taplin ME, Sternberg CN, Miller K, et al. mitosis. J Biol Chem 2011;286:12796–802. Increased survival with enzalutamide in prostate cancer after chemother- 18. Kong X, Shen Y, Jiang N, Fei X, Mi J. Emerging roles of DNA-PK besides apy. N Engl J Med 2012;367:1187–97. DNA repair. Cell Signal 2011;23:1273–80. 41. Merseburger AS, Haas GP, von Klot CA. An update on enzalutamide in the 19. Ruis BL, Fattah KR, Hendrickson EA. The catalytic subunit of DNA- treatment of prostate cancer. Ther Adv Urol 2015;7:9–21. dependent protein kinase regulates proliferation, telomere length, and 42. Schiewer MJ, Goodwin JF, Han S, Brenner JC, Augello MA, Dean JL, et al. genomic stability in human somatic cells. Mol Cell Biol 2008;28: Dual roles of PARP-1 promote cancer growth and progression. 6182–95. Cancer Discov 2012;2:1134–49. 20. Stronach EA, Chen M, Maginn EN, Agarwal R, Mills GB, Wasan H, et al. 43. Medunjanin S, Weinert S, Schmeisser A, Mayer D, Braun-Dullaeus RC. DNA-PK mediates AKT activation and apoptosis inhibition in clinically Interaction of the double-strand break repair kinase DNA-PK and estrogen acquired platinum resistance. Neoplasia 2011;13:1069–80. receptor-alpha. Mol Biol Cell 2010;21:1620–8. 21. Kotula E, Berthault N, Agrario C, Lienafa MC, Simon A, Dingli F, et al. 44. Trevino LS, Bolt MJ, Grimm SL, Edwards DP, Mancini MA, DNA-PKcs plays role in cancer metastasis through regulation of secreted Weigel NL. Differential regulation of progesterone receptor-medi- proteins involved in migration and invasion. Cell Cycle 2015;14: ated transcription by CDK2 and DNA-PK. Mol Endocrinol 2016; 1961–72. 30:158–72. 22. Davidson D, Amrein L, Panasci L, Aloyz R. Small molecules, inhibitors 45. Hill R, Lee PW. The DNA-dependent protein kinase (DNA-PK): more than of DNA-PK, targeting DNA repair, and beyond. Front Pharmacol 2013; just a case of making ends meet? Cell Cycle 2010;9:3460–9. 4:5. 46. Taccioli GE, Amatucci AG, Beamish HJ, Gell D, Xiang XH, Torres Arzayus 23. Zhao Y, Thomas HD, Batey MA, Cowell IG, Richardson CJ, Griffin RJ, et al. MI, et al. Targeted disruption of the catalytic subunit of the DNA-PK gene in Preclinical evaluation of a potent novel DNA-dependent protein kinase mice confers severe combined immunodeficiency and radiosensitivity. inhibitor NU7441. Cancer Res 2006;66:5354–62. Immunity 1998;9:355–66. 24. Tsuji T, Sapinoso LM, Tran T, Gaffney B, Wong L, Sankar S, et al. CC-115, a 47. Ju J, Naura AS, Errami Y, Zerfaoui M, Kim H, Kim JG, et al. Phosphorylation dual inhibitor of mTOR kinase and DNA-PK, blocks DNA damage of p50 NF-kappaB at a single serine residue by DNA-dependent protein repair pathways and selectively inhibits ATM-deficient cell growth in vitro. kinase is critical for VCAM-1 expression upon TNF treatment. J Biol Chem Oncotarget 2017;8:74688–702. 2010;285:41152–60. 25. Mortensen DS, Perrin-Ninkovic SM, Shevlin G, Elsner J, Zhao J, Whitefield 48.FergusonBJ,MansurDS,PetersNE,RenH,SmithGL.DNA-PKisa B, et al. Optimization of a series of triazole containing mammalian target of DNA sensor for IRF-3-dependent innate immunity. Elife 2012;1: rapamycin (mTOR) kinase inhibitors and the discovery of CC-115. J Med e00047. Chem 2015;58:5599–608. 49. Wang Y, Sun H, Wang J, Wang H, Meng L, Xu C, et al. DNA-PK-mediated 26. Thijssen R, Ter Burg J, Garrick B, van Bochove GG, Brown JR, Fernandes SM, phosphorylation of EZH2 regulates the DNA damage-induced apoptosis to et al. Dual TORK/DNA-PK inhibition blocks critical signaling pathways in maintain T-cell genomic integrity. Cell Death Dis 2016;7:e2316. chronic lymphocytic leukemia. Blood 2016;128:574–83. 50. Tsai AK, Khan AY, Worgo CE, Wang LL, Liang Y, Davila E. A multi- 27. Centenera MM, Raj GV, Knudsen KE, Tilley WD, Butler LM. Ex vivo culture kinase and DNA-PK inhibitor combination immunomodulates mel- of human prostate tissue and drug development. Nat Rev Urol 2013;10: anomas, suppresses tumor progression, and enhances immunothera- 483–7. pies. Cancer Immunol Res 2017;5:790–803. 28. Shafi AA, Schiewer MJ, de Leeuw R, Dylgjeri E, McCue PA, Shah N, et al. 51. Wong RH, Chang I, Hudak CS, Hyun S, Kwan HY, Sul HS. A role of DNA-PK Patient-derived models reveal impact of the tumor microenvironment on for the metabolic gene regulation in response to insulin. Cell 2009;136: therapeutic response. Eur Urol Oncol 2018;1:325–37. 1056–72.

5636 Clin Cancer Res; 25(18) September 15, 2019 Clinical Cancer Research

Combined DNA-PK/TORK/AR Targeting in Prostate Cancer

52. Amatya PN, Kim HB, Park SJ, Youn CK, Hyun JW, Chang IY, et al. A role of 55. Motzer RJ, Escudier B, Oudard S, Hutson TE, Porta C, Bracarda S, et al. DNA-dependent protein kinase for the activation of AMP-activated protein Efﬁcacy of everolimus in advanced renal cell carcinoma: a double-blind, kinase in response to glucose deprivation. Biochim Biophys Acta 2012; randomised, placebo-controlled phase III trial. Lancet 2008;372:449–56. 1823:2099–108. 56. Park JC, Eisenberger MA. Advances in the treatment of metastatic prostate 53. Park SJ, Gavrilova O, Brown AL, Soto JE, Bremner S, Kim J, et al. DNA-PK cancer. Mayo Clin Proc 2015;90:1719–33. promotes the mitochondrial, metabolic, and physical decline that occurs 57. Rathkopf DE, Larson SM, Anand A, Morris MJ, Slovin SF, Shaffer DR, et al. during aging. Cell Metab 2017;26:447. Everolimus combined with geﬁtinib in patients with metastatic castration- 54. Dutcher JP, de Souza P, McDermott D, Figlin RA, Berkenblit A, Thiele A, resistant prostate cancer: phase 1/2 results and signaling pathway implica- et al. Effect of temsirolimus versus interferon-alpha on outcome of patients tions. Cancer 2015;121:3853–61. with advanced renal cell carcinoma of different tumor histologies. 58. Statz CM, Patterson SE, Mockus SM. mTOR inhibitors in castration- Med Oncol 2009;26:202–9. resistant prostate cancer: a systematic review. Target Oncol 2017;12:47–59.

www.aacrjournals.org Clin Cancer Res; 25(18) September 15, 2019 5637

Pleiotropic Impact of DNA-PK in Cancer and Implications for Therapeutic Strategies

Emanuela Dylgjeri, Christopher McNair, Jonathan F. Goodwin, et al.

Clin Cancer Res 2019;25:5623-5637. Published OnlineFirst July 2, 2019.

Updated version Access the most recent version of this article at: doi:10.1158/1078-0432.CCR-18-2207

Supplementary Access the most recent supplemental material at: Material http://clincancerres.aacrjournals.org/content/suppl/2021/02/25/1078-0432.CCR-18-2207.DC1

Cited articles This article cites 58 articles, 18 of which you can access for free at: http://clincancerres.aacrjournals.org/content/25/18/5623.full#ref-list-1

Citing articles This article has been cited by 1 HighWire-hosted articles. Access the articles at: http://clincancerres.aacrjournals.org/content/25/18/5623.full#related-urls

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Department at Subscriptions [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://clincancerres.aacrjournals.org/content/25/18/5623. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.