Predictive Biomarkers for Cancer Therapy with PARP Inhibitors

REVIEW Predictive biomarkers for cancer therapy with PARP inhibitors

J Michels1,2,3, I Vitale4,5, M Saparbaev2,3,6, M Castedo1,2,3,11 and G Kroemer1,2,3,7,8,9,10,11

Poly(ADP-ribose) polymerase (PARP) inhibitors have raised high expectations for the treatment of multiple malignancies. PARP inhibitors, which can be used as monotherapies or in combination with DNA-damaging agents, are particularly efficient against tumors with defects in DNA repair mechanisms, in particular the homologous recombination pathway, for instance due to BRCA mutations. Thus, deficient DNA repair provides a framework for the success of PARP inhibitors in medical oncology. Here, we review encouraging results obtained in recent clinical trials investigating the safety and efficacy of PARP inhibitors as anticancer agents. We discuss emerging mechanisms of regulation of homologous recombination and how inhibition of DNA repair might be used in cancer therapy. We surmise that the identification of patients that are likely to benefit from PARP inhibition will improve the clinical use of PARP inhibitors in a defined target population. Thus, we will place special emphasis on biomarker discovery.

Oncogene (2014) 33, 3894–3907; doi:10.1038/onc.2013.352; published online 16 September 2013 Keywords: PAR; base excision repair; homologous recombination; BRCAness; synthetic lethality

INTRODUCTION PARP1 IN PHYSIOLOGICAL AND PATHOLOGICAL RESPONSES The superfamily of poly(ADP-ribose) (PAR) polymerases (PARPs),1 PARP1 is historically known as a sensor of DNA nicks, contributing whose discovery dates back to the early 1960s, nowadays en- to the single-strand break repair, the orchestration of the DNA compasses 17 distinct members that participate in a wide array damage response (DDR) and the maintenance of genomic of cell functions, including DNA transcription, DNA repair, stability.10,11 Nevertheless, several recent studies have led to the genomic stability maintenance, cell cycle regulation as well as identification of a plethora of novel roles for PARP1, placing its non-apoptotic cell death (Table 1).2–5 activity at the center of the cellular stress response network. PARP1, a multifunctional enzyme of 113 kDa, is the most PARP1 À / À knockout mice exhibit (i) increased genetic instability abundant, ubiquitously expressed and best-characterized member as measured by the frequency of micronuclei, sister chromatid of the superfamily.6 PARP1 has a highly conserved structural and exchanges and other chromosome aberrations as well as (ii) major functional organization including: an N-terminal double zinc finger DNA repair defects coupled to hypersensitivity to alkylating agents 12 DNA-binding domain, a central auto-regulation domain and a and g-radiation. Nonetheless, the inhibition of PARP is not lethal 13 À / À C-terminal catalytic domain that binds oxidized nicotinamide for mammals, and Parp1 mice are viable and fertile, even adenine dinucleotide (NAD þ ).7 PARP1 is a DNA nick sensor that though they manifest accelerated aging and exhibit a higher þ incidence of spontaneous or carcinogen-induced tumors than utilizes NAD as a substrate to catalyze the covalent attachment 14 of ADP-ribose units to the g-carboxyl group of Glu residues of a wild-type controls. However, the simultaneous loss of PARP1 and PARP2 is lethal at the onset of gastrulation, suggesting that the two wide array of target proteins. This post-translational modification 15 known as poly-ADP-ribosylation or PARylation increases more PARP proteins have redundant roles in embryogenesis. than 500-fold in the presence of DNA-strand breaks.8 Although PARP1 accounts for the vast majority (B85%) of PAR polymer PARP1 and the DDR synthesis, five other members have catalytic activity of PARylation, Due to the continuous exposure to endogenous and exogenous namely PARP2, PARP3, PARP4 and PARP5A/B (also known as DNA-damaging insults, cells experience DNA lesions such as SSBs 9 tankyrase1/2). Recently, Hottiger reported a new classification for and double-strand breaks (DSBs) that require continuous activa- the ADP-ribosyl transferases grouping the PARP family members tion of DNA repair pathways (Box 1). and distinguishing proteins with mono or poly(ADP)ribosylation The levels of PARP1 protein increase in response to DNA activity.9 damage,16 and PARP1 contributes to both SSB and DSB repair. Here, we provide an overview on the roles of PARP1, focusing PARP1 associates with persisting SSBs or DSBs where it becomes on its involvement in cellular stress management, and the activated and catalyzes the PARylation of multiple nuclear acceptor possibility to target PARP1 for the treatment of cancer. We will proteins adjacent to the DNA break including that of PARP1 itself, devote special attention to the discovery of predictive biomarkers histones, DNA repair proteins, chromatin regulators and transcription that may improve patient selection. factors.7,17 PARylation facilitates the chromatin decondensation,

1INSERM, U848, Villejuif, France; 2Institut Gustave Roussy, Villejuif, France; 3Universite´ de Paris Sud, Villejuif, France; 4Regina Elena National Cancer Institute, Roma, Italy; 5National Institute of Health, Rome, Italy; 6CNRS UMR8200, Villejuif, France; 7Metabolomics Platform, Institut Gustave Roussy, Villejuif, France; 8Centre de Recherche des Cordeliers, Paris, France; 9Poˆle de Biologie, Hoˆpital Europe´en Georges Pompidou, AP-HP, Paris, France and 10Faculty of Medicine, Universite´ Paris Descartes, Paris, France. Correspondence: Dr M Castedo or Professor G Kroemer, INSERM U848, Institut Gustave Roussy, 114 rue Edouard Vaillant, F-94805 Villejuif, France. E-mail: [email protected] or [email protected] 11These authors share senior co-authorship. Received 17 June 2013; revised 9 July 2013; accepted 12 July 2013; published online 16 September 2013 Anticancer proﬁle of PARP inhibitors J Michels et al 3895

Table 1. Multiple PARP functions Box 1 DNA repair pathways Actor Function Reference Base excision repair (BER) is a DNA repair pathway that DNA repair XRCC1 BER 11,22 removes non-bulky base lesions in form of the free bases, as DDB2, XPA NER 24–26 they occur after spontaneous, oxidative and alkylating DNA MRE11 HRR 27 damage.181 BER includes different effectors acting at distinct 35 DNA-PKC NHEJ steps. In human cells DNA glycosylases (for example,

88 8-oxoguanine-DNA glycosylase (OGG1), alkyl-N-purine-DNA Genomic Topoisomerase I Genomic glycosylase (ANPG), endonuclease III DNA glycosylase hNTH1) stability maintenance 182–185 Telomere Genomic stability 199 retrieve the damaged bases generating abasic sites that then are cleaved by AP endonuclease 1, APE1, inducing single- endings 186 Mitotic spindle Mitosis 199 strand breaks (SSBs). PARP1 and PARP2 attract different poles repair enzymes to the DNA lesion site (for example, X-ray cross-complementing factor 1 XRCC1, polymerase b,pro- DNA Histone H1 Decondensation of 200 liferating cell nuclear antigen (PCNA), ligase 3).37 transcription the double DNA Nucleotide excision repair (NER) is a pluripotent pathway helix 201 for recognition and removal of bulky helix disturbing DNA Cell cycle c-Fos, c-Myc G0-G1 transition lesions generated by UV irradiation and anticancer agents progression Inflammation TNF-a, NF-kB, Acute and chronic 62 such as cisplatin (CDDP). NER involves dual incisions that HMGB1 inflammation bracket lesion site and release of the 24–32 nucleotide-long Cell NAD þ Mitochondrial 66 oligomer containing damaged residues. Two sub-pathways metabolism respiration are identified, (i) the transcription-coupled repair (TC-NER) in transcriptionally active genes and (ii) the global genomic Cell death/ AIF Parthanatos 47 repair (GG-NER) operating on the entire genome.187 RNA survival polymerase II is stalled at the DNA lesions are sensed by ROS Apoptosis 202 66,202 cockayne syndrome WD repeat proteins (CSA and CSB) in TC- Glycolysis Necrosis NER and by Xeroderma pigmentosum (XPC) in GG-NER.188 In inhibition, NAD þ depletion both pathways, the lesions are excised by endonucleases (for AMPK activation Autophagy 52 example, XPG and ERCC1-XPF) and resulting DNA gap is then filled and sealed by DNA polymerases d, e and ligase III, respectively. NER also contributes to repair intra and Abbreviations: AIF, apoptosis-inducing factor; AMPK, AMP kinase; ATM, interstrand crosslinks (ICL) and oxidative DNA damage. ataxia-telangiectasia mutated; BER, base excision repair; DDB2, damaged Mismatch repair (MMR) is a system that recognizes and DNA-binding protein; DNA-PKc, DNA-dependant protein kinase; HRR, homologous recombinational repair; NAD þ, nicotinamide adenine dinu- repairs DNA replication and recombination errors, such as cleotide; NER, nucleotide excision repair; NHEJ, non-homologous end mismatched bases, deletions and insertions. In human cells joining; PARP, poly-(ADP-ribose) polymerase; ROS, reactive oxygen species; two major heterodimers: Msh2/Msh6 (MutSa), and Msh2/ TNF-a, tumor necrosis factor a; XPA, xeroderma pigmentosum comple- Msh3 (MutSb) recognize DNA mismatches and trigger their mentary group A; XRCC1, X-ray repair cross-complementing 1. removal by recruiting MutLa (MLH1/PMS2) and MutLb (MLH1/PMS1) complexes. Direct repair (DR) is the direct reversal of a damaged base to its native state without excision and de novo DNA synthesis. thereby rendering DNA lesions accessible to the repair machinery. In At present, there are three types of DNA damage known to addition, PARP1 may favor the assembly of protein complexes be repaired by chemical reversion including: photoreactiva- involved in DDR.18,19 Although controversial,20,21 PARP1 may operate tion process by photolyases, O-methylation (in O6-Guanine, in base excision repair (BER) by detecting the presence of SSBs, as O4-Thymine and phosphates) is directly reversed by the well as by recruiting and modulating the activity of repair proteins protein O6-methylguanine DNA methyltransferase (MGMT) such as X-ray cross-complementing gene 1 (XRCC1), DNA ligase III and oxidative demethylation of N-methyl groups by AlkB and 8-oxoguanine-DNA glycosylase.11,22 family proteins. MGMT transfers the methyl/alkyl groups PARP1 may contribute to nucleotide excision repair, as from mutagenic DNA base lesions O6-meG and O4-meT suggested by the finding that ultraviolet radiation-induced direct generated by chemotherapeutic agents like dacarbazine and DNA lesions including thymine dimers—which are removed by TMZ onto one of its nucleophilic cysteine residues situated in the nucleotide excision repair machinery—activate PARP1.23 the protein active site. Importantly, the cysteine acceptor site PARP1 operates in nucleotide excision repair by stabilizing is not regenerated after the transfer, and therefore these damaged DNA-binding protein 224,25 or by associating with proteins are not true enzymes. The self-methylated DNA- Xeroderma Pigmentosum Complementation Group A.26 methyltransferases are referred to as suicidal DNA repair PARP1 is believed to promote the repair and restart of stalled proteins, as they are irreversibly inactivated during this replication forks during homologous recombinational repair (HRR) stoichiometric repair reaction.73,189 by attracting MRE11, a component of the MRN (Mre11-Rad50- Homologous recombination (HR) is a prominent DNA Nbs1) complex, to collapsed replication forks.27 The combined homology-directed double-strand break (DSB) repair pathway action of PARP1 and BRCA2 may also protect stalled replication that in general proceeds in error-free manner. In HR repair forks from Mre11-mediated degradation.28,29 Moreover, PARP1 pathway, Rad51 protein catalyzes DNA-strand exchange activates HRR by antagonizing the suppressive impact of the between similar or identical DNA molecules. DSB is repaired KU70–KU80 heterodimer and/or ligase IV—two key players of using the homologous non-damaged sister chromatid or non-homologous end joining (NHEJ)—on HRR.30–32 homologous chromosome as a template. In human cells, HR PARP1 is essential to accomplish alternative (A) NHEJ.33,34 requires BRCA1 and BRCA2, RAD51 and FANC proteins. HR Moreover, PARP1 may cooperate with DNA-PKcs in the initiating only acts during the S and G2 phases of the cell cycle.190 step of canonical (C) NHEJ.35 However, the precise contribution of

& 2014 Macmillan Publishers Limited Oncogene (2014) 3894 – 3907 Anticancer profile of PARP inhibitors J Michels et al 3896 the mitochondria would lead to the mitochondrial release of Translesion synthesis (TLS) is a DNA damage tolerance apoptosis-inducing factor, which then translocates to the nucleus pathway that allows the DNA replication machinery to and participates in chromatin degradation.45,46 The binding of replicate past DNA lesions by switching on the specialized apoptosis-inducing factor to PAR may be important for the translesion DNA polymerases (for example, Y-family poly- activation of parthanatos.47 merase H and A-family polymerase Q), which is often In striking contrast with its role in the execution of cell death, overactivated in cancer cells. TLS operates if the S-phase PARP1 also has a pivotal function in cell survival processes. Indeed, has been initiated in spite of DNA damage. PARP1 activity is required for the maintenance of genomic Non-homologous end joining (NHEJ) is a pathway that stability, and PARylation is implicated in both DNA repair (see repairs double-strand breaks generated by closel-spaced above) and mitosis.48 Moreover, growing evidence suggests that nicks on the complementary strands. In NHEJ DSB is ligated PARP1 activation might be involved in the decision of cells to without need for a homologous template this process induce a cytoprotective autophagic response, thereby avoiding involves KU70 (also known as XRCC6), KU80 (also known as cell death.49–51 Reportedly, PARP1 may mediate autophagy XRCC5), DNA-dependant protein kinase catalytic subunit induction through the activation of AMP kinase.52–54 (DNA-PKcs), artemis, ligase 4 and XRCC4-XPF. NHEJ is cell 191 Beyond its inhibitory action on the DDR, PARP inactivation may cycle independent and rapid though highly error-prone. be cytotoxic for additional reasons. Thus, PARP inhibition may 16 (i) elicit a cell cycle arrest in the G2/M phase; (ii) increase the level of genomic instability, including a hyperrecombinant phenotype,55 56,57 PARP1 to C-NHEJ—at least in human cells—remains to be telomere shortening and chromosome aberrations (for established24, and a role of PARP3, rather than PARP1, in example, chromosome fusion, sister chromatid exchange, micro- 13,58 promoting the DSB repair during C-NHEJ has recently been nuclei and aneu-/tetraploidization); and (iii) activate the proposed.36 intrinsic pathway of apoptosis, as recently demonstrated by our 59 Given its multiple roles in DDR, pharmacological inhibitors of group on a panel of CDDP-resistant NSCLC cell lines. the enzymatic activity of PARP are considered as promising agents for anticancer therapy (see below). PARP inhibitors subvert SSB Other functions of PARP1 repair, thereby inducing the formation of stalled replication forks and/or suppress DSB repair. Alternatively, PARP inhibitors may act Apart from its contribution in recovering from DNA damage and in regulating cell death, PARP has an important role in inflammatory by trapping PARP1 at the SSB intermediate formed during BER, 60,61 thus impairing the access of BER proteins to the DNA lesion and processes and in ischemia-reperfusion injury. PARP1 acts provoking the obstruction of replicative forks.37 PARP inhibitors during acute and chronic inflammation by (i) stimulating the are particularly effective against BRCA1- and BRCA2-defective secretion of pro-inflammatory mediators (for example, TNFa, 16 NF-kB and HMGB1), (ii) regulating the differentiation of T cells and tumors by provoking synthetic lethality. In this context, a current 62 model suggests that, in conditions of HRR deficiency (HRD), which (iii) controlling the maturation of B lymphocytes. In line with may result from loss-of-function mutations in BRCA1/2, DNA these observations, some preclinical studies demonstrated that the inhibition of PARP has antidiabetic effects and confers cardio- damage sensors (NBS1, ataxia-telangiectasia mutated (ATM), ATR, 63–65 CHK1 and CHK2) or RAD51, the inhibition of PARP would trigger protection and neuroprotection in vascular diseases. Both PARP1 and PARP2 have been implicated in the regulation the error-prone NHEJ mechanism, resulting in genetic instability 66 and/or cell death.38 Consistent with this hypothesis, HRR-deficient of bioenergetic metabolism. Through yet ill-defined mechanisms, PARP1 hyperactivation may also contribute to the pathogenesis of cells become insensitive to PARP inhibitors when NHEJ is disabled 67,68 by DNA-PKcs inhibitors.38 Alzheimer’s disease. Finally, PARP activation downregulates mitochondrial respiration and oxidative metabolism, thereby enhancing age-related diseases (for example, diabetes and PARP1 as a modulator of cell survival and death obesity). This effect could be ascribed to three different mecha- PARP1 is believed to have key functions in the modulation of the nisms, namely (i) direct PARylation of some metabolic enzymes cellular response to several stressors.19 Upon exposure to DNA- potentially affecting their catalytic activity, (ii) effects on metabolism-relevant transcription factors and (iii) alteration of damaging agents or endogenous oxidative stress, PARP1 becomes þ 66 active and generates signals that either favor adaptation to stress cellular NAD concentrations. In this context, PARP1 inhibitors or stimulate cell death. The cell’s ultimate fate depends on the have been evaluated as agents that may avoid excessive necrosis nature of the stimulus, the severity of the damage, the cellular after cardiac or brain ischemia. context and the balance between the activation versus inhibition of PARP. Uncontrolled PARP1 activation may result in the induction of cell PARP INHIBITORS—PRECLINICAL DATA death. Upon extensive DNA damage that saturates the DNA repair Several distinct PARP inhibitors may synergize with conventional machinery, prolonged PARP activation can result in the depletion of DNA-damaging agents, such as ionizing radiation and chemothera- NAD þ and subsequently of ATP pools, thus inducing necrotic cell peutics, in increasing their in vitro and in vivo effectiveness death.39–41 Moreover, PARP1 may act downstream of the kinases against tumor cells. This applies to investigational agents such as RIPK1 and RIPK3 in TRAIL-induced necroptosis.42 As compared with 3-aminobenzamide, NU1025 (8-hydroxy-2-methylquinazolin-4- wild-type controls, Parp1 À / À mice are more resistant to several (3H)one), NU1085 (2-(4-hydroxyphenyl)benzamidazole-4-carboxa- paradigms of injury-induced tissue degeneration, reinforcing the mide), PJ34 (phenanthridinone), as well as to recently developed notion that PARP1 contributes to the execution of regulated PARP inhibitors including ABT888 (also known as Veliparib), necrosis.43 Although the cleavage and inactivation of PARP1 by AG014699 (also known as Rucaparib, AG14447 or PF-01367338), caspases is a well-recognized byproduct of apoptosis,5 the CEP8983 (a 4-methoxy-carbazole derivate), E7016, GPI 15427 and implication of PARP1 in cell death is controversial and limited (for olaparib (also known as AZD2281 or KU-0059436). more details see Virag et al.44). Nevertheless, PARP overactivation Inhibition of PARP1 by olaparib or similar agents increases the seems to trigger a peculiar form of cell death called ‘parthanatos’ radiosensitivity of glioblastoma69,70 and non-small-cell lung that is dependent from apoptosis-inducing factor but independent carcinoma,71 both in vitro and in vivo. The underlying mecha- from caspases. According to one proposed model, the activation of nism of olaparib-mediated radiosensitization was found to involve PARP1 and the translocation of PAR polymers from the nucleus to DNA replication-generating DSBs.70

Oncogene (2014) 3894 – 3907 & 2014 Macmillan Publishers Limited Anticancer profile of PARP inhibitors J Michels et al 3897 Temozolomide (TMZ) is an alkylating agent currently used for constitutive PARP hyperactivation96 leading to an increase in the the treatment of malignant glioma and metastatic melanoma that intracellular PAR levels without any consistent increase in the level leads to the formation of a wide spectrum of methyl DNA adducts, of PARP protein expression.97 As a result, increased levels of PAR including O-methyl adducts, which are repaired by the DR may predict responsiveness to PARP inhibitors. pathway,72 and N-methyl adducts, which are removed by DNA Apart from BRCA mutations, constitutive modifications of the glycosylase-initiated BER and by the AlkB family of oxidative DNA expression of other DDR effectors may induce HRD, a condition demethylases during DR.73 To date, TMZ resistance is believed to that has been referred to ‘BRCAness’ and that may be efficiently arise from two distinct mechanisms, namely (i) the increased targeted by PARP inhibitors (Box 2). In this context, Aurora A activity of the O6-methylguanine DNA methyltransferase that overexpression and MRE11 downregulation may be explored as eliminates the O-methyl adducts or (ii) a deficiency in MMR, which surrogate markers of BRCAness. Aurora A is a mitotic kinase with results in microsatellite instability, and, consequently, in the lack of essential roles in cell cycle control and mitosis.98,99 Aurora A is repair of—and increased tolerance to—O-methyl adducts.74,75 overexpressed in multiple cancers, and this deregulation has been Inhibition of PARP with consequent defective BER enhances the causally linked to tumorigenesis.100 The overexpression of Aurora cytotoxicity of TMZ, correlating with an increase in TMZ-induced A may affect DSB repair by preventing the recruitment of RAD51 N-methyl adducts.76 Such synergistic effects were observed in to DNA DSBs by a mechanism that would require the inhibition of different cell types responding to TMZ, namely (i) in melanoma the checkpoint kinase CHK1.101 As a result, DSB repair by HR is and cervical carcinoma cells, if combined with the pharmaco- impaired and cells become susceptible to PARP inhibitors. Aurora logical PARP inhibitor GPI 15427 or small interfering RNAs for the A overexpression is a candidate predictive biomarker of PARP depletion of PARP1 and PARP2, (ii) in neuroblastoma cells, if inhibitor sensitivity. combined with AG014699, both in vitro and in vivo, in xenografted The phosphatase and tensin homolog (PTEN) is a tumor tumor models,77,78 and (iii) in a other tumor types, including those suppressor gene that inhibits the oncogenic phosphoinositide a priori non-responsive to TMZ, if combined with ABT888.79 3-kinase (PI3K)–AKT–mTOR pathway downstream of epidermal Remarkably, in leukemia cells, the combination between ABT888 growth factor receptor (EGFR) signaling. Loss of PTEN is most and TMZ inhibited the growth of MMR-deficient and MMR- proficient cells regardless of the O6-methylguanine DNA methyltransferase level.80 Intriguingly enough, GPI 15427 enhanced the in vivo antitumor effects of TMZ even against PARP1-depleted Box 2 BRCAness cells.78 This finding indicates that GPI 1542 might act on other Germline-transmitted inactivating mutations of the BRCA1 or PARP isoforms than PARP1 or might have yet-to-be-elucidated BRCA2 genes compromise. HR proficiency, predispose to off-target effects. female breast (85% lifetime risk), ovarian (10–40%), male The association of PARP inhibitors with cisplatin (CDDP) showed 192 breast, pancreatic and prostate cancer. Sporadic tumors additive or hyperadditive effects on cervical cancer, liver cancer 81–83 may display ‘BRCAness’, that is, HR defects without mutations and testicular germ cell tumor cell lines. Moreover, when 171,193 in BRCA1 or BRCA2. BRCAness can be explained by added to NSCLC cells, CDDP synergizes with PJ34, CEP8983 or epigenetic inactivation (for instance due to promoter PARP1-depleting small interfering RNAs with regard to the methylation) of BRCA1 or BRCA2, (epi)mutations in other inhibition of cell proliferation, induction of DNA damage foci genes, post-translational protein modifications, as well as and activation of the intrinsic apoptotic pathway.84 BRCA1-/2- pharmaclogical inhibitors that inactivate HR-dependent repair: mutated breast cancer xenografts were also more sensitive to the combination of AG014699 and carboplatin (which is chemically related to CDDP) than to each drug alone.85 Sporadic ovarian (5–31%) (55% of high-grade serous ovarian Topoisomerases I and II have pivotal roles in DNA replication carcinoma) and breast cancers (11–14%) (in particular TNBC) present an epigenetic silencing of BRCA1 by methylation of and transcription, in thus far that they ‘unwind’ the double DNA 171,194 86 it’s promoter. helix. Topoisomerase inhibitors act by stabilizing this unwinded 195 form of the DNA helix, thus causing the subsequent conversion ATM and MRE11 mutations in lymphocytic leukemia cause of transient SSBs into permanent DSBs during DNA replication.87 defects in the early steps of the HR pathway. PARP1 interferes with the DNA relaxation activity of topo- Aurora A, a mitotic kinase, overexpressed in a diverse array of cancers may inhibit the recruitment of RAD51 at the DNA isomerase I either through direct protein–protein interaction or 101 through PARylation.88 Importantly, PARP inactivation potentiates damage site. Fanconi gene promoter methylation has been reported in the cytotoxicity of topoisomerase I inhibitors (for example, 196–198 camptothecin and topotecan) on neuroblastoma, colorectal lung, lung, cervical and ovarian cancer. breast and ovary human cancer cells,77,89,90 as well as that of The amplification of EMSY a protein capable of silencing BRCA2 is observed in sporadic breast (13%) and high-grade topoisomerase II inhibitors (for example doxorubicin, C-1305) on 172 liver, ovarian and BRCA1-positive breast cancer cells.91–93 ovarian cancer (17%). ETS transcription factors are aberrantly expressed in different cancers and may repress BRCA1/2 and subsequently impair 140 Targeting DNA repair deficiencies HRR. PARP inhibitors given as standalone therapies have been shown to PARP1 binding protein PARPBP (also known as PARI or CI2orf48) may inhibit HR by interacting with RAD51 at mediate some degree of cytotoxicity against human cancer cell 177 lines grown in vitro,16 as well as in a few in vivo human and murine replication forks. PARPBP is overexpressed in pancreatic 85 ductal carcinoma and positively regulate the poly(ADP-ribosyl) tumor models. PARP inhibitors are particularly toxic when they 175,176 are administered to BRCA-deficient cells, suggesting that PARP ation activity of PARP1. inhibition and HR defects might be synthetically lethal for cancer Pharmacological inhibitors of the HR are being developed for cells.94 Thus, pharmacological or genetic inactivation of either cancer therapy. In particular, specific inhibitors of ataxia- PARP1, BRCA1 or BRCA2 alone is compatible with cell survival, telangiectasia mutated (ATM) kinase, cyclin-dependent whereas their simultaneous inhibition is not.16 This discovery kinase 1 (CDK1), epidermal growth factor receptor (EGFR), raised the possibility to exploit PARP inhibitors for the treatment heat-shock protein of 90 KDa (HSP90), histone deacetylases of breast or ovarian cancers with defects in BRCA1 or BRCA2.95 Of (HDAC), phosphoinositide 3-kinase (PI3K) pathway and note, HR-deficient cells susceptible to PARP inhibitors present BCR-ABL (imatinib) may compromise HR (see main text).

& 2014 Macmillan Publishers Limited Oncogene (2014) 3894 – 3907 Anticancer profile of PARP inhibitors J Michels et al 3898 frequently found in glioblastoma, NSCLC, as well as endometrial Imatinib downregulates the HRR and renders human prostate and (up to 80% of cases), breast and prostate cancers.102 The impact of pancreatic cancer cell lines susceptible to radiotherapy and PTEN loss on DSBs repair is controversial.103–105 Some authors chemotherapy (that is, gemcitabine and mitomycine C).126 As a reported that the absence of PTEN sensitized cancer cells to result, it may be worthwhile to explore the association of imatinib the inhibition of PARP, presumably by downregulating RAD51 with PARP inhibition. and consequently impairing HR.106,107 Accordingly, KU0058948 Some recent results indicate that BRCA1 is functional only upon selectively killed endometrioid endometrial cancer cells displaying phosphorylation by CDK1, a core component of the cell cycle PTEN loss and HR deficiency.108 Nevertheless, a recent study machinery.127 The depletion of CDK1 by means of a doxycycline performed on primary prostate cancers has weakened this inducible shRNA compromises the ability of H1299 NSCLC cells to conclusion, as the authors failed to observe a significant form RAD51 foci in response to g-irradiation.128 Accordingly, the correlation between, on one hand, the PTEN status and, on the simultaneous inhibition of CDK1 and PARP synergistically inhibited other hand: (i) RAD51 expression levels, (ii) sensitivity to PARP the growth of human NSCLC xenografts and of transgene-induced inhibitors and (iii) the expression of genes associated to synthetic lung adenocarcinomas.128 lethality, such as RAD51, ATM, PRKDC, BRCA1, BRCA2, MRE11, CDK6 EGFR inhibitors, namely tyrosine kinase inhibitors (that is, and MSH2.105 Instead, PARP I sensitivity was associated with a erlotinib and gefitinib), and monoclonal antibodies (for example, defect in MRE11 expression coupled to PARP hyperactivation.105 cetuximab) are effective treatments in EGFR mutated NSCLC and MRE11 is a key component of the MRN complex (Mre11-Rad50- KRAS wild-type colorectal cancers respectively.129,130 Intriguingly, Nbs1) that promotes the restart of stalled replication forks and EGFR inhibitors attenuate the repair of DSBs by HR (that is, by induces HRR by recruiting and activating the ATM kinase.27,109,110 downregulation of BRCA1) and NHEJ (that is, by inhibiting the Thus, MRE11 deficiency may be a more accurate predictive marker phosphorylation and translocation of DNA-PK to the for PARP inhibition sensitivity than PTEN deficiency. nucleus),131,132 conferring susceptibility to PARP inhibition. It will PARP inhibitors can kill cells lacking DDR pathways other than be interesting to learn whether these preclinical results can be HRR. Thus, HRR-proficient human mesothelioma, NSCL, epithelial confirmed in vivo, in suitable xenograft models. ovarian and cervix cancer cell lines that had acquired CDDP Inhibitors of the chaperone heat-shock protein 90 such as resistance due to prolonged culture in the presence of the drug, allylamino-17-demethoxygeldanamycin 17-AAG—a compound upregulated the activity of PARP (and hence promoted PARylation that is currently under clinical investigation133,134—may inhibit of cellular, mostly nuclear proteins).59 When exposed to PARP HR by provoking the degradation of some proteins such as BRCA2 inhibitors, these cells increased the number of BRCA1 and RAD51 and RAD51.135 Consistent with this notion, 17-AAG exacerbates foci, which are both surrogate markers of HRR activation.111 the cytotoxic effects of both irradiation and olaparib on Although these cells were proficient in the initiation of BER, glioblastoma multiforme.136 Along similar lines, histone they were deficient in XRCC1 and polymerase b.59 Intriguingly, deacetylase inhibitors can affect the DDR by modifying the synthetic lethality has been observed for Chinese hamster ovarian expression of proteins involved in HR.137 Indeed, the histone cells and human ovarian, breast and cervix cancer cells when they deacetylase inhibitor suberoylanilide hydroxamic acid efficiently were first XRCC1-depleted and then treated with either PARP or inhibits the growth of hepatocellular carcinoma cells when DSB inhibitors (for example, ATM and DNA PKC inhibitors).21,112 associated with olaparib.138 These findings are encouraging, but These results suggest that the absence of XRCC1 expression may need further confirmation. constitute a biomarker that predicts sensitivity to PARP inhibitors. ETS gene fusions are found in a wide array of cancers, including Ewing’s sarcoma, acute myeloid leukemia and prostate cancer.139 PARP inhibition suppresses the growth of ETS gene fusion-positive Manipulating HRR for sensitization to PARP inhibitors but not ETS gene fusion-negative xenografts, along with the One approach to extend the indications of PARP inhibitors to HRR- accumulation of double-strand DNA lesions.140 ETS was recently proficient cancers might consist in combining them with chemical identified as the transcription factor with the highest number of inhibitors of HRR. Prospective targets related to the HRR pathway BRCA1/2 binding motifs.141 ETS may repress the BRCA promoter include ATM, the cyclin-dependent kinase 1 (CDK1), EGFR, heat- upon its activation by the mitogen-activated protein kinase shock protein of 90 KDa, histone deacetylase and the PI3K pathway.142,143 PI3K inhibitors have been reported to stimulate pathway (see below). Moreover, gimeracil has been discovered ERK-dependent ETS1 activity and subsequent HR impairment due as a drug that inhibits the early steps of HRR.113 to decreased expression of BRCA1/2 and concomitant upregulation The combined inhibition of PARP1 and ATM is incompatible of PAR levels indicative of increased PARP activity.141,144 As a result, with embryogenesis, as revealed by the phenotype of double triple-negative breast cancer (TNBC) BRCA-proficient cells knockout mice.114 ATM, a kinase triggered by DNA damage (that succumbed to the combined pharmacological inhibition of PI3K is, DSB), induces DDR (that is, HRR and NHEJ) and cell cycle and PARP.141 These results indicate that the presence of ETS gene checkpoint activation that is often dysregulated in cancer.115 PARP fusions predict sensitivity to PARP inhibitors. Furthermore, they inhibitors are extremely efficient against cancer cells bearing ATM suggest that therapeutic measures that upregulate ETS1 function defects, as shown in the context of Mantle cell lymphoma.116–118 may synergize with PARP inhibitors. PARP-inhibited cells were also found to be sensitive to the inhibition of the DNA-PK variant involved in NHEJ.32,119,120 Notable HER2 as a modulator of the response to PARP inhibitors differences exist between different proteins involved in NHEJ. HER2 overexpressing (HER2 þ ) human breast cancer cell lines are This applies to DNA-PK and KU80 in the sense that Parp1 À / À susceptible to PARP inhibition by ABT888 and AZD2281 regardless DNA-PK À / À mice are viable, while Parp1 À / À KU80 À / À mice are of their HR status. HER2 knockdown abrogated their PARP not.121 Of note, the association of DNA-PK and ATM inhibitors was hypersensitivity.145 Of note, one third of human breast cancers unable to kill PARP1-deficient cells more efficiently than ATM are HER2 þ , and the overexpression of HER is a negative inhibitors alone.121 The synergistic effect of PARP inhibitors and prognostic factor.146 HER2 overexpression has been linked to ATM inhibitors is encouraging and warrants further investigation. the activation of NF-kB.147 Given the fact that PARP is a co- Another downstream target of DNA-PK is the non-receptor activator of NF-kB,148 it is well possible that the susceptibility of protein kinase c-ABL that phosphorylates RAD51.122,123 c-ABL is HER2 þ breast cancer cells to PARP inhibition can be explained by specifically targeted by imatinib mesylate, a tyrosine kinase their dependence on the NF-kB signaling pathway.145 As a result, inhibitor that transformed the prognosis of patients with gastro- HER2 overexpression might predict the response of breast cancer intestinal stromal tumors and chronic myeloid leukemia.124,125 cells to PARP inhibition.

Oncogene (2014) 3894 – 3907 & 2014 Macmillan Publishers Limited Anticancer profile of PARP inhibitors J Michels et al 3899 53BP1 and acquired PARP resistance confirmed or clinically suspected HRR deficiencies (BRCA muta- Upon long-term exposure to PARP inhibitors, BRCA-deficient tions, basal-like or TNBC, high-grade EOC with platinum sensitivity), tumors develop an acquired resistance to these compounds and rather few studies included biomarker analyses (Tables 2–3). through a mechanism that often implies the restoration of HR The pharmacokinetic and pharmacodynamic characteristics of proficiency.149 Thus, pancreatic cancer cells carrying a BRCA2 olaparib have first been evaluated in a phase I trial on 60 patients mutation became resistant to PARP inhibition due to intragenic with solid tumors (including ovarian, breast and prostate cancer), 95 deletions restoring the open-reading frame of BRCA2.150,151 Along 22 of whom carried BRCA1 or BRCA2 mutations. Importantly, similar lines, loss of 53BP1 (a DNA damage-responsive intranuclear objective tumor responses were observed only among BRCA protein involved in HR and NHEJ), an event that restores HR, can carriers with 63% clinical benefit rate. Encouraged by these results, render BRCA1-deficient cells resistant to PARP inhibitors152 and a supplemental cohort of 50 ovarian patients bearing BRCA1 or other DNA-damaging agents such as CDDP.153 Thus, 53BP1 BRCA2 mutations was included. Among them, 13 were platinum- expression levels (or functionality) may constitute an interesting refractory (progressive disease, PD, during previous platinum- biomarker to identify HR-deficient tumors that would respond to based chemotherapy), 24 were platinum-resistant (PD o6 months PARP inhibitors. after previous platinum-based chemotherapy) and 13 were platinum sensitive (PD 46 months after previous platinum- based chemotherapy). These patients exhibited a response rate (according to RECIST criteria or reduction in the circulating levels CLINICAL EVALUATION OF PARP INHIBITORS of the ovarian cancer-associated marker CA-125) of 23%, 45% and Current clinical trials involve multiple PARP inhibitors, such as 69%, respectively.155 The effectiveness of olaparib correlated with AZD2461 and olaparib (also known as AZD2281 or KU-0059436) BRCA mutation and platinum sensitivity. from AstraZeneca (Cambridge, UK), veliparib (also known as The aim of a subsequent phase II trial was to determine the ABT888) from Abbott (Green Oaks, IL, USA), CEP9722 from efficacy of olaparib in BRCA1 or BRCA2 mutation carriers with Cephalon (Frazer, PA, USA), rucaparib (also known as AG014699, advanced refractory breast cancer (54 patients) or epithelial AG14447 or PF-01367338) from Clovis Oncology Inc. (Boulder, CO, ovarian cancer (EOC, 57 patients). For two cohorts receiving 400 USA) Pfizer, GPI-21016 from Eisai/MGI Pharma (Tokyo, Japan), MK- or 100 mg olaparib per day, the ORR was 41% and 21% 4827 from Merck (Whitehouse Station, NJ, USA), INO-1001 from (clinical benefit rate 52% and 26%) in breast cancer and the Inotek/Genentech (San Francisco, CA, USA), BMN-673 from ORR was 33% and 13% (clinical benefit rate 66% and 42%) in BioMarin Pharmaceutical Inc. (Novato, CA, USA) and E7449 from EOC, respectively.156,157 In a recent non randomized phase II Eisai Inc., (Woodcliff Lake, NJ, USA) (http://www.clinicaltrials.gov/). study, women with advanced recurrent high-grade serous or All these compounds are administered orally. We will not discuss undifferentiated ovarian carcinoma or TNBC (63 patients) were the highly deceptive clinical evaluation of iniparib (also known as stratified according to the presence of BRCA1 or BRCA2 mutations BSI-201 or SAR240550) from BiPar/Sanofi-Aventis (Paris, France) and treated with olaparib 400 mg twice daily. ORR was 41% for that was erroneously considered as a PARP inhibitor, yet lacks any BRCA-mutated patients and surprisingly high—24%—for non- specific activity at this level.154 mutated patients.158 Thus, a fraction of ovarian and breast cancers PARP inhibitors are being evaluated in oncology either as that lack BRCA mutations can respond to PARP inhibitors. monotherapeutic agents or in combination with other anticancer A randomized study has compared Olaparib monotherapy (200 therapies, seeking higher efficiency (Figure 1). In only one quarter or 400 mg bid) with pegylated liposomal doxorubicin (50 mg/m2) of clinical trials, patients were selected based on the presence of both given for 28 days, in heavily pretreated EOC with confirmed

Figure 1. Clinical trials with PARP inhibitors. The ﬁgure illustrates 93 oncological trials involving PARP listed in the US National Institute of Health registry of Clinical Trials (http://www.clinicaltrials.gov). (a) Distribution according to the study phase. (b) Distribution according to tumor types: non-speciﬁed solid tumors (30%), ovarian cancer (22%), breast cancer (17%), hematological malignancies (9%), lung cancer (3%), cerebral tumors (3%) and other cancers (16%). (c) In most trials (61%) patients were not selected according to predictive features for the response to PARP inhibition. However, a minority of trials focused on patients with BRCA mutations (18%) or clinical features of presumed ‘BRCAness’ (21%) (e.g., high serous epithelial ovarian cancers, triple-negative breast cancers). (d) Distribution according to therapeutic strategies: monotherapy (28%) combination therapy with chemotherapy (63%), targeted therapies (7%) or radiotherapy (2%). (e) Combination therapies with conventional chemotherapy included platinum compounds (42%), temozolomide (28%), topotecan (7%), anthracyclines (4%) or others (19%).

& 2014 Macmillan Publishers Limited Oncogene (2014) 3894 – 3907 Anticancer proﬁle of PARP inhibitors J Michels et al 3900 Table 2. Single agent trials involving the PARP inhibitor olaparib

Phase Patient population Trial design Results Grade 3/4 toxicity Reference

I BRCA1 or BRCA2 mutated Phase I (60 patients) with MTD 400 mg bid Lymphopenia 8% 95,155 ovarian cancer olaparib Radiological RR 28% Anemia 8% Expansion phase, 200 mg CBR according to PFI in refractory, GI symptoms 10% bid in BRCA-mutated resistant and sensitive BRCA mutated Fatigue 4% advanced ovarian cancer EOC: 23%, 45% and 69%, respectively (50 patients) II BRCA-mutated advanced Olaparib 100 (27 patients) ORR, 21% vs 41% Fatigue 15% 156 recurrent breast cancer or 400 mg (27 patients) bid Nausea 15% Vomiting 11% Anemia 11% II BRCA-mutated advanced Olaparib 100 (24 patients) ORR, 13% vs 33% Fatigue 3% 157 ovarian cancer or 400 (33 patients) mg bid CBR, 21% vs 52% Nausea 6% PFS median, 1.9 (95% CI: 1.8–3.6) Anemia 3% vs 5.8 (95% CI: 2.8–10.6) II Advanced recurrent high- Olpaparib 400 mg bid OR 41% BRCA ovarian cancer Fatigue 11% 158 grade ovarian or triple- BRCA mutations in 17/64 OR 24% non-mutated ovarian cancer Dyspnea 12% negative breast cancer EOC patients and 10/26 No OR in breast cancer (TNBC). Stratification TNBC patients PFS in EOC with BRCA mutation according to BRCA1/2 (221 daily) or without (192 daily) mutation PFS in breast cancer with BRCA mutation (109 daily) or without (54 daily) II BRCA1- or BRCA2-mutated Randomised 1:1:1 olaparib PFS 6.5, 8.8 and 7.1 months Anemia 13% 159 ovarian cancer recurrent 200 (32 patients) or 400 mg HR 0.88, P ¼ 0.66 Fatigue 9% within 12 months of prior bid (32 patients) or PLD ORR 25, 31 and 18% Nausea 6% platinum therapy 50 mg/m2 (33 patients) GCIC 34, 56 and 33% 2 Toxic death (CV accident, MDS syndrome) II Platinum-sensitive relapsed Maintenance therapy with PFS median, 8.4 months vs 4.8 Anemia 5.1%, 161 high-grade EOC with two olaparib 400 mg bid (136 months; HR 0.35, 95% CI: 0.25–0.49, Fatigue 6.6% or more platinum-based patients) versus placebo Po0.001 prior regimens (129 patients). PFS as a OS interim analysis HR, 0.94; 95% primary EP CI: 0.63–1.39; P ¼ 0.75 Abbreviations: bid, twice a day; CBR, clinical benefit rate; CI, confidence interval; CV, cerebrovascular accident; EOC, epithelial ovarian cancer; EP, endpoint; GI, gastro-intestinal; HR, hazard ratio; MDS, myelodysplastic syndrome; OS, overall survival; OR, objective response rate; ORR, overall response rate; PARP, poly(ADP-ribose) polymerase; PD, pharmacodynamics; PFS, progression-free survival; PFI, platinum-free interval; RR, response rate; TNBC, triple-negative breast cancer.

BRCA1/2 mutation and a platinum-free interval shorter than revealed the greatest clinical benefit in (germline or somatic) 12 months (reflecting platinum resistance or intermediate BRCA-mutated patients (median: 11.2 vs 4.3 m; HzR, 0.19; 95% CI: responses). The ORR of 31% for the group receiving 400 mg 0.11–0.32; Po0.0001) (Ledermann, ASCO 2013, abstract 5505).161 olaparib was consistent with previous studies.155,156 However, the One of the major challenges of regimens combining olaparib trial failed to reveal significant differences in progression-free with conventional chemotherapeutics is myelotoxicity.162 However, survival (PFS) or ORR between the groups,159 with the particularity the association of olaparib with metronomic cyclophosphamide or that PFS for pegylated liposomal doxorubicin was rather high TMZ is rather well tolerated, allowing full-dose administration of compared with prior studies in EOC.160 There are indications that chemotherapy.163,164 In chemonaı¨ve metastatic melanoma, the BRCA1/2-mutated tumors are more anthracycline-sensitive than combination therapy with TMZ seems of real interest. In a phase I non-mutated tumors,111 and this might explain the high efficacy trial, one patient benefited from a complete response that has of pegylated liposomal doxorubicin in this study, which only been lasting for 45 years post therapy.164 A phase II trail yielded enrolled BRCA1/2-mutated patients. However, the patients in this encouraging results with a response rate of 17.4% and stable study were poor responders to CDDP and hence belonged to a disease in 36% of the patients 6 months post therapy.165 subgroup of patients that reportedly exhibit low responses to Several studies have associated olaparib with targeted therapies. PARP inhibitors.155 Ongoing phase I studies are combining PARP inhibition with (i) the PARP inhibitors have been evaluated in NSCLC (Oza, ASCO 2012, PI3K inhibitor BKM120 in recurrent TNBC and high-grade EOC abstract 5001) and EOC as a maintenance therapy after platinum (NCT01623349), (ii) the VEGFR inhibitor cediranib in ovarian and treatment.161 PFS was significantly longer in patients with breast cancer (NCT01116648), (iii) bevacizumab plus carboplatin and platinum-sensitive high-grade serous EOC treated with 400 mg paclitaxel in EOC (NCT00989651), (iv) the CDK inhibitor dinaciclib bid olaparib compared to placebo after two or more platinum- plus carboplatin in advanced solid tumors (NCT01434316) or (v) the based regimens (median, 8.4 months vs 4.8 months; HzR, 0.35; anti-CD20 monoclonal antibody rituximab plus bendamustine in 95% CI, 0.25–0.49; Po0.001). The study groups were well balanced hematological malignancies (NCT01326702) (www.clinicaltrials.gov). with 22% of BRCA mutations, and subgroup analyses led to the The results of these studies are still elusive. conclusion that all patient groups profited from olaparib In synthesis, PARP inhibitors have been yielding promising independently from the BRCA mutational status, age, ethnic clinical results in several trials. When administered alone, PARP origin or the baseline response to platinum. However, an interim inhibitors are well tolerated and efficient provided that they are analysis of OS revealed no significant improvement (HzR, 0.94; used against DNA repair deficient tumors. The subgroup of 95% CI, 0.63–1.39; P ¼ 0.75). The preplanned subgroup analysis patients targeted by most ongoing clinical trials of PARP inhibitors

Oncogene (2014) 3894 – 3907 & 2014 Macmillan Publishers Limited Anticancer proﬁle of PARP inhibitors J Michels et al 3901 Table 3. Combination trials involving PARP inhibitors

PARP Other Phase Patient Trial design Results Grade 3/4 toxicity Reference inhibitor therapeutic population agents

Olaparib Dacarbazine I Advanced Dacarbazine (600–800 mg/m2) OTD Lymphopenia 15% 203 cancer, d1, plus olaparib (20–200 mg) dacarbazine 600 mg/m2 d1 Neutropenia 23% Expansion bid d1–d7, 21-day cycle (40 plus olpaparib cohort with patients) 100 mg bid d1–d7, chemonaı¨ve 21-Day cycle melanoma Tumor response: 2 PR, 8 SD, 30 PD Topotecan I Advanced solid Topotecan 0.5 mg/m2/day, 3 MTD Neutropenia 204 tumors days or 1 mg/m2/day, 3 days, plus olaparib 50, 100 or 200 mg Topotecan 1 mg/m2/day 3 days bid, d1 ¼ d21 (19 patients) olaparib 100 mg bid CDDP and I Advanced solid Olaparib 100 mg bid d1-4 plus MTD, Olaparib 100 mg/m2 d1 Neutropenia 61% 162 Gemcitabine tumors CDDP 60 mg/m2 d3 plus plus CDDP 60 mg/m2 d1 plus Lymphopenia 61% Gemcitabine 500 mg/m2 d3, Gemcitabine 500 mg/m2 d1, d8 Thrombo-cytopenia d10 (23 patients) Tumor response, 2 PR, 13 SD, 6 57% PD Bevacizumab I Advanced solid Olaparib 100 (4 patients), 200 No DLT None 205 tumors (4 patients) or 400 mg (4 patients) bid plus bevacizumab 10 mg/m2 d1 ¼ d15 Carboplatin II Platinum- 6 Â 21 daily cycles P(175 mg/ PFS median, 12.2 months vs 9.6 65% vs 57% during (Oza, and paclitaxel sensitive m2 d1) þ C(AUC4 months; HR, 0.51, 95% CI: 0.34– combination phase ASCO recurrent serous d1) þ olaparib (200 mg bid 0.77; P ¼ 0.0012; 29% vs 16% during 2012, ovarian cancer d1-10/21) (81 patients) then ORR, 64% vs 58% maintenance abstract olaparib (400 mg bid OS, immature (14% total 5001) continuous) (66 patients), events) or 6 Â 21 daily cycles P(175 mg/m2 d1) þ C(AUC6 d1) (75 patients) then no further (55 patients) Veliparib Topotecan I Advanced or Topotecan IV varying MTD Neutropenia 206 refractory solid schedules plus veliparib 10 mg topotecan 0.6 mg/m2/day plus Thrombo-cytopenia tumors, bid (24 patients) veliparib lymphoma 10 mg bid d1-5, 21-day cycle tumor response: 8 SD Cyclo I Refractory solid Metronomic dosing of MTD, cyclophosphamide Lymphopenia 163 phosphamide tumors and cyclophosphamide d1–d21 50 mg plus veliparib 60 mg lymphoid plus veliparib d1–d7 or d1–d14 daily malignancies or d1–d21 (35 patients) Tumor response: 8 PR, 6 SD Rucaparib Temozolomide I Advanced solid Escalating doses of rucaparib PID 12 mg/m2/days Thrombo-cytopenia 164 tumors plus temozolomide 100 mg/m2 MTD temozolomide 200 mg/ Neutropenia (18 patients) or 200 mg/m2 m2/days d1–d5, 28-day cycle d1–d5, 28-day cycle, for an Tumor response: 1 CR, 2 PR, 6 expansion cohort of metastatic SD malignant melanoma (15 patients) Temozolomide II Chemotherapy Rucaparib 12 mg/m2/day plus RR, 17.4% Myelo-suppression 165 naı¨ve metastatic temozolomide 200 mg/m2 TTP median, 3.5 months requiring a 25% dose melanoma d1–d5, 28-day cycle, OS median, 9.9 months reduction of (46 patients) temozolomide INO-1001 Temozolomide I Unresectable Escalating doses of INO-1001 Tumor response: 1 PR, 4 SD, Thrombo-cytopenia 207 stage III/IV plus temozolomide 200 mg/m2 TTP median, 2.2 months Neutropenia melanoma d1–d5, 28-day cycle, OS median, 17 months Asthenia (12 patients) Half live of INO-1001 mean, 5 h Abbreviations: bid, twice a day; C, carboplatin; CB, clinical benefit; CDDP, cisplatin; CR, complete response; DLT, dose limiting toxicity; EOC, epithelial ovarian cancer; MTD, maximal tolerated dose; NSCLC, non-small-cell lung cancer; ORR, overall response rate; OS, overall survival; OTD, optimal tolerated dose; P, paclitaxel; PARP, poly(ADP-ribose) polymerase; PD, progressive disease; PFS, progression-free survival; PID, PARP inhibitory dose; PI3K, PI3kinase; PLD, pegylated liposomal doxorubicin; PR, partial response; SD, stable disease; TTP, time to progression. carries BRCA mutations and platinum-sensitive tumors, because For instance, as reported by the Cancer Genome Atlas, only 50% of this subgroup showed the most promising results in early clinical high-grade EOCs are HRR deficient.167 Moreover, only 20% of development.155 Although BRCA mutations render cells sensitive TNBC may be BRCA mutated.168 These findings underscore the to platinum and PARP inhibition,166 there is no absolute need for suitable biomarkers that diagnose DNA repair defects correlation between the BRCA mutational status, platinum and predict responses to PARP inhibitors. sensitivity and clinical responses. Thus, a significant fraction of patients with BRCA non-mutated EOC demonstrated a prolonged PFS with olaparib.161 Moreover, a significant fraction of BRCA- CONCLUDING REMARKS: BIOMARKER DISCOVERY mutated platinum-resistant EOC remained sensitive to PARP As detailed in the clinical section of this review, PARP inhibitors inhibitors.155 Clinical features considered as surrogate selectively target HR-deficient tumors, including tumors from biomarkers of HRD such as unfavorable pathobiological patients bearing hereditary BRCA1 and 2 mutations. Approxi- parameters (TNBC, high-grade serous EOC) and platinum mately 15% of EOC bear mutations in BRCA1 or BRCA2.169,170 responsiveness in a BRCA-mutated context may not be Moreover, an elevated proportion of sporadic EOC (up to 55%)— sufficiently specific in the prediction for the response to PARP I. and in particular high-grade serous EOC—are HRR deficient for a

& 2014 Macmillan Publishers Limited Oncogene (2014) 3894 – 3907 Anticancer profile of PARP inhibitors J Michels et al 3902 variety of reasons, including BRCA1 promoter methylation, MRE11 caused by BRCA mutations, constituted the subgroup of breast mutation and EMSY amplification.167,171–173 Thus, reliable cancers that most frequently exhibited a low RAD51 score. biomarkers for the identification of patients with defects in the However, the use of RAD51 foci as a biomarker for HRR HR pathway would greatly facilitate the clinical development of proficiency has two major intrinsic limits that would have to be PARP inhibitor-based therapies (Figure 2). overcome in future studies. First, RAD51 foci cannot be detected at Recently, 24 EOCs were classified as HRR-competent or HRR- baseline and must be induced by DNA damage. Second, the deficient, based on the ability of primary tumor cells from ascitic expression of RAD51 is restricted to the S and G2 phases of fluid cultures to form RAD51 foci, in vitro, upon short exposure to proliferating cells, meaning that it cannot be detected in tumor AG014699.174 In this study, the amount of RAD51 foci per nucleus cells that are dormant or arrested in the G1 phase.111 It remains to inversely correlated with cytotoxicity of AG014699 in vitro. be determined whether other types of DNA damage-elicited foci, However, it remains unknown whether this functional marker of such as Mre11 foci,29 might yield a more accurate biomarker to HRR proficiency affects therapeutic outcome in vivo.174 In a cohort predict sensitivity to PARP inhibitors than RAD51 foci. of 68 sporadic breast cancer patients, reduced in vivo formation Several studies reported that PARP1 binding protein PARPBP of RAD51 foci (upon neoadjuvant anthracycline-based chemo- (also known as PARI or CI2orf48), which is overexpressed in therapy) correlated with high histological grade and intense pancreatic human carcinoma cells,175 inhibited HRR and induced proliferation rate (as determined by the Ki67 marker), yet was an increase in PAR levels.176,177 This result suggests that PARPBP predictive of pathologic complete responses.111 Remarkably, might predict both HRD and sensitivity to PARP inhibition. PARP is TNBCs, whose basal-like phenotype closely resembles that also hyperactivated in BRCA2 mutated—and thus HRR-deficient— cancer cell lines, as shown for instance for V-C8 hamster and CAPAN1 pancreatic cell lines, and this correlated with sensitivity to PARP inhibitors.96 Furthermore, the knockdown of HRR proteins (for example, RAD54, RAD52, BLM, WRN and XRCC3) in U2OS osteosarcoma cells induced HRD, PARP hyperactivation and increased sensitivity to PARP inhibitors.96 Indeed, high levels of PAR-modified proteins detectable by western blotting or immunohistochemistry predicted the cytotoxic response to PARP inhibitors in vitro and in vivo, in CDDP-resistant human cancers cells of different histological origins.59 These finding suggest that immunohistochemical methods assessing the abundance of PAR in cells (which correlates with PARP activity) may be useful to predict sensitivity to PARP inhibitors (Figure 3). The genomic instability of HRR-deficient tumors might be used as a predictive biomarker for the response to PARP inhibitors.

Figure 2. Predicitve biomarkers for PARP inhibitors. BRCA1/2 loss-of- function mutation impairs HRR and induces a PARP hyperactivation reflected by an increased abundance of poly(ADP)ribose polymers (PAR). HRD may occur without BRCA mutation in the context of ‘BRCAness’. EMSY genes and ETS fusion genes that are reported in different tumors inhibit BRCA2. 53BP1 is a DDR factor that interacts with BRCA1 upstream of HRR and NHEJ. Concomitant 53BP1 and BRCA1 loss may reinitiate HRR and subsequent resistance to PARP inhibition. Genetic instability correlates with HRD according to a HRD score. PARPBP overexpression is reported in pancreatic cancers, inducing genetic instability and PARP hyperactivation. MRE11 from the MRN protein complex (MRE11, RAD50 and NBS1) acts at the stalled replication fork upstream of the HRR. RAD51 is one of the main effectors of HRR. Aurora A is an essential actor of mitosis and cell cycle regulation and is often overexpressed in tumors. Overexpression of Aurora A inhibits RAD51 recruitment. Apart from HRR deficiencies, two main actors of the final steps of the BER pathway may be of interest. Downregulation of polymerase b and XRCC1 may induce increased PARylation. HER2 and PARP overexpression initiate an overexpression of the transcription factor NF- kB, actor of the immune response. Note that none of the depicted putative biomarkers is entirely specific for the response of tumor cells to PARP inhibitors. Aurora A, AURKA; BER, base excision repair; 53BP1, p53-binding protein-1; BRCA, breast cancer gene; CDK1, cyclin-dependent kinase 1; DDR, DNA damage response; HER2, Figure 3. PARP hyperactivation. Preclinical data report that the human epidermal growth factor receptor 2; HRD score, homologous abundance of PAR may reflect cellular DNA repair deficiencies or recombination deficiency score; HRR, homologous recombination; DDR (red box), thus constituting a universal predictive biomarker HR, homologous recombination repair; I PARP, inhibitors de PARP; for the response to PARP inhibitors. PAR is detectable by different MAPK, mitogen-activated protein kinase; MRE11, meiotic recombi- methods listed in the figure. BRCA, breast cancer gene; DDR, DNA nation protein 11; NF-kB, nuclear factor-kappa B; PAR, poly-(ADP- damage response; HRR, homologous recombination repair; MRE11, ribose); PARPBP, poly-(ADP-ribose) polymerase-binding protein; meiotic recombination protein 11; NAD þ, nicotinamide dinucleo- PLK3, polo-like kinase-3; POLB, polymerase b; XRCC1, X-ray repair tide; PAR, poly-(ADP-ribose); PI3K, phosphoinositide 3-kinase; POLB, cross-complementing 1. polymerase b; XRCC1, X-ray repair cross-complementing 1.

Oncogene (2014) 3894 – 3907 & 2014 Macmillan Publishers Limited Anticancer profile of PARP inhibitors J Michels et al 3903 A HRD score has been defined by assessing mutations of BRCA1 2 Rouleau M, Patel A, Hendzel MJ, Kaufmann SH, Poirier GG. PARP inhibition: and BRCA2, the methylation of their promoters, the level of PARP1 and beyond. Nat Rev Cancer 2010; 10: 293–301. expression of BRCA1 at the mRNA level, as well as LOH affecting 3 Krishnakumar R, Kraus WL. The PARP side of the nucleus: molecular actions, the loci coding for BRCA1 and BRCA2. This could be validated in physiological outcomes, and clinical targets. Mol Cell 2010; 39: 8–24. two independent human EOC cohorts.178–180 However, apart from 4 Herceg Z, Wang ZQ. Failure of poly(ADP-ribose) polymerase cleavage by cas- the BRCA promoter methylation, these analyses cannot detect pases leads to induction of necrosis and enhanced apoptosis. Mol Cell Biol 1999; HRR deficiencies that fall into the category of ‘BRCAness’, and 19: 5124–5133. hence must be considered as intrinsically suboptimal. 5 Galluzzi L, Vitale I, Abrams JM, Alnemri ES, Baehrecke EH, Blagosklonny MV et al. Molecular definitions of cell death subroutines: recommendations of the In conclusion, there is ample awareness of the urgent need of Nomenclature Committee on Cell Death 2012. Cell Death Differ 2012; 19: 107–120. defining reliable biomarkers that predict the clinical efficacy of PARP 6 Hassa PO, Haenni SS, Elser M, Hottiger MO. Nuclear ADP-ribosylation reactions in inhibitors. The definition and practical implementation of such mammalian cells: where are we today and where are we going? Microbiol Mol biomarkers is still in its infancy. However, it is our hope that constant Biol Rev 2006; 70: 789–829. progress in the comprehension of the complexities of DDRs will 7 Langelier MF, Planck JL, Roy S, Pascal JM. Structural basis for DNA damage- eventually lead to the development of simple and reproducible dependent poly(ADP-ribosyl)ation by human PARP-1. Science 2012; 336: tests that facilitate the clinical application of PARP inhibitors. 728–732. 8 Jankevicius G, Hassler M, Golia B, Rybin V, Zacharias M, Timinszky G et al. A family ABBREVIATIONS of macrodomain proteins reverses cellular mono-ADP-ribosylation. Nat Struct Mol Biol 2013; 20: 508–514. ART, ADP-ribosyl transferases; ATM, ataxia-telangiectasia mutated; 9 Hottiger MO, Hassa PO, Luscher B, Schuler H, Koch-Nolte F. Toward a unified BER, base excision repair; bid, twice a day; CBR, clinical benefit rate; nomenclature for mammalian ADP-ribosyltransferases. Trends Biochem Sci 2010; CDDP, cisplatin; CDK1, cyclin-dependent kinase 1; DDR, DNA 35: 208–219. damage response; DiOC6(3), 3,30-dihexyloxacarbocyanine iodide; 10 Durkacz BW, Omidiji O, Gray DA, Shall S. (ADP-ribose)n participates in DNA DLT, dose limiting toxicity; DNA-PK, DNA-dependent protein excision repair. Nature 1980; 283: 593–596. kinase; DNA-PKcs, catalytic subunit of DNA-dependent protein 11 Dantzer F, Schreiber V, Niedergang C, Trucco C, Flatter E, De La Rubia G et al. kinase; DR, direct repair; DSB, double-strand break; DSBR, double- Involvement of poly(ADP-ribose) polymerase in base excision repair. Biochimie strand break repair; EGBFr, epidermal growth factor receptor; EOC, 1999; 81: 69–75. epithelial ovarian cancer; GI, gastro-intestinal; HDAC, histone 12 Shall S, de Murcia G. Poly(ADP-ribose) polymerase-1: what have we learned from the deficient mouse model? Mutat Res 2000; 460: 1–15. deacetylase; HzR, hazard ratio; HR, homologous recombination; 13 Wang ZQ, Stingl L, Morrison C, Jantsch M, Los M, Schulze-Osthoff K et al. PARP is HRD, homologous recombination deficiency; HRR, homologous important for genomic stability but dispensable in apoptosis. Genes Dev 1997; recombinational repair; HSP90, heat-shock protein of 90 KDa; LOH, 11: 2347–2358. loss of heterozygosity; MAPK, mitogen-activated protein kinase; 14 Tsutsumi M, Masutani M, Nozaki T, Kusuoka O, Tsujiuchi T, Nakagama H et al. MGMT, O6-methylguanine DNA methyltransferase; MMR, mis- Increased susceptibility of poly(ADP-ribose) polymerase-1 knockout mice to match repair; MTD, maximum tolerated dose; NAD, nicotinamide nitrosamine carcinogenicity. Carcinogenesis 2001; 22:1–3. adenine dinucleotide; NER, nucleotide excision repair; NHEJ, non- 15 Menissier de Murcia J, Ricoul M, Tartier L, Niedergang C, Huber A, Dantzer F et al. homologous end joining; NSCLC, non-small-cell lung carcinoma; Functional interaction between PARP-1 and PARP-2 in chromosome stability and ORR, overall response rate; OS, overall survival; PAR, poly-(ADP- embryonic development in mouse. EMBO J 2003; 22: 2255–2263. 16 Farmer H, McCabe N, Lord CJ, Tutt AN, Johnson DA, Richardson TB et al. ribose); PARP, poly-(ADP-ribose) polymerase; PARPBP, PARP1- Targeting the DNA repair defect in BRCA mutant cells as a therapeutic strategy. binding protein; PARP I, PARP inhibitor; PLD, pegylated liposomal Nature 2005; 434: 917–921. doxorubicin; PD, progressive disease; PFI, platinum-free interval; 17 D’Amours D, Desnoyers S, D’Silva I, Poirier GG. Poly(ADP-ribosyl)ation reactions PFS, progression-free survival; PI, propidium iodide; PI3K, phos- in the regulation of nuclear functions. Biochem J 1999; 342(Pt 2): 249–268. phoinositide 3-kinase; PS, performance status; PSA, prostate- 18 De Vos M, Schreiber V, Dantzer F. The diverse roles and clinical relevance of specific antigen; PTEN, phosphatase and tensin homolog; RECIST, PARPs in DNA damage repair: current state of the art. Biochem Pharmacol 2012; response evaluation criteria in solid tumors; RNAi, RNA inter- 84: 137–146. ference; SNP, single-nucleotide polymorphism; SSB, single-strand 19 Luo X, Kraus WL. On PAR with PARP: cellular stress signaling through poly(ADP- break; SSBR, single-strand break repair; TMZ, temozolomide; TNBC, ribose) and PARP-1. Genes Dev 2012; 26: 417–432. 20 Campalans A, Kortulewski T, Amouroux R, Menoni H, Vermeulen W, Radicella JP. triple-negative breast cancer; XRCC1, X-ray repair cross-comple- Distinct spatiotemporal patterns and PARP dependence of XRCC1 recruitment menting 1 to single-strand break and base excision repair. Nucleic Acids Res 2013; 41: 3115–3129. 21 Strom CE, Johansson F, Uhlen M, Szigyarto CA, Erixon K, Helleday T. Poly CONFLICT OF INTEREST (ADP-ribose) polymerase (PARP) is not involved in base excision repair but The authors declare no conflict of interest. PARP inhibition traps a single-strand intermediate. Nucleic Acids Res 2011; 39: 3166–3175. ACKNOWLEDGEMENTS 22 Noren Hooten N, Kompaniez K, Barnes J, Lohani A, Evans MK. Poly(ADP-ribose) polymerase 1 (PARP-1) binds to 8-oxoguanine-DNA glycosylase (OGG1). J Biol The authors would like to thank Lorenzo Galluzzi (Universite´ Paris Descartes, Paris, Chem 2011; 286: 44679–44690. France) for assistance with manuscript preparation. The authors are supported by the 23 Vodenicharov MD, Ghodgaonkar MM, Halappanavar SS, Shah RG, Shah GM. Ligue contre le Cancer (e´quipe labelliseé), Agence National de la Recherche, AXA Mechanism of early biphasic activation of poly(ADP-ribose) polymerase-1 in Chair for Longevity Research, Association pour la Recherche sur le Cancer, response to ultraviolet B radiation. J Cell Sci 2005; 118(Pt 3): 589–599. Associazione Italiana per la Ricerca sul Cancro (AIRC), Cance´ropoˆ le Ile-de-France, 24 Pines A, Vrouwe MG, Marteijn JA, Typas D, Luijsterburg MS, Cansoy M et al. Institut National du Cancer (INCa), Fondation Bettencourt-Schueller, Fondation de PARP1 promotes nucleotide excision repair through DDB2 stabilization and France, Fondation pour la Recherche Me´dicale, the European Commission (ArtForce), recruitment of ALC1. J Cell Biol 2012; 199: 235–249. the European Research Council, the LabEx Immuno-Oncology, Action Lions ‘Vaincre 25 Robu M, Shah RG, Petitclerc N, Brind’Amour J, Kandan-Kulangara F, Shah GM. Role le Cancer’, Luxembourg, the SIRIC Stratified Oncology Cell DNA Repair and Tumor of poly(ADP-ribose) polymerase-1 in the removal of UV-induced DNA lesions by Immune Elimination (Socrate) and Cancer Research and Personalized Medicine nucleotide excision repair. Proc Natil Acad Sci USA 2013; 110: 1658–1663. (Carpem) and the Paris Alliance of Cancer Research Institutes. 26 King BS, Cooper KL, Liu KJ, Hudson LG. Poly(ADP-ribose) contributes to an association between poly(ADP-ribose) polymerase-1 and xeroderma pigmentosum complementation group A in nucleotide excision repair. J Biol Chem 2012; REFERENCES 287: 39824–39833. 1 Chambon P, Weill JD, Mandel P. Nicotinamide mononucleotide activation of new 27 Bryant HE, Petermann E, Schultz N, Jemth AS, Loseva O, Issaeva N et al. PARP is DNA-dependent polyadenylic acid synthesizing nuclear enzyme. Biochem activated at stalled forks to mediate Mre11-dependent replication restart and Biophys Res Commun 1963; 11: 39–43. recombination. EMBO J 2009; 28: 2601–2615.

& 2014 Macmillan Publishers Limited Oncogene (2014) 3894 – 3907 Anticancer profile of PARP inhibitors J Michels et al 3904 28 Yang YG, Cortes U, Patnaik S, Jasin M, Wang ZQ. Ablation of PARP-1 does not 53 Huang Q, Wu YT, Tan HL, Ong CN, Shen HM. A novel function of poly(ADP- interfere with the repair of DNA double-strand breaks, but compromises the ribose) polymerase-1 in modulation of autophagy and necrosis under oxidative reactivation of stalled replication forks. Oncogene 2004; 23: 3872–3882. stress. Cell Death Differ 2009; 16: 264–277. 29 Ying S, Hamdy FC, Helleday T. Mre11-dependent degradation of stalled 54 Zhou J, Ng S, Huang Q, Wu YT, Li Z, Yao SQ et al. AMPK mediates a pro-survival DNA replication forks is prevented by BRCA2 and PARP1. Cancer Res 2012; 72: autophagy downstream of PARP-1 activation in response to DNA alkylating 2814–2821. agents. FEBS Lett 2013; 587: 170–177. 30 Hochegger H, Dejsuphong D, Fukushima T, Morrison C, Sonoda E, Schreiber V 55 Schultz N, Lopez E, Saleh-Gohari N, Helleday T. Poly(ADP-ribose) polymerase et al. Parp-1 protects homologous recombination from interference by Ku and (PARP-1) has a controlling role in homologous recombination. Nucleic Acids Res Ligase IV in vertebrate cells. EMBO J 2006; 25: 1305–1314. 2003; 31: 4959–4964. 31 Sugimura K, Takebayashi S, Taguchi H, Takeda S, Okumura K. PARP-1 ensures 56 d’Adda di Fagagna F, Hande MP, Tong WM, Lansdorp PM, Wang ZQ, Jackson SP. regulation of replication fork progression by homologous recombination on Functions of poly(ADP-ribose) polymerase in controlling telomere length and damaged DNA. J Cell Biol 2008; 183: 1203–1212. chromosomal stability. Nat Genet 1999; 23: 76–80. 32 Couto CA, Wang HY, Green JC, Kiely R, Siddaway R, Borer C et al. PARP regulates 57 Samper E, Goytisolo FA, Menissier-de Murcia J, Gonzalez-Suarez E, Cigudosa JC, nonhomologous end joining through retention of Ku at double-strand breaks. de Murcia G et al. Normal telomere length and chromosomal end capping in J Cell Biol 2011; 194: 367–375. poly(ADP-ribose) polymerase-deficient mice and primary cells despite increased 33 Wang M, Wu W, Wu W, Rosidi B, Zhang L, Wang H et al. PARP-1 and Ku compete chromosomal instability. J Cell Biol 2001; 154: 49–60. for repair of DNA double strand breaks by distinct NHEJ pathways. Nucleic Acids 58 Morrison C, Smith GC, Stingl L, Jackson SP, Wagner EF, Wang ZQ. Genetic Res 2006; 34: 6170–6182. interaction between PARP and DNA-PK in V(D)J recombination and tumor- 34 Mansour WY, Rhein T, Dahm-Daphi J. The alternative end-joining pathway for igenesis. Nat Genet 1997; 17: 479–482. repair of DNA double-strand breaks requires PARP1 but is not dependent upon 59 Michels J, Vitale I, Galluzzi L, Adam J, Olaussen KA, Kepp O et al. Cisplatin microhomologies. Nucleic Acids Res 2010; 38: 6065–6077. Resistance Associated with PARP Hyperactivation. Cancer Res 2013; 73: 35 Spagnolo L, Barbeau J, Curtin NJ, Morris EP, Pearl LH. Visualization of a DNA-PK/ 2271–2280. PARP1 complex. Nucleic Acids Res 2012; 40: 4168–4177. 60 Ba X, Garg NJ. Signaling mechanism of poly(ADP-ribose) polymerase-1 (PARP-1) 36 Rulten SL, Fisher AE, Robert I, Zuma MC, Rouleau M, Ju L et al. PARP-3 and APLF in inflammatory diseases. Am J Pathol 2011; 178: 946–955. function together to accelerate nonhomologous end-joining. Mol Cell 2011; 41: 61 Strosznajder RP, Czubowicz K, Jesko H, Strosznajder JB. Poly(ADP-ribose) 33–45. metabolism in brain and its role in ischemia pathology. Mol Neurobiol 2010; 41: 37 Helleday T. The underlying mechanism for the PARP and BRCA synthetic 187–196. lethality: clearing up the misunderstandings. Mol Oncol 2011; 5: 387–393. 62 Rosado MM, Bennici E, Novelli F, Pioli C. Beyond dna repair,the immunological 38 Patel AG, Sarkaria JN, Kaufmann SH. Nonhomologous end joining drives poly role of parp-1 and its siblings. Immunology 2013; 139: 428–437. 63 Szabo C. Roles of poly(ADP-ribose) polymerase activation in the pathogenesis of (ADP-ribose) polymerase (PARP) inhibitor lethality in homologous recombi- diabetes mellitus and its complications. Pharmacol Res 2005; 52: 60–71. nation-deficient cells. Proc Natl Acad Sci USA 2011; 108: 3406–3411. 64 Pacher P, Szabo C. Role of poly(ADP-ribose) polymerase 1 (PARP-1) in cardio- 39 Ha HC, Snyder SH. Poly(ADP-ribose) polymerase is a mediator of necrotic cell vascular diseases: the therapeutic potential of PARP inhibitors. Cardiovas Drug death by ATP depletion. Proc Natl Acad Sci USA 1999; 96: 13978–13982. Rev 2007; 25: 235–260. 40 Zong WX, Ditsworth D, Bauer DE, Wang ZQ, Thompson CB. Alkylating DNA 65 Moroni F, Cozzi A, Chiarugi A, Formentini L, Camaioni E, Pellegrini-Giampietro DE damage stimulates a regulated form of necrotic cell death. Genes Dev 2004; 18: et al. Long-lasting neuroprotection and neurological improvement in stroke 1272–1282. models with new, potent and brain permeable inhibitors of poly(ADP-ribose) 41 Ethier C, Tardif M, Arul L, Poirier GG. PARP-1 modulation of mTOR signaling in polymerase. Br J Pharmacol 2012; 165: 1487–1500. response to a DNA alkylating agent. PLoS One 2012; 7: e47978. 66 Bai P, Canto C. The role of PARP-1 and PARP-2 enzymes in metabolic regulation 42 Jouan-Lanhouet S, Arshad MI, Piquet-Pellorce C, Martin-Chouly C, and disease. Cell Metab 2012; 16: 290–295. Le Moigne-Muller G, Van Herreweghe F et al. TRAIL induces necroptosis 67 Strosznajder JB, Czapski GA, Adamczyk A, Strosznajder RP. Poly(ADP-ribose) involving RIPK1/RIPK3-dependent PARP-1 activation. Cell Death Differ 2012; 19: polymerase-1 in amyloid beta toxicity and Alzheimer’s disease. Mol Neurobiol 2003–2014. 2012; 46: 78–84. 43 Sahaboglu A, Tanimoto N, Kaur J, Sancho-Pelluz J, Huber G, Fahl E et al. PARP1 68 Abeti R, Abramov AY, Duchen MR. Beta-amyloid activates PARP causing astro- gene knock-out increases resistance to retinal degeneration without affecting cytic metabolic failure and neuronal death. Brain 2011; 134(Pt 6): 1658–1672. retinal function. PLoS One 2010; 5: e15495. 69 Russo AL, Kwon HC, Burgan WE, Carter D, Beam K, Weizheng X et al. In vitro and 44 Virag L, Robaszkiewicz A, Vargas JM, Javier Oliver F. Poly(ADP-ribose) signaling in in vivo radiosensitization of glioblastoma cells by the poly (ADP-ribose) poly- cell death. Mol Aspects Med S0098-2997: 00008-3. merase inhibitor E7016. Clin Cancer Res 2009; 15: 607–612. 45 Yu SW, Wang H, Poitras MF, Coombs C, Bowers WJ, Federoff HJ et al. Mediation 70 Dungey FA, Loser DA, Chalmers AJ. Replication-dependent radiosensitization of of poly(ADP-ribose) polymerase-1-dependent cell death by apoptosis-inducing human glioma cells by inhibition of poly(ADP-Ribose) polymerase: mechanisms factor. Science 2002; 297: 259–263. and therapeutic potential. Int J Radiat Oncol Biol Phys 2008; 72: 1188–1197. 46 Yu SW, Andrabi SA, Wang H, Kim NS, Poirier GG, Dawson TM et al. Apoptosis- 71 Senra JM, Telfer BA, Cherry KE, McCrudden CM, Hirst DG, O’Connor MJ et al. inducing factor mediates poly(ADP-ribose) (PAR) polymer-induced cell death. Inhibition of PARP-1 by olaparib (AZD2281) increases the radiosensitivity of a Proc Natil Acad Sci USA 2006; 103: 18314–18319. lung tumor xenograft. Mol Cancer Ther 2011; 10: 1949–1958. 47 Wang Y, Kim NS, Haince JF, Kang HC, David KK, Andrabi SA et al. 72 Taverna P, Catapano CV, Citti L, Bonfanti M, D’Incalci M. Influence of Poly(ADP-ribose) (PAR) binding to apoptosis-inducing factor is critical for O6-methylguanine on DNA damage and cytotoxicity of temozolomide in L1210 PAR polymerase-1-dependent cell death (parthanatos). Science Signal 2011; 4: mouse leukemia sensitive and resistant to chloroethylnitrosoureas. Anti-Cancer ra20. Drugs 1992; 3: 401–405. 48 Kashima L, Idogawa M, Mita H, Shitashige M, Yamada T, Ogi K et al. CHFR protein 73 Sedgwick B. Repairing DNA-methylation damage. Nat Rev Mol Cell Biol 2004; 5: regulates mitotic checkpoint by targeting PARP-1 protein for ubiquitination and 148–157. degradation. J Biol Chem 2012; 287: 12975–12984. 74 Pan Q, Yang XJ, Wang HM, Dong XT, Wang W, Li Y et al. Chemoresistance to 49 Munoz-Gamez JA, Rodriguez-Vargas JM, Quiles-Perez R, Aguilar-Quesada R, temozolomide in human glioma cell line U251 is associated with increased Martin-Oliva D, de Murcia G et al. PARP-1 is involved in autophagy induced by activity of O6-methylguanine-DNA methyltransferase and can be overcome by DNA damage. Autophagy 2009; 5: 61–74. metronomic temozolomide regimen. Cell Biochem Biophys 2012; 62: 185–191. 50 Yoon JH, Ahn SG, Lee BH, Jung SH, Oh SH. Role of autophagy in chemoresis- 75 Taverna P, Liu L, Hanson AJ, Monks A, Gerson SL. Characterization of MLH1 and tance: regulation of the ATM-mediated DNA-damage signaling pathway MSH2 DNA mismatch repair proteins in cell lines of the NCI anticancer drug through activation of DNA-PKcs and PARP-1. Biochem Pharmacol 2012; 83: screen. Cancer Chemother Pharmacol 2000; 46: 507–516. 747–757. 76 Tentori L, Turriziani M, Franco D, Serafino A, Levati L, Roy R et al. Treatment with 51 Rodriguez-Vargas JM, Ruiz-Magana MJ, Ruiz-Ruiz C, Majuelos-Melguizo J, temozolomide and poly(ADP-ribose) polymerase inhibitors induces early apop- Peralta-Leal A, Rodriguez MI et al. ROS-induced DNA damage and PARP-1 are tosis and increases base excision repair gene transcripts in leukemic cells required for optimal induction of starvation-induced autophagy. Cell Res 2012; resistant to triazene compounds. Leukemia 1999; 13: 901–909. 22: 1181–1198. 77 Daniel RA, Rozanska AL, Thomas HD, Mulligan EA, Drew Y, Castelbuono DJ et al. 52 Huang Q, Shen HM. To die or to live: the dual role of poly(ADP-ribose) poly- Inhibition of poly(ADP-ribose) polymerase-1 enhances temozolomide and merase-1 in autophagy and necrosis under oxidative stress and DNA damage. topotecan activity against childhood neuroblastoma. Clin Cancer Res 2009; 15: Autophagy 2009; 5: 273–276. 1241–1249.

Oncogene (2014) 3894 – 3907 & 2014 Macmillan Publishers Limited Anticancer profile of PARP inhibitors J Michels et al 3905 78 Tentori L, Muzi A, Dorio AS, Scarsella M, Leonetti C, Shah GM et al. Pharmaco- 101 Sourisseau T, Maniotis D, McCarthy A, Tang C, Lord CJ, Ashworth A et al. logical inhibition of poly(ADP-ribose) polymerase (PARP) activity in PARP-1 Aurora-A expressing tumour cells are deficient for homology-directed DNA silenced tumour cells increases chemosensitivity to temozolomide and to a double strand-break repair and sensitive to PARP inhibition. EMBO Mol Med N3-adenine selective methylating agent. Curr Cancer Drug Targets 2010; 10: 2010; 2: 130–142. 368–383. 102 Salmena L, Carracedo A, Pandolfi PP. Tenets of PTEN tumor suppression. Cell 79 Palma JP, Wang YC, Rodriguez LE, Montgomery D, Ellis PA, Bukofzer G et al. 2008; 133: 403–414. ABT-888 confers broad in vivo activity in combination with temozolomide in 103 Shen WH, Balajee AS, Wang J, Wu H, Eng C, Pandolfi PP et al. Essential role for diverse tumors. Clin Cancer Res 2009; 15: 7277–7290. nuclear PTEN in maintaining chromosomal integrity. Cell 2007; 128: 157–170. 80 Horton TM, Jenkins G, Pati D, Zhang L, Dolan ME, Ribes-Zamora A et al. Poly(ADP- 104 Kumar A, Fernandez-Capetillo O, Carrera AC. Nuclear phosphoinositide 3-kinase ribose) polymerase inhibitor ABT-888 potentiates the cytotoxic activity of beta controls double-strand break DNA repair. Proc Natl Acad Sci USA 2010; 107: temozolomide in leukemia cells: influence of mismatch repair status and 7491–7496. O6-methylguanine-DNA methyltransferase activity. Mol Cancer Ther 2009; 8: 105 Fraser M, Zhao H, Luoto KR, Lundin C, Coackley C, Chan N et al. PTEN deletion in 2232–2242. prostate cancer cells does not associate with loss of RAD51 function: implica- 81 Huang SH, Xiong M, Chen XP, Xiao ZY, Zhao YF, Huang ZY. PJ34, an inhibitor of tions for radiotherapy and chemotherapy. Clin Cancer Res 2012; 18: 1015–1027. PARP-1, suppresses cell growth and enhances the suppressive effects of cisplatin 106 Mendes-Pereira AM, Martin SA, Brough R, McCarthy A, Taylor JR, Kim JS et al. in liver cancer cells. Oncol Rep 2008; 20: 567–572. Synthetic lethal targeting of PTEN mutant cells with PARP inhibitors. EMBO Mol 82 Gambi N, Tramontano F, Quesada P. Poly(ADPR)polymerase inhibition and Med 2009; 1: 315–322. apoptosis induction in cDDP-treated human carcinoma cell lines. Biochem 107 McEllin B, Camacho CV, Mukherjee B, Hahm B, Tomimatsu N, Bachoo RM et al. Pharmacol 2008; 75: 2356–2363. PTEN loss compromises homologous recombination repair in astrocytes: impli- 83 Cavallo F, Graziani G, Antinozzi C, Feldman DR, Houldsworth J, Bosl GJ et al. cations for glioblastoma therapy with temozolomide or poly(ADP-ribose) poly- Reduced proficiency in homologous recombination underlies the high sensitivity merase inhibitors. Cancer Res 2010; 70: 5457–5464. of embryonal carcinoma testicular germ cell tumors to Cisplatin and poly (adp- 108 Dedes KJ, Wetterskog D, Mendes-Pereira AM, Natrajan R, Lambros MB, Geyer FC ribose) polymerase inhibition. PLoS One 2012; 7: e51563. et al. PTEN deficiency in endometrioid endometrial adenocarcinomas predicts 84 Michels J, Vitale I, Senovilla L, Enot DP, Garcia P, Lissa D et al. Synergistic sensitivity to PARP inhibitors. Sci Translation Med 2010; 2: 53ra75. interaction between cisplatin and PARP inhibitors in non-small cell lung cancer. 109 Fraser M, Harding SM, Zhao H, Coackley C, Durocher D, Bristow RG. MRE11 Cell Cycle 2013; 12: 877–883. promotes AKT phosphorylation in direct response to DNA double-strand breaks. 85DrewY,MulliganEA,VongWT,ThomasHD,KahnS,KyleSet al. Therapeutic Cell Cycle 2011; 10: 2218–2232. 110 Haince JF, McDonald D, Rodrigue A, Dery U, Masson JY, Hendzel MJ et al. PARP1- potential of poly(ADP-ribose) polymerase inhibitor AG014699 in human cancers dependent kinetics of recruitment of MRE11 and NBS1 proteins to multiple DNA with mutated or methylated BRCA1 or BRCA2. JNatlCancerInst2011; 103: 334–346. damage sites. J Biol Chem 2008; 283: 1197–1208. 86 Wang JC. Cellular roles of DNA topoisomerases: a molecular perspective. Nat Rev 111 Graeser M, McCarthy A, Lord CJ, Savage K, Hills M, Salter J et al. A marker Mol Cell Biol 2002; 3: 430–440. of homologous recombination predicts pathologic complete response to 87 Liu LF, Desai SD, Li TK, Mao Y, Sun M, Sim SP. Mechanism of action of camp- neoadjuvant chemotherapy in primary breast cancer. Clin Cancer Res 2010; 16: tothecin. Ann N Y Acad Sci 2000; 922:1–10. 6159–6168. 88 Park SY, Cheng YC. Poly(ADP-ribose) polymerase-1 could facilitate the religation 112 Sultana R, Abdel-Fatah T, Abbotts R, Hawkes C, Albarakati N, Seedhouse C et al. of topoisomerase I-linked DNA inhibited by camptothecin. Cancer Res 2005; 65: Targeting XRCC1 Deficiency in Breast Cancer for Personalized Therapy. Cancer 3894–3902. Res 2013; 73: 1621–1634. 89 Thomas HD, Calabrese CR, Batey MA, Canan S, Hostomsky Z, Kyle S et al. 113 Sakata K, Someya M, Matsumoto Y, Tauchi H, Kai M, Toyota M et al. Gimeracil, an Preclinical selection of a novel poly(ADP-ribose) polymerase inhibitor for clinical inhibitor of dihydropyrimidine dehydrogenase, inhibits the early step in trial. Mol Cancer Ther 2007; 6: 945–956. homologous recombination. Cancer Sci 2011; 102: 1712–1716. 90 Delaney CA, Wang LZ, Kyle S, White AW, Calvert AH, Curtin NJ et al. Potentiation 114 Menisser-de Murcia J, Mark M, Wendling O, Wynshaw-Boris A, de Murcia G. of temozolomide and topotecan growth inhibition and cytotoxicity by novel Early embryonic lethality in PARP-1 Atm double-mutant mice suggests a poly(adenosine diphosphoribose) polymerase inhibitors in a panel of human functional synergy in cell proliferation during development. Mol Cell Biol 2001; tumor cell lines. Clin Cancer Res 2000; 6: 2860–2867. 21: 1828–1832. 91 Magan N, Isaacs RJ, Stowell KM. Treatment with the PARP-inhibitor PJ34 causes 115 Lavin MF. Ataxia-telangiectasia: from a rare disorder to a paradigm for cell enhanced doxorubicin-mediated cell death in HeLa cells. Anti-cancer Drugs 2012; signalling and cancer. Nat Rev Mol Cell Biol 2008; 9: 759–769. 23: 627–637. 116 Williamson CT, Muzik H, Turhan AG, Zamo A, O’Connor MJ, Bebb DG et al. 92 Munoz-Gamez JA, Quiles-Perez R, Ruiz-Extremera A, Martin-Alvarez AB, ATM deficiency sensitizes mantle cell lymphoma cells to poly(ADP-ribose) Sanjuan-Nunez L, Carazo A et al. Inhibition of poly (ADP-ribose) polymerase-1 polymerase-1 inhibitors. Mol Cancer Ther 2010; 9: 347–357. enhances doxorubicin activity against liver cancer cells. Cancer Lett 2011; 301: 117 Weston VJ, Oldreive CE, Skowronska A, Oscier DG, Pratt G, Dyer MJ et al. The 47–56. PARP inhibitor olaparib induces significant killing of ATM-deficient lymphoid 93 Wesierska-Gadek J, Zulehner N, Ferk F, Skladanowski A, Komina O, Maurer M. tumor cells in vitro and in vivo. Blood 2010; 116: 4578–4587. PARP inhibition potentiates the cytotoxic activity of C-1305, a selective inhibitor 118 Schaffner C, Idler I, Stilgenbauer S, Dohner H, Lichter P. Mantle cell lymphoma is of topoisomerase II, in human BRCA1-positive breast cancer cells. Biochem characterized by inactivation of the ATM gene. Proc Natl Acad Sci USA 2000; 97: Pharmacol 2012; 84: 1318–1331. 2773–2778. 94 Kaelin Jr. WG. The concept of synthetic lethality in the context of anticancer 119 Veuger SJ, Curtin NJ, Richardson CJ, Smith GC, Durkacz BW. Radiosensitization therapy. Nat Rev Cancer 2005; 5: 689–698. and DNA repair inhibition by the combined use of novel inhibitors of DNA- 95 Fong PC, Boss DS, Yap TA, Tutt A, Wu P, Mergui-Roelvink M et al. Inhibition of dependent protein kinase and poly(ADP-ribose) polymerase-1. Cancer Res 2003; poly(ADP-ribose) polymerase in tumors from BRCA mutation carriers. NEnglJ 63: 6008–6015. Med 2009; 361: 123–134. 120 Veuger SJ, Curtin NJ, Smith GC, Durkacz BW. Effects of novel inhibitors of 96 Gottipati P, Vischioni B, Schultz N, Solomons J, Bryant HE, Djureinovic T et al. poly(ADP-ribose) polymerase-1 and the DNA-dependent protein kinase on Poly(ADP-ribose) polymerase is hyperactivated in homologous recombination- enzyme activities and DNA repair. Oncogene 2004; 23: 7322–7329. defective cells. Cancer Res 2010; 70: 5389–5398. 121 Bryant HE, Helleday T. Inhibition of poly (ADP-ribose) polymerase activates ATM 97 Zaremba T, Ketzer P, Cole M, Coulthard S, Plummer ER, Curtin NJ. Poly(ADP- which is required for subsequent homologous recombination repair. Nucleic ribose) polymerase-1 polymorphisms, expression and activity in selected human Acids Res 2006; 34: 1685–1691. tumour cell lines. Br J Cancer 2009; 101: 256–262. 122 Yuan ZM, Huang Y, Ishiko T, Nakada S, Utsugisawa T, Kharbanda S et al. Reg- 98 Seki A, Coppinger JA, Jang CY, Yates JR, Fang G. Bora and the kinase Aurora a ulation of Rad51 function by c-Abl in response to DNA damage. J Biol Chem cooperatively activate the kinase Plk1 and control mitotic entry. Science 2008; 1998; 273: 3799–3802. 320: 1655–1658. 123 Cousineau I, Belmaaza A. EMSY overexpression disrupts the BRCA2/RAD51 99 Cazales M, Schmitt E, Montembault E, Dozier C, Prigent C, Ducommun B. CDC25B pathway in the DNA-damage response: implications for chromosomal instabil- phosphorylation by Aurora-A occurs at the G2/M transition and is inhibited by ity/recombination syndromes as checkpoint diseases. Mol Genet Genomics 2011; DNA damage. Cell Cycle 2005; 4: 1233–1238. 285: 325–340. 100 Yang G, Chang B, Yang F, Guo X, Cai KQ, Xiao XS et al. Aurora kinase A promotes 124 Verweij J, Casali PG, Zalcberg J, LeCesne A, Reichardt P, Blay JY et al. Progression- ovarian tumorigenesis through dysregulation of the cell cycle and suppression free survival in gastrointestinal stromal tumours with high-dose imatinib: ran- of BRCA2. Clin Cancer Res 2010; 16: 3171–3181. domised trial. Lancet 2004; 364: 1127–1134.

& 2014 Macmillan Publishers Limited Oncogene (2014) 3894 – 3907 Anticancer profile of PARP inhibitors J Michels et al 3906 125 O’Brien SG, Guilhot F, Larson RA, Gathmann I, Baccarani M, Cervantes F et al. 148 Veuger SJ, Hunter JE, Durkacz BW. Ionizing radiation-induced NF-kappaB acti- Imatinib compared with interferon and low-dose cytarabine for newly vation requires PARP-1 function to confer radioresistance. Oncogene 2009; 28: diagnosed chronic-phase chronic myeloid leukemia. N Engl J Med 2003; 348: 832–842. 994–1004. 149 Montoni A, Robu M, Pouliot E, Shah GM. Resistance to PARP-Inhibitors in Cancer 126 Choudhury A, Zhao H, Jalali F, Al Rashid S, Ran J, Supiot S et al. Targeting Therapy. Front Pharmacol 2013; 4:18. homologous recombination using imatinib results in enhanced tumor cell 150 Edwards SL, Brough R, Lord CJ, Natrajan R, Vatcheva R, Levine DA et al. Resis- chemosensitivity and radiosensitivity. Mol Cancer Ther 2009; 8: 203–213. tance to therapy caused by intragenic deletion in BRCA2. Nature 2008; 451: 127 Johnson N, Cai D, Kennedy RD, Pathania S, Arora M, Li YC et al. Cdk1 participates 1111–1115. in BRCA1-dependent S phase checkpoint control in response to DNA damage. 151 Sakai W, Swisher EM, Karlan BY, Agarwal MK, Higgins J, Friedman C et al. Mol Cell 2009; 35: 327–339. Secondary mutations as a mechanism of cisplatin resistance in BRCA2-mutated 128 Johnson N, Li YC, Walton ZE, Cheng KA, Li D, Rodig SJ et al. Compromised CDK1 cancers. Nature 2008; 451: 1116–1120. activity sensitizes BRCA-proficient cancers to PARP inhibition. Nat Med 2011; 17: 152 Bunting SF, Callen E, Wong N, Chen HT, Polato F, Gunn A et al. 53BP1 inhibits 875–882. homologous recombination in Brca1-deficient cells by blocking resection of DNA 129 Lynch TJ, Bell DW, Sordella R, Gurubhagavatula S, Okimoto RA, Brannigan BW breaks. Cell 2010; 141: 243–254. et al. Activating mutations in the epidermal growth factor receptor underlying 153 Bouwman P, Aly A, Escandell JM, Pieterse M, Bartkova J, van der Gulden H et al. responsiveness of non-small-cell lung cancer to gefitinib. N Engl J Med 2004; 350: 53BP1 loss rescues BRCA1 deficiency and is associated with triple-negative and 2129–2139. BRCA-mutated breast cancers. Nat Struct Mol Biol 2010; 17: 688–695. 130 Karapetis CS, Khambata-Ford S, Jonker DJ, O’Callaghan CJ, Tu D, Tebbutt NC 154 Patel AG, De Lorenzo SB, Flatten KS, Poirier GG, Kaufmann SH. Failure of et al. K-ras mutations and benefit from cetuximab in advanced colorectal cancer. iniparib to inhibit poly(ADP-Ribose) polymerase in vitro. Clin Cancer Res 2012; 18: N Engl J Med 2008; 359: 1757–1765. 1655–1662. 131 Li L, Wang H, Yang ES, Arteaga CL, Xia F. Erlotinib attenuates homologous 155 Fong PC, Yap TA, Boss DS, Carden CP, Mergui-Roelvink M, Gourley C et al. recombinational repair of chromosomal breaks in human breast cancer cells. Poly(ADP)-ribose polymerase inhibition: frequent durable responses in BRCA Cancer Res 2008; 68: 9141–9146. carrier ovarian cancer correlating with platinum-free interval. J Clin Oncol 2010; 132 Nowsheen S, Bonner JA, Lobuglio AF, Trummell H, Whitley AC, Dobelbower MC 28: 2512–2519. et al. Cetuximab augments cytotoxicity with poly (adp-ribose) polymerase 156 Tutt A, Robson M, Garber JE, Domchek SM, Audeh MW, Weitzel JN et al. Oral inhibition in head and neck cancer. PLoS One 2011; 6: e24148. poly(ADP-ribose) polymerase inhibitor olaparib in patients with BRCA1 or BRCA2 133 Modi S, Stopeck A, Linden H, Solit D, Chandarlapaty S, Rosen N et al. HSP90 mutations and advanced breast cancer: a proof-of-concept trial. Lancet 2010; inhibition is effective in breast cancer: a phase II trial of tanespimycin (17-AAG) 376: 235–244. 157 Audeh MW, Carmichael J, Penson RT, Friedlander M, Powell B, Bell-McGuinn KM plus trastuzumab in patients with HER2-positive metastatic breast cancer pro- et al. Oral poly(ADP-ribose) polymerase inhibitor olaparib in patients with BRCA1 gressing on trastuzumab. Clin Cancer Res 2011; 17: 5132–5139. or BRCA2 mutations and recurrent ovarian cancer: a proof-of-concept trial. 134 Hong DS, Banerji U, Tavana B, George GC, Aaron J, Kurzrock R. Targeting the Lancet 2010; 376: 245–251. molecular chaperone heat shock protein 90 (HSP90): lessons learned and future 158 Gelmon KA, Tischkowitz M, Mackay H, Swenerton K, Robidoux A, Tonkin K et al. directions. Cancer Treat Rev 2013; 39: 375–387. Olaparib in patients with recurrent high-grade serous or poorly differentiated 135 Noguchi M, Yu D, Hirayama R, Ninomiya Y, Sekine E, Kubota N et al. Inhibition of ovarian carcinoma or triple-negative breast cancer: a phase 2, multicentre, open- homologous recombination repair in irradiated tumor cells pretreated with label, non-randomised study. Lancet Oncol 2011; 12: 852–861. Hsp90 inhibitor 17-allylamino-17-demethoxygeldanamycin. Biochem Biophys Res 159 Kaye SB, Lubinski J, Matulonis U, Ang JE, Gourley C, Karlan BY et al. Phase II, Commun 2006; 351: 658–663. open-label, randomized, multicenter study comparing the efficacy and safety of 136 Dungey FA, Caldecott KW, Chalmers AJ. Enhanced radiosensitization of human olaparib, a poly (ADP-ribose) polymerase inhibitor, and pegylated liposomal glioma cells by combining inhibition of poly(ADP-ribose) polymerase with doxorubicin in patients with BRCA1 or BRCA2 mutations and recurrent ovarian inhibition of heat shock protein 90. Mol Cancer Ther 2009; 8: 2243–2254. cancer. J Clin Oncol 2012; 30: 372–379. 137 Maeta Y, Shiota G, Okano J, Murawaki Y. Effect of promoter methylation of the 160 Gordon AN, Fleagle JT, Guthrie D, Parkin DE, Gore ME, Lacave AJ. Recurrent p16 gene on phosphorylation of retinoblastoma gene product and growth of epithelial ovarian carcinoma: a randomized phase III study of pegylated lipo- hepatocellular carcinoma cells. Tumour Biol 2005; 26: 300–305. somal doxorubicin versus topotecan. J Clin Oncol 2001; 19: 3312–3322. 138 Zhang JX, Li DQ, He AR, Motwani M, Vasiliou V, Eswaran J et al. Synergistic 161 Ledermann J, Harter P, Gourley C, Friedlander M, Vergote I, Rustin G et al. inhibition of hepatocellular carcinoma growth by cotargeting chromatin Olaparib maintenance therapy in platinum-sensitive relapsed ovarian cancer. modifying enzymes and poly (ADP-ribose) polymerases. Hepatology 2012; 55: N Engl J Med 2012; 366: 1382–1392. 1840–1851. 162 Rajan A, Carter CA, Kelly RJ, Gutierrez M, Kummar S, Szabo E et al. A phase I 139 Brenner JC, Chinnaiyan AM. Translocations in epithelial cancers. Biochim Biophys combination study of olaparib with cisplatin and gemcitabine in adults with Acta 2009; 1796: 201–215. solid tumors. Clin Cancer Res 2012; 18: 2344–2351. 140 Brenner JC, Ateeq B, Li Y, Yocum AK, Cao Q, Asangani IA et al. Mechanistic 163 Kummar S, Ji J, Morgan R, Lenz HJ, Puhalla SL, Belani CP et al. A phase I study of rationale for inhibition of poly(ADP-ribose) polymerase in ETS gene fusion- veliparib in combination with metronomic cyclophosphamide in adults with positive prostate cancer. Cancer cell 2011; 19: 664–678. refractory solid tumors and lymphomas. Clin Cancer Res 2012; 18: 1726–1734. 141 Ibrahim YH, Garcia-Garcia C, Serra V, He L, Torres-Lockhart K, Prat A et al. 164 Plummer R, Jones C, Middleton M, Wilson R, Evans J, Olsen A et al. Phase I study PI3K inhibition impairs BRCA1/2 expression and sensitizes BRCA-proficient of the poly(ADP-ribose) polymerase inhibitor, AG014699, in combination with triple-negative breast cancer to PARP inhibition. Cancer Discov 2012; 2: temozolomide in patients with advanced solid tumors. Clin Cancer Res 2008; 14: 1036–1047. 7917–7923. 142 Baker KM, Wei G, Schaffner AE, Ostrowski MC. Ets-2 and components of mam- 165 Plummer R, Lorigan P, Steven N, Scott L, Middleton MR, Wilson RH et al. A phase malian SWI/SNF form a repressor complex that negatively regulates the BRCA1 II study of the potent PARP inhibitor, Rucaparib (PF-01367338, AG014699), with promoter. J Biol Chem 2003; 278: 17876–17884. temozolomide in patients with metastatic melanoma demonstrating evidence of 143 Sharrocks AD. The ETS-domain transcription factor family. Nat Rev Mol Cell Biol chemopotentiation. Cancer Chemother Pharmacol 2013; 71: 1191–1199. 2001; 2: 827–837. 166 Adams SF, Marsh EB, Elmasri W, Halberstadt S, Vandecker S, Sammel MD et al. A 144 Juvekar A, Burga LN, Hu H, Lunsford EP, Ibrahim YH, Balmana J et al. Combining high response rate to liposomal doxorubicin is seen among women with BRCA a PI3K inhibitor with a PARP inhibitor provides an effective therapy for BRCA1- mutations treated for recurrent epithelial ovarian cancer. Gynecol Oncol 2011; related breast cancer. Cancer Discover 2012; 2: 1048–1063. 123: 486–491. 145 Nowsheen S, Cooper T, Bonner JA, LoBuglio AF, Yang ES. HER2 overexpression 167 Cancer Genome Atlas Research Network. Integrated genomic analyses of ovarian renders human breast cancers sensitive to PARP inhibition independently of carcinoma. Nature 2011; 474: 609–615. any defect in homologous recombination DNA repair. Cancer Res 2012; 72: 168 Gonzalez-Angulo AM, Timms KM, Liu S, Chen H, Litton JK, Potter J et al. Incidence 4796–4806. and outcome of BRCA mutations in unselected patients with triple receptor- 146 Slamon DJ, Leyland-Jones B, Shak S, Fuchs H, Paton V, Bajamonde A et al. Use of negative breast cancer. Clin Cancer Res 2011; 17: 1082–1089. chemotherapy plus a monoclonal antibody against HER2 for metastatic breast 169 Moynahan ME, Chiu JW, Koller BH, Jasin M. Brca1 controls homology-directed cancer that overexpresses HER2. N Engl J Med 2001; 344: 783–792. DNA repair. Mol Cell 1999; 4: 511–518. 147 Merkhofer EC, Cogswell P, Baldwin AS. Her2 activates NF-kappaB and induces 170 Moynahan ME, Cui TY, Jasin M. Homology-directed dna repair, mitomycin-c invasion through the canonical pathway involving IKKalpha. Oncogene 2010; 29: resistance, and chromosome stability is restored with correction of a Brca1 1238–1248. mutation. Cancer Res 2001; 61: 4842–4850.

Oncogene (2014) 3894 – 3907 & 2014 Macmillan Publishers Limited Anticancer profile of PARP inhibitors J Michels et al 3907 171 Turner N, Tutt A, Ashworth A. Hallmarks of ’BRCAness’ in sporadic cancers. 189 Pegg AE. Mammalian O6-alkylguanine-DNA alkyltransferase: regulation and Nat Rev Cancer 2004; 4: 814–819. importance in response to alkylating carcinogenic and therapeutic agents. 172 Hughes-Davies L, Huntsman D, Ruas M, Fuks F, Bye J, Chin SF et al. EMSY Cancer Res 1990; 50: 6119–6129. links the BRCA2 pathway to sporadic breast and ovarian cancer. Cell 2003; 115: 190 Gudmundsdottir K, Ashworth A. The roles of BRCA1 and BRCA2 and associated 523–535. proteins in the maintenance of genomic stability. Oncogene 2006; 25: 173 Press JZ, De Luca A, Boyd N, Young S, Troussard A, Ridge Y et al. Ovarian 5864–5874. carcinomas with genetic and epigenetic BRCA1 loss have distinct molecular 191 Jackson SP, Bartek J. The DNA-damage response in human biology and disease. abnormalities. BMC Cancer 2008; 8: 17. Nature 2009; 461: 1071–1078. 174 Mukhopadhyay A, Elattar A, Cerbinskaite A, Wilkinson SJ, Drew Y, Kyle S et al. 192 Wooster R, Weber BL. Breast and ovarian cancer. N Engl J Med 2003; 348: Development of a functional assay for homologous recombination status in 2339–2347. primary cultures of epithelial ovarian tumor and correlation with sensitivity to 193 Cerbinskaite A, Mukhopadhyay A, Plummer ER, Curtin NJ, Edmondson RJ. poly(ADP-ribose) polymerase inhibitors. Clin Cancer Res 2010; 16: 2344–2351. Defective homologous recombination in human cancers. Cancer Treat Rev 2012; 175 Kashiwaya K, Hosokawa M, Eguchi H, Ohigashi H, Ishikawa O, Shinomura Y et al. 38: 89–100. Identification of C2orf18, termed ANT2BP (ANT2-binding protein), as one of 194 Jacinto FV, Esteller M. Mutator pathways unleashed by epigenetic silencing in the key molecules involved in pancreatic carcinogenesis. Cancer Sci 2009; 100: human cancer. Mutagenesis 2007; 22: 247–253. 457–464. 195 Bullrich F, Rasio D, Kitada S, Starostik P, Kipps T, Keating M et al. ATM mutations 176 Piao L, Nakagawa H, Ueda K, Chung S, Kashiwaya K, Eguchi H et al. C12orf48, in B-cell chronic lymphocytic leukemia. Cancer Res 1999; 59: 24–27. termed PARP-1 binding protein, enhances poly(ADP-ribose) polymerase-1 196 Marsit CJ, Liu M, Nelson HH, Posner M, Suzuki M, Kelsey KT. Inactivation of (PARP-1) activity and protects pancreatic cancer cells from DNA damage. Genes the Fanconi anemia/BRCA pathway in lung and oral cancers: implications for treatment and survival. Oncogene 2004; 23: 1000–1004. Chromosome Cancer 2011; 50: 13–24. 197 Narayan G, Arias-Pulido H, Nandula SV, Basso K, Sugirtharaj DD, Vargas H et al. 177 O’Connor KW, Dejsuphong D, Park E, Nicolae CM, Kimmelman AC, D’Andrea AD Promoter hypermethylation of FANCF: disruption of Fanconi Anemia-BRCA et al. PARI Overexpression Promotes Genomic Instability and Pancreatic pathway in cervical cancer. Cancer Res 2004; 64: 2994–2997. Tumorigenesis. Cancer Res 2013; 73: 2529–2539. 198 Taniguchi T, Tischkowitz M, Ameziane N, Hodgson SV, Mathew CG, Joenje H 178 Abkevich V, Timms KM, Hennessy BT, Potter J, Carey MS, Meyer LA et al. Patterns et al. Disruption of the Fanconi anemia-BRCA pathway in cisplatin-sensitive of genomic loss of heterozygosity predict homologous recombination repair ovarian tumors. Nat Med 2003; 9: 568–574. defects in epithelial ovarian cancer. Br J Cancer 2012; 107: 1776–1782. 199 Riffell JL, Lord CJ, Ashworth A. Tankyrase-targeted therapeutics: expanding 179 Stefansson OA, Jonasson JG, Johannsson OT, Olafsdottir K, Steinarsdottir M, opportunities in the PARP family. Nat Rev Drug Discov 2012; 11: 923–936. Valgeirsdottir S et al. Genomic profiling of breast tumours in relation to BRCA 200 Kraus WL, Lis JT. PARP goes transcription. Cell 2003; 113: 677–683. abnormalities and phenotypes. Br Cancer Res 2009; 11: R47. 201 Carbone M, Rossi MN, Cavaldesi M, Notari A, Amati P, Maione R. Poly(ADP- 180 Birkbak NJ, Wang ZC, Kim JY, Eklund AC, Li Q, Tian R et al. Telomeric allelic ribosyl)ation is implicated in the G0-G1 transition of resting cells. Oncogene imbalance indicates defective DNA repair and sensitivity to DNA-damaging 2008; 27: 6083–6092. agents. Cancer Dis 2012; 2: 366–375. 202 Aurelius J, Thoren FB, Akhiani AA, Brune M, Palmqvist L, Hansson M et al. 181 Gros L, Saparbaev MK, Laval J. Enzymology of the repair of free radicals-induced Monocytic AML cells inactivate antileukemic lymphocytes: role of NADPH oxi- DNA damage. Oncogene 2002; 21: 8905–8925. dase/gp91(phox) expression and the PARP-1/PAR pathway of apoptosis. Blood 182 Saparbaev M, Laval J. Excision of hypoxanthine from DNA containing dIMP 2012; 119: 5832–5837. residues by the Escherichia coli, yeast, rat, and human alkylpurine DNA glyco- 203 Khan OA, Gore M, Lorigan P, Stone J, Greystoke A, Burke W et al. A phase I study sylases. Proc Natl Acad Sc USA 1994; 91: 5873–5877. of the safety and tolerability of olaparib (AZD2281, KU0059436) and dacarbazine 183 Gros L, Ishchenko AA, Saparbaev M. Enzymology of repair of etheno-adducts. in patients with advanced solid tumours. Br J Cancer 2011; 104: 750–755. Mutat Res 2003; 531: 219–229. 204 Samol J, Ranson M, Scott E, Macpherson E, Carmichael J, Thomas A et al. 184 Boiteux S, Radicella JP. The human OGG1 gene: structure, functions, and its Safety and tolerability of the poly(ADP-ribose) polymerase (PARP) inhibitor, implication in the process of carcinogenesis. Arch Biochem Biophys 2000; 377: olaparib (AZD2281) in combination with topotecan for the treatment of patients 1–8. with advanced solid tumors: a phase I study. Invest New Drugs 2012; 30: 185 Ocampo MT, Chaung W, Marenstein DR, Chan MK, Altamirano A, Basu AK et al. 1493–1500. Targeted deletion of mNth1 reveals a novel DNA repair enzyme activity. Mol Cell 205 Dean E, Middleton MR, Pwint T, Swaisland H, Carmichael J, Goodege-Kunwar P Biol 2002; 22: 6111–6121. et al. Phase I study to assess the safety and tolerability of olaparib in combi- 186 Wilson 3rd DM, Barsky D. The major human abasic endonuclease: formation, nation with bevacizumab in patients with advanced solid tumours. Br J Cancer consequences and repair of abasic lesions in DNA. Mutat Res 2001; 485: 2012; 106: 468–474. 283–307. 206 Kummar S, Chen A, Ji J, Zhang Y, Reid JM, Ames M et al. Phase I study of PARP 187 Andressoo JO, Hoeijmakers JH, de Waard H. Nucleotide excision repair and its inhibitor ABT-888 in combination with topotecan in adults with refractory solid connection with cancer and ageing. Ad Exp Med Biol 2005; 570: 45–83. tumors and lymphomas. Cancer Res 2011; 71: 5626–5634. 188 Sugasawa K, Ng JM, Masutani C, Iwai S, van der Spek PJ, Eker AP et al. Xeroderma 207 Bedikian AY, Papadopoulos NE, Kim KB, Hwu WJ, Homsi J, Glass MR et al. A phase pigmentosum group C protein complex is the initiator of global genome IB trial of intravenous INO-1001 plus oral temozolomide in subjects with nucleotide excision repair. Mol Cell 1998; 2: 223–232. unresectable stage-III or IV melanoma. Cancer Invest 2009; 27: 756–763.

& 2014 Macmillan Publishers Limited Oncogene (2014) 3894 – 3907