Cancer therapy usually involves exposing the body to agents that kill cancer cells more efficiently than normal tissue cells. Such therapies must therefore exploit specific molecular and cellular features of the cancer they are aim-ing to eliminate. Most cancer cells proliferate more rapidly than their normal counterparts so most cancer drugs target the cell cycle. Cell division can be targeted directly by inhibitors of the mitotic spindle, thus preventing equal division of DNA to the two daughter cells. The growth sig-nals that result in entry into the cell cycle can be targeted by hormonal manipulation, therapeutic antibodies and drugs that inhibit growth signalling pathways. However, the most common means of targeting the cell cycle is to exploit the effect of DNA-damaging drugs. DNA damage causes cell-cycle arrest and cell death either directly or following DNA replication during the S phase of the cell cycle. Cellular attempts to replicate damaged DNA can cause increased cell killing, thus making DNA-damaging treatments more toxic to replicating cells than to non-replicating cells. However, the toxicity of DNA-damaging drugs can be reduced by the activities of several DNA repair pathways that remove lesions before they become toxic. The efficacy of DNA damage-based cancer therapy can thus be modulated by DNA repair pathways. In addi-tion, some of these pathways are inactivated in some cancer types. These two features make DNA repair mechanisms a promising target for novel cancer treatments.

DNA-damaging agents in cancer treatmentMany cancer drugs employed in the clinic have been used for several decades and are highly efficient in kill-ing proliferating cells (FIG. 1). High levels of DNA damage

cause cell-cycle arrest and cell death. Furthermore, DNA lesions that persist into the S phase of the cell cycle can obstruct replication fork progression, resulting in the formation of replication-associated DNA double-strand breaks (DSBs). DSBs are generally considered to be the most toxic of all DNA lesions1,2.

Common types of DNA damage that interfere with replication fork progression are chemical modifications (adducts) of DNA bases, which are created by reactive drugs that covalently bind DNA either directly or after being metabolized in the body. These alkylating agents are grouped in two categories: monofunctional alkylating agents with one active moiety that modifies single bases and bifunctional alkylating agents that have two reactive sites and crosslink DNA with proteins or, alternatively, crosslink two DNA bases within the same DNA strand (intra-strand crosslinks) or on opposite DNA strands (inter-strand crosslinks). Inter-strand crosslinks pose a severe block to replication forks.

DNA synthesis is sometimes targeted by inhibitors of DNA replication, such as aphidicolin, which directly inhibits DNA polymerases3, whereas the radical scav-enger hydroxyurea inhibits ribonucleotide reductase, which is required for production of the dNTPs that are used for DNA synthesis4. These two replication inhibi-tors can be regarded as DNA-damaging agents because they impair replication fork progression and cause DNA lesions, including DSBs5,6.

Antimetabolites, such as 5-fluorouracil (5FU) and thio-purines, resemble nucleotides, nucleotide precursors or cofactors required for nucleotide biosynthesis and act by inhibiting nucleotide metabolism pathways, thus

*Radiation Oncology & Biology, University of Oxford, Old Road Campus Research Building, off Roosevelt Drive, Headington, Oxford, OX3 7DQ, UK.‡Department of Genetics Microbiology and Toxicology, Stockholm University, Svante Arrhenius väg 16, S-106 91 Stockholm, Sweden.Correspondence to T.H. e-mail: thomas.helleday@rob.ox.ac.ukdoi:10.1038/nrc2342Published online7 February 2008

Alkylating agentsElectrophilic compounds that are reactive either directly or following metabolism and bind covalently to electron-rich atoms in DNA bases (that is, oxygen and nitrogen).

AntimetabolitesCompounds with similar chemical structures to nucleotide metabolites that interfere with nucleotide biosynthesis or are incorporated into DNA.

DNA repair pathways as targets for cancer therapyThomas Helleday*‡, Eva Petermann*, Cecilia Lundin*, Ben Hodgson*and Ricky A. Sharma*

Abstract | DNA repair pathways can enable tumour cells to survive DNA damage that is induced by chemotherapeutic treatments; therefore, inhibitors of specific DNA repair pathways might prove efficacious when used in combination with DNA-damaging chemotherapeutic drugs. In addition, alterations in DNA repair pathways that arise during tumour development can make some cancer cells reliant on a reduced set of DNA repair pathways for survival. There is evidence that drugs that inhibit one of these pathways in such tumours could prove useful as single-agent therapies, with the potential advantage that this approach could be selective for tumour cells and have fewer side effects.

R E V I E W S

NATUre revIewS | cancer volUMe 8 | MArCH 2008 | 193

© 2008 Nature Publishing Group

Non-homologous end joiningConnection and resealing of the two ends of a DNA double-strand break without the need for sequence homology between the ends.

Homologous recombinationA process that can copy a DNA sequence from an intact DNA molecule (often the newly synthesized sister chromatid) to repair or bypass replication lesions.

Base-excision repairA repair pathway that replaces missing or modified DNA bases, such as those produced by alkylating agents or in spontaneously degraded DNA, with the correct DNA base.

Nucleotide-excision repairA process that removes large DNA adducts or base modifications that distort the double helix and uses the opposite strand as template for repair.

AlkyltransferasesA class of enzymes that directly reverse DNA base modifications that are induced by alkylating agents by transferring the alkyl group from the base onto the protein.

depleting cells of dNTPs. They can also impair replication fork progression by becoming incorporated into the DNA7. In general, the molecular mechanisms through which anti-metabolites induce cell death are poorly understood.

Another means of interfering with replication is to exploit DNA strand breaks that arise naturally during the process of DNA synthesis. Topoisomerases are a group of enzymes that resolve torsional strains imposed on the double helix during DNA replication. They induce transient DNA breaks to relax supercoiled DNA or allow DNA strands to pass through each other8. resealing of these breaks can be prevented by the use of topoisomer-ase poisons that trap the enzymes in complex with the DNA. The nature of the damage that is caused depends on which type of enzyme is targeted. Topoisomerase II poisons cause DSBs, and topoisomerase I poisons cause positive supercoils in advance of replication forks134 and replication-associated DSBs1,2. This is a strategy commonly used for cancer treatment.

Ionizing radiation and radiomimetic agents such as bleomycin cause replication-independent DSBs that can kill non-replicating cells. In addition, such treatments can also rapidly prevent DNA replication by activation of cell-cycle checkpoints to avoid formation of toxic DNA replication lesions9.

Cell-cycle checkpoints are regulated by effector kinases, such as ataxia telangiectasia mutated (ATM) and ATM and rad3-related (ATr)10–12, which regulate the activities of downstream checkpoint proteins, such as checkpoint kinases 1 (CHK1) and 2 (CHK2). Defects in DNA dam-age checkpoint pathways result in sensitivity to a range of anticancer treatments, for example, loss of ATM results in sensitivity to ionizing radiation13. The triggering of these checkpoints and subsequent DNA repair activity largely determines the efficacy of anticancer drugs in causing tumour regression.

Efficient repair of chemotherapy lesionsDirect DSBs are mainly repaired by non-homologous end  joining14, whereas replication-associated DSBs are repaired by homologous recombination (Hr)15 and related replication repair pathways. DNA adducts, such as those created by alkylating agents, may be excised and repaired before they are confronted by the replica-tion machinery. This is achieved by base-excision repair, excising a single damaged DNA base or a short strand containing the damaged base16 or nucleotide-excision repair (Ner), which excises a single-stranded DNA molecule of approximately 24–30 base pairs contain-ing the DNA lesion17,18. Damaged DNA can also be repaired without removal of the damaged base, in a process that directly reverses the DNA alkylation19. The o-6-methylguanine-DNA methyltransferase (MGMT) is an alkyltransferase that removes alkylations on the o6 position of guanine produced by anticancer drugs such as temozolomide20, and the DNA dioxygenases ABH2 (also known as AlKBH2) and ABH3 (also known as AlKBH3) revert 1-methyladenine and 3-methyl-cytosine back to adenine or cytosine respectively21. The repair of alkylated lesions is thought to be quick, with the majority of lesions probably being repaired within one hour. If the lesions are removed before the initia-tion of replication, the efficiency of alkylating agents in killing the tumour is significantly reduced. Thus, modu-lation of DNA repair clearly influences the efficacy of alkylating agents, and resistance to alkylating agents is often explained by increased expression and/or activity of DNA repair proteins.

whereas most DNA repair pathways mediate resistance to DNA damage, mismatch repair is actually required for the toxicity of several anticancer drugs (FIG. 1). This has been explained by the ‘futile repair cycle’ model in which mismatch repair removes the newly inserted intact base instead of the damaged base, triggering subsequent rounds of futile repair which can be deleterious to the cell22. It is also possible that mis-match repair might have an important role in triggering checkpoint signalling and apoptosis, which might medi-ate increased cytotoxicity23. It has been established that a defect in mismatch repair is associated with resistance to many, but not all, DNA-damaging anticancer agents, such as monofunctional alkylating agents and cispla-tin, as well as the antimetabolite 6-thioguanine7,22,24. It should be noted that mismatch repair acts directly at replication forks and can therefore not prevent them from encountering damage.

Collapse of replication forks during DNA synthesis can be avoided by bypassing DNA lesions in a process called translesion synthesis25,26. This process is carried out by switching the regular polymerases, ε and δ, which are responsible for leading and lagging strand synthesis, respectively27,28, to polymerases with different substrate specificities, thus enabling them to bypass different types of damaged bases29.

once a replication fork stalls or collapses, other repair pathways are required to permit resumption of replication. Collapsed replication forks are recognized by the checkpoint machinery, which will in turn trigger

At a glance

•SeveralcancerchemotherapydrugsworkbyproducingexcessiveDNAdamagethatcausescelldeathdirectlyorfollowingDNAreplication.SurvivalispromotedthroughrepairoftheselesionsbyanumberofDNArepairpathways.

•TheefficacyofanticancerdrugsishighlyinfluencedbycellularDNArepaircapacity.InhibitorsofDNArepairincreasetheefficacyofDNA-damaginganticancerdrugsinpreclinicalmodels.Small-moleculeinhibitorsofDNArepairhavebeencombinedwithconventionalchemotherapydrugsinseveralphaseI–IIclinicaltrials.

•TumourdevelopmentcanbeassociatedwithperturbedDNAdamageresponseandrepairpathways.ThisperturbationresultsinreducedDNArepaircapacityandincreasedgeneticinstabilityintumourcells.DefectsinoneDNArepairpathwaycanbecompensatedforbyotherpathways.SuchcompensatingpathwayscanbeidentifiedinsyntheticlethalityscreensandthenspecificallytargetedfortreatmentofDNArepair-defectivetumours.

•EvidenceindicatesthatinhibitorsofDNArepairpathwayscanworkassingleagentsforthetargetedtreatmentofDNArepair-defectivecancers.ThishypothesisiscurrentlybeingtestedinphaseIItrialsinwhichpatientswithbreastorovariancancersthataredefectiveinhomologousrecombinationarebeingtreatedwithapoly(ADP-ribose)polymeraseinhibitor.

•Tumoursoftenexhibitreplicationstressasaconsequenceofoncogene-inducedgrowthsignalsorhypoxia-inducedreplicationarrest.WeproposethatDNArepairinhibitorscouldbeusedtopreventtherepairofreplicationlesionspresentintumourcellsandconvertthemintofatalreplicationlesionsthatspecificallykillcancercells.

R E V I E W S

194 | MArCH 2008 | volUMe 8 www.nature.com/reviews/cancer

© 2008 Nature Publishing Group

Nature Reviews | Cancer

OH•

Toxic lesions

Single-strand breaks

Double-strand breaks

Base damage

Includes mismatch repair-mediated toxicity

No

Major repair pathways

HR

HR

HR

HR

HR

YesAlkylsulphonates

Nitrosourea compounds

Temozolomide

b Monofunctional alkylators

c Bifunctional alkylators

Nitrogen mustard

Mitomycin C

CisplatinNER

NER

AT

Yes

?

Cancer treatment

Yes BER

BER

BER

d Antimetabolites

5-Fluorouracil (5FU)

Thiopurines

Folate analogues

NHEJ

NHEJ

NHEJ

SSBR

SSBR

e Topoisomerase inhibitors

Camptothecins

Etoposide (VP16)

f Replication inhibitors

Aphidicolin

Hydroxyurea

a Radiotherapy and radiomimetics

Ionizing radiation

Bleomycin

TLS

TLS

No

No

FA

FA

FA

FA

O2G

RecQ

RecQ

RecQ

RecQ

Double-strand breaks

Replication lesions

Double-strand breaks

Single-strand breaks

Replication lesions

Double-strand breaks

DNA crosslinks

Bulky adducts

Replication lesions

Base damage

Bulky adducts

Replication lesions

ENDO

ENDO

ENDO

ENDO

Replication lesions

Uncharacterized

Base damage

Pt

H3CS

O–

OO

H3CS

OCH3

OO

CH3

N

N

NH

N

SH

H2N

H3NPt

NH3

Cl Cl

DNA dioxygenasesA class of enzymes that directly reverse DNA base methylations through an oxidation mechanism. The human DNA dioxygenase ABH2 is thought to act at replication forks.

Mismatch repairA process that acts during DNA replication to correct base-pairing errors made by the DNA polymerases.

Figure 1 |OverviewofDnarepairpathwaysinvolvedinrepairingtoxicDnalesionsformedbycancertreatments.The DNA-damaging agents that are used in cancer treatment induce a diverse spectrum of toxic DNA lesions. These lesions are recognized by a variety of DNA repair pathways which are lesion-specific but are complementary in some respects. a | Ionizing radiation and radiomimetic drugs induce double-strand breaks (DSBs) that are predominantly repaired by non-homologous end joining (NHEJ). b, c | Monofunctional alkylators (b)and bifunctional alkylators (c) induce DNA base modifications, which interfere with DNA synthesis. Lesions produced by some alkylators are processed into toxic lesions in a mismatch repair-dependent manner. The base-excision repair (BER) and nucleotide-excision repair (NER) pathways are, together with alkyltransferases (ATs), major repair pathways, whereas other repair pathways repair toxic replication lesions, such as those produced by interstrand crosslinks. d | Antimetabolites interfere with nucleotide metabolism and DNA synthesis, causing replication lesions which have not yet been characterized. Mismatch repair mediates the toxicity of some antimetabolites (for example, thiopurines). The repair pathways involved in repair of antimetabolite-induced lesions are, apart from BER, poorly characterized. e | Topoisomerase poisons trap topoisomerase I or II in transient cleavage complexes with DNA, thus creating DNA breaks and interfering with replication. f | Replication inhibitors induce replication fork stalling and collapse, resulting in indirect DSBs. The relative contributions of the major repair pathways to the respective types of DNA damage outlined are indicated by the sizes of the boxes. This is based on the extent of sensitivity of repair-deficient cells to anticancer drugs in each category. ENDO, endonuclease-mediated repair; FA, Fanconi anaemia repair pathway; HR, homologous recombination; O2G, DNA dioxygenases; RecQ, RecQ-mediated repair; SSBR, DNA single-strand break repair; TLS, translesion synthesis.

R E V I E W S

NATUre revIewS | cancer volUMe 8 | MArCH 2008 | 195

© 2008 Nature Publishing Group

Translesion synthesisA mechanism during DNA replication in which the standard DNA polymerase is temporarily exchanged for a specialized polymerase that can synthesize DNA across base damage on the template strand.

Fanconi anaemia repair pathwayProteins of this pathway, including BRCA2, are mutated in the hereditary disorder Fanconi anaemia (FA), resulting in hypersensitivity to inter-strand crosslinks. Evidence suggests that the FA pathway promotes the repair of stalled replication forks, possibly by activating HR and facilitating ATR- and ATM-dependent checkpoint signalling.

Endonuclease-mediated repairA repair pathway that introduces a DNA single-strand break in a DNA structure to facilitate continuous repair.

RecQ-mediated repairA repair pathway that unwinds complex DNA structure to facilitate repair.

Therapeutic indexThe therapeutic index describes the ability of a treatment strategy to kill cancer cells in preference to cells in normal tissues.

cell-cycle arrest12, DNA repair30 or cell death through apoptosis or senescence31–33. Although we know little of the nature of replication lesions, there is an increas-ing body of information concerning pathways that repair them. Hr has a central role in the repair of most replication lesions formed by anticancer drugs5,6,15,34. There are several ways by which Hr is used to restart replication. The sequence identity between two newly synthesised DNA molecules can be used to restart rep-lication behind the replication block. Also, recombina-tion can be used to bypass DNA lesions in a process called template switching35. other repair pathways active at replication forks involve the Fanconi anaemia (FA) repair pathway36, endonuclease-mediated repair, such as that mediated by the MUS81 endonuclease37, and RecQ-mediated repair, which involves DNA helicases such as Bloom syndrome (BlM)38, werner syndrome (wrN)39,40 and other members of the recQ family of helicases41. Several of the proteins in these pathways have been found to be directly linked with Hr42 or the resolution of recombination products such as Holliday junctions38,40,43,44. However, cells that are defective in these pathways show distinct differences from Hr-defective cells, indicating that they represent different but overlapping repair pathways45.

Cells defective in a specific DNA repair pathway exhibit sensitivity to drugs producing DNA lesions that are normally repaired by that pathway. This sensitivity has been exploited to isolate hamster cell lines showing hypersensitivity to cancer treatments such as etoposide, mitomycin C and ionizing radiation, and also to allow cloning of genes involved in DNA repair46. The DNA repair pathways involved in the repair of damage caused by various anticancer agents are summarized in FIG. 1. These DNA repair pathways can have increased activity in tumour cells, resulting in resistance to chemotherapeu-tic drugs47. Importantly, these DNA repair pathways can be inhibited pharmacologically to potentially increase the efficacy or specificity of anticancer agents.

DNA repair inhibitors in combination therapyThe basic understanding of DNA repair, from the principles of the DNA lesions created to the pathways that are capable of repairing these lesions, has increased considerably during recent years. This knowledge permits rational combination of cytotoxic agents and inhibitors of DNA repair to enhance tumour-cell killing.

Understanding lesions and repair pathways enables the use of DNA-repair inhibitors to exploit tumour defects or cancer-specific replication lesions (BOX 1). Several inhibitors of DNA repair have been developed as clinical agents and clinical trials are ongoing (TABlE 1).

Sensitizers to alkylating agents. Despite the adverse side effects caused by alkylating agents on bone marrow and other normal tissues, drugs such as cyclophosphamide, ifosfamide, chlorambucil, melphalan and dacarbazine remain some of the most commonly prescribed chemotherapies in adults and children with various solid and haematological malignancies, particularly in combination with anthracyclines and steroids in multi-agent regimens. More recently, a DNA alkylator and methylator developed in the 1980s, temozolomide (an oral prodrug that crosses the blood–brain barrier), has changed clinical practice in the treatment of high-grade gliomas in adults and children48,49.

A class of agents currently being tested in clinical trials in combination with temozolomide therapy con-sists of the pseudosubstrates for MGMT. The lead com-pounds in this class have been O6-benzylguanine50 and lomeguatrib (AstraZeneca, london, UK); the latter is also known as O6-(4-bromothenyl)guanine or PaTrin-2. resistance to o6-alkylating agents can be overcome in preclinical models by depletion of MGMT51 and a rela-tionship exists between MGMT activity and resistance to chloroethylating nitrosoureas and methylating agents in tumour cells grown in vitro and in xenograft mod-els52. O6-Benzylguanine and lomeguatrib have recently been tested in phase I–II clinical trials and biologically effective doses have been established for both agents53. In the case of O6-benzylguanine, a phase I clinical trial not only defined the maximum tolerated dose (MTD) of a single dose of temozolomide when combined with O6-benzylguanine, but it also determined the dose of O6-benzylguanine that was effective in producing complete depletion of tumour MGMT activity for 48 h54. However, results obtained so far indicate that, when used in combination with cytotoxic chemo-therapy, myelosuppression is significantly enhanced by O6-benzylguanine and lomeguatrib, necessitating significant reductions in the doses of alkylating agents prescribed from those used in standard chemotherapy55. on account of this lack of selectivity for malignant tis-sue versus normal bone marrow, no improvement in the therapeutic index has so far been demonstrated in clinical trials of these agents.

The combination of temozolomide with inhibitors of poly(ADP-ribose) polymerase 1 (PArP1) is currently under investigation in several clinical trials (TABlE 1). PArP1 is required for the efficient base-excision repair of apurinic sites, the intermediate DNA lesions induced by temozolomide, and inhibition of PArP1 retards the repair of these lesions. It was originally shown in 1980 by Durkacz and Shall that specific inhibitors of PArP can prevent the rejoining of DNA strand breaks that are caused by dimethyl sulphate, resulting in demonstrable cytotoxicity in vitro56. However, in the absence of PArP1 inhibition, apurinic sites are not generally regarded

Box 1 | Strategies using DNA repair inhibitors in cancer treatment

•DNArepairinhibitorscanbeusedincombinationwithaDNA-damaginganticanceragent.ThiswillincreasetheefficiencyofthecancertreatmentbyinhibitingDNArepair-mediatedremovaloftoxicDNAlesions.

•DNArepairinhibitorscanbeusedasmonotherapytoselectivelykillcancercellswithadefectintheDNAdamageresponseorDNArepair.SyntheticlethalinteractionsbetweenatumourdefectandDNArepairpathwaycanbeusedtoidentifynoveltreatmentstrategies.

•Inthefuture,DNArepairinhibitorscouldpotentiallybeusedtoamplifytumour-specificreplicationlesionstoselectivelykillcancercells.Thisstrategywouldtakeadvantageofcancer-specificreplicationstresscausedbyoncogenesorthetumourmicroenvironment.

R E V I E W S

196 | MArCH 2008 | volUMe 8 www.nature.com/reviews/cancer

© 2008 Nature Publishing Group

as major contributors to the cytotoxicity induced by temozolomide. The success of the treatment rationale adopted by current clinical trials of GPI-21016 (Guilford Pharmaceuticals, Baltimore, Maryland, USA), INo-1001 (Inotek Pharmaceuticals, Beverly, Massachusetts, USA) and AG-014699 (Pfizer GrD, la Jolla, California, USA) depends on the overall biological role of and necessity for PArP in cancer cells that are trying to repair the DNA damage induced by temozolomide. The role of PArP in DNA repair has not been fully elucidated and additional roles for PArP in DNA damage signalling or repair might explain the increased toxicity of combina-tion treatments57,58.

Platinum chemotherapies. Cisplatin, carboplatin and oxaliplatin have become three of the most commonly prescribed chemotherapeutic drugs used to treat solid cancers in patients59. Platinum resistance, either intrin-sic or acquired during cyclical treatment, is a major clinical problem as additional agents that can be added to therapy in order to circumvent tumour resistance do not currently exist.

Platinum chemotherapy is now being tested with PArP inhibition in two clinical trials (TABlE 1). The rationale for combining PArP inhibition with platinum chemotherapy is based on preclinical observations that PArP inhibitors preferentially kill neoplastic cells and induce complete or partial regression of a wide variety of human tumour xenografts in nude mice treated with platinum chemo-

therapy60–62. For example, ABT-888 (Abbott laboratories, Chicago, Illinois, USA), a potent inhibitor of PArP1 and PArP2, has been shown to potentiate the regression of established tumours induced by temozolomide, cisplatin, carboplatin or cyclophosphamide therapy in rodent ortho-topic and xenograft models63. However, as stated above, the full function of PArP in DNA repair is not clear57,58, and as a result the biological mechanisms of chemosensiti-zation of cancer cells to platinum chemotherapy by PArP inhibition remain to be resolved. Interestingly, as a mono-therapy in these preclinical models, ABT-888 exhibits no significant anticancer activity.

DNA demethylating agents such as 2′-deoxy-5-azacytidine (decitabine; MGI Pharma, Bloomington, Minnesota, USA) have been combined with cisplatin or carboplatin to reverse drug resistance caused by the silencing of mismatch repair genes by hypermethylation. The toxicity of agents such as cisplatin depends at least partly on functional mismatch repair (FIG. 1). Preclinical data from xenograft models and translational studies from drug-resistant cells and tissues that are mismatch repair-deficient owing to MLH1 hypermethylation have demonstrated increased chemotherapeutic efficacy when a demethylating agent is combined with platinum chemotherapy64,65. Decitabine is currently being tested in combination with carboplatin in a phase II clinical trial in patients with ovarian cancer (see Decitabine and Carboplatin in relapsed ovarian Cancer in Further information).

Table 1 | Ongoing clinical trials of small-molecule inhibitors of the DNA damage response and related signalling pathways

agent(company)

Targetmoleculeorpathway

Monotherapyorcombinationtherapyagent(s)

Phaseofclinicaltrialplanned,ongoingorrecentlycompleted

reference

AZD-2281 (Astra Zeneca)

PARP Gemcitabine Carboplatin Topotecan Monotherapy

Phase I Phase I Phase I Phase II

http://www.astrazenecaclinicaltrials.com/article/525925.aspx

AG014699 (Pfizer)

PARP Temozolomide antibody Temozolomide

Phase I

Phase II

http://www.eddn.org/clinicalTr_caResUK.html

INO-1001 (Inotek) PARP Temozolomide Phase I http://www.inotekcorp.com/content/ino-1001.asp

BSI-201 (Bipar Sciences)

PARP Monotherapy Gemcitabine–carboplatin

Phase I Phase II

http://www.biparsciences.com/BSI201.html

ABT-888 (Abbott Laboratories)

PARP Temozolomide Phase I NCT00526617

TRC-102 (Tracon Pharma)

BER Temozolomide Pemetrexed

Phase I Phase I planned

http://www.traconpharma.com/content/pipeline_overview.html

Lomeguatrib (Astra Zeneca)

MGMT Irinotecan Temozolomide

Phase I Phase II

http://www.astrazenecaclinicaltrials.com/article/525925.aspx

O6-Benzylguanine MGMT Temozolomide Phase II http://clinicalstudies.info.nih.gov/cgi/detail.cgi?A_2006-C-0089.html

Decitabine (MGI Pharma121)

Hypermethylation of mismatch repair genes

Epirubicin, cisplatin, 5-fluorouracil Carboplatin

Phase I

Phase II

http://pfsearch.ukcrn.org.uk/StudyDetail.aspx?TopicID=&StudyID=2192

XL844 (Exilixis122) CHK1, CHK2 Gemcitabine Phase I planned http://www.exelixis.com/pipeline_xl844.shtmlThe recent or current stage of development of clinical trials is indicated for individual compounds, which are grouped by molecular target. BER, base excision repair; CHK, checkpoint kinase; MGMT, O-6-methylguanine methyltransferase; PARP, poly(ADP-ribose) polymerase.

R E V I E W S

NATUre revIewS | cancer volUMe 8 | MArCH 2008 | 197

© 2008 Nature Publishing Group

Synthetic lethalityA genetic phenomenon in which the combination of two otherwise non-lethal mutations results in an inviable cell. Synthetic lethal phenotypes are indicative of an interaction between the products of the two mutant genes within the cell.

It has been shown in preclinical models that the major cisplatin intra-strand crosslinks formed in DNA are rec-ognized and repaired by the mammalian Ner pathway66. one biological predictor of clinical outcome for patients with completely resected non-small-cell lung cancer receiving adjuvant cisplatin-based chemotherapy is expression of the excision repair cross-complementation group 1 (erCC1) protein67. An immunohistochemical study of 761 operative specimens of non-small-cell lung cancer tissue taken before a proportion of the patients in the study received adjuvant chemotherapy showed that a significant benefit from cisplatin-based adjuvant chemo-therapy was associated with the absence of erCC1 in 56% of the samples studied. The results suggested that patients with completely resected non-small-cell lung cancer and erCC1-negative tumours appeared to benefit from adjuvant cisplatin-based chemotherapy, whereas patients with erCC1-positive tumours did not. However, it should be noted that, among the patients who did not receive adjuvant chemotherapy, those with erCC1-positive tumours survived significantly longer than those with erCC1-negative tumours. It is therefore conceivable from the results of this study that erCC1 might not represent a biomarker specific to cisplatin-based chemotherapy, but it might represent a marker of prognosis and response to combination chemotherapy (not necessarily specific to cisplatin) in a poor-prognosis subset of patients, as has recently been demonstrated for p53 overexpression in a similar patient subgroup68.

Although it has been suggested that erCC1 is a potential anticancer drug target, the protein has no known enzymatic activity, making the means of regu-lating its activity harder to decipher. Pharmacological modulation of erCC1 might therefore be less desirable than a greater understanding of the clinical relevance of protein–protein interactions within the Ner machin-ery or between erCC1 and other repair pathways, as potential targets for improving the efficacy of platinum chemotherapies. For example, UCN-01 (7-hydroxy-staurosporine) is an anticancer agent that potentiates cisplatin and carboplatin toxicity (shown in preclinical models and a phase I clinical trial, respectively), which has been shown to interfere with the interaction of erCC1 and another component of the Ner pathway, xeroderma pigmentosum A (XPA), in vitro69.

Attenuators of checkpoint signalling. An alternative approach to modulating DNA repair activity and poten-tially improving the therapeutic index is to interfere with cell cycle checkpoint signalling. Xl844 (eXel-9844) is a small-molecule inhibitor of CHK1 and CHK2. This drug causes inhibition of cell-cycle arrest, progressive DNA damage, inhibition of DNA repair and, ultimately, tumour cell apoptosis in cancer cells grown in vitro70, although the outcome of inhibiting CHK1 and CHK2 in general can vary in a cell type-dependent manner. Although concerns have been raised about the potential toxicity to normal cells of an approach which inhibits both CHK1 and CHK2 kinases71, preclinical data have suggested that intermittent dosing with Xl844 in combination with the deoxycytidine analogue gemcitabine

is well tolerated by female athymic nude mice used as a xenograft model70. Xl844 is currently being tested in a clinical trial (NCT00475917) in combination with gemcitabine, which normally causes cell-cycle arrest and apoptosis by its incorporation into DNA.

Two kinases from the phosphatidylinositol 3-kinase (PI3K)-related protein kinase (PIKKs) family, ATM and ATr, are central to cellular responses to DSBs. when activated, ATM and ATr phosphorylate a multitude of proteins, which initiate a cascade that induces cell-cycle arrest and facilitates DNA repair. An inhibitor of ATM kinase activity, KU55933 (AstraZeneca), is currently in preclinical development. The rationale behind the clinical use of ATM inhibitors depends on the premise that ATM inhibition should improve the therapeutic index by hypersensitizing tumour cells to agents that cause DSBs. Here again, the feasibility of inhibiting ATM kinase in patients will depend on the level of toxicity that such agents cause in normal tissues when they are used in combination with ionizing radiation or cytotoxic chemotherapy.

Radiosensitizers. DNA-dependent protein kinase (DNAPK, also known as PrKDC), a member of the PIKK family, is important for DSB repair by non-homologous end joining following ionizing radiation72. Cells defective in DNAPK are highly sensitive to ionizing radiation73 indicating that inhibition of DNAPK might sensitize tumours to radiation treatment. wortmannin, a known inhibitor of PIKKs at low nanomolar concentrations, has antiproliferative effects and is a radiosensitizer in preclinical models74, but is unsuitable for clinical appli-cations owing to its inherent toxicity and instability in cells75. other small molecules that reversibly inhibit DNAPK kinase activity at low micromolar concentra-tions have been synthesized. They are currently in tran-sition from late preclinical development to early clinical trials. In particular, NU7441 (REF. 76) has been shown to sensitize cells to topoisomerase II poisons and can also function as a radiosensitizer in a manner consistent with DNAPK inhibition77.

DNA repair inhibitors as monotherapyAs discussed above, most of the current small-molecule inhibitors of DNA repair have so far been tested in early clinical trials as sensitizers of tumour cells to chemo-therapy. However, DNA damage also occurs spontane-ously in cells in the absence of treatments and DNA repair pathways are therefore essential for the survival of untreated cells. As several cancers are defective in DNA damage response and repair pathways (TABlE 2), the concept of synthetic lethal interactions can be used to advocate the use of DNA repair inhibitors as monothera-pies (FIG. 2). DNA repair is an ideal target for inhibition in cancer cells as the inhibitors should be exclusively toxic to cancer cells and therefore be associated with minimal side effects for patients (BOX 2).

Indeed, DNA repair inhibitors have been demon-strated to work as single agents in patients with DNA repair-defective tumours. The most notable example so far is the use of PArP inhibitors to treat patients with

R E V I E W S

198 | MArCH 2008 | volUMe 8 www.nature.com/reviews/cancer

© 2008 Nature Publishing Group

Table 2 | Synthetic lethal interactions of DNA repair and cell cycle checkpoint genes implicated in cancer by pathway

Protein Syndrome Primarycancers Biomarker SLIs S. cerevisiae homologue

SLIsinS. cerevisiae (n)

HR

BRCA1, BRCA2 – Breast, ovarian112, 135 – PARP1 (REFS 78,79) – –

RAD54B – NHL, colon cancer113 – PARP1 (REF. 87) rdh54 –

RAD51B (RAD51L1) – Lipoma, uterine leiomyoma114

– PARP1 (REF. 87) rad51 31

CtIP (RBBP8) – Colorectal cancer115 – – sae2 5

NHEJ

MRE11 Ataxia telangiectasia-like disorder

Colorectal cancer108 IR sensitivity – mre11 40

LIG4 LIG4 Leukaemia116 IR sensitivity – lig4 –

Artemis (DCLRE1C) Omenn Lymphoma117 – – pso2 –

MMR

MSH2, MLH1, MSH6, PMS1, PMS2, MLH3

– Hereditary nonpolyposis colorectal cancer118–120

Microsatellite instability

– msh2, mlh1, msh6, pms1, mlh3

3

RecQ

BLM Bloom Various121 Increased SCE – sgs1 43

WRN Werner Various122 Increased telomeric SCE

– sgs1 43

RECQL4 Rothmund–Thomson

Skin basal and sqamous cell, osteosarcoma122

– – sgs1 43

Damage signalling

ATM Ataxia-telangiectasia

Leukaemia123 IR sensitivity PARP1 (REFS 124,125), FANC89

tel1 2

NBS1 Nijmegen breakage

Various126 IR sensitivity – xrs2 30

p53 Li–Fraumeni Various127 – – - –

CHK2 Li–Fraumeni Various128 – – dun1/rad53 37

NER

XPA, XPC, DDB1, ERCC4, ERCC5, POLH

XP Skin cancers129 UV sensitivity – rad14, rad4, rad1, rad2, rad30

11

ERCC2, ERCC3 XP, Cockayne, trichothio-dystrophy

Skin cancers129 – – rad25, rad3 8

ERCC1 Cerebro-oculo-facio-skeletal

Squamous cell carcinoma, head and neck130

– – rad10 6

FA

FANCA, FANCC, FANCD2, FANCE, FANCG

Fanconi anaemia

Various131 Impaired FANCD2 ubiqutination131

ATM89 – –

FANCB, FANCF, FANCL Fanconi anaemia

Various131 Impaired FANCD2 ubiqutination131

– – –

BRIP1 Fanconi anaemia

Various131 – – – –

FANCM Fanconi anaemia

Various131 Impaired FANCD2 ubiqutination131

– mph1 1

BER

POLB – Various132 – – pol4 –

FEN1 – Various133 – – rad27 104Synthetic lethalities observed in mammalian cells and the number of synthetic lethal interactions (SLIs) for their Saccharomyces cervisiae homologues are shown. The genes showing SLIs with DNA repair genes can potentially be used as targets for novel drugs that then may selectively kill cancer cells in monotherapy. A complete list of the SLIs in S. cervisiae can be found in Supplementary information S1 (table). ATM, ataxia telangiectasia mutated; BER, base-excision repair; CHK, checkpoint kinase; ERCC, excision repair cross-complementation; FA, Fanconi anaemia-mediated repair; FANC, Fanconi anemia, complementation group; FEN1, flap structure-specific endonuclease 1; HR, homologous recombination; IR, infrared; MLH, mutL homologue; MMR, mismatch repair; MSH, mutS homologue; NER, nucleotide-excision repair; NHEJ, non-homologous end joining; PARP, poly(ADP-ribose) polymerase; RecQ, RecQ-mediated repair; SCE, sister chromatid exchange; UV, ultraviolet; XP, xeroderma pigmentosum.

R E V I E W S

NATUre revIewS | cancer volUMe 8 | MArCH 2008 | 199

© 2008 Nature Publishing Group

Nature Reviews | Cancer

Targeted monotherapy Cancer mutation

Replication lesion

Survival

a Initial cancer mutation

Lethal

b Inhibitor of Top3 in cancer cells

Survival

c Drug-resistant mutation d Second therapy in resistant cells

Lethal

SAE2mre11

SGS1

TOP3 mre11

SGS1

TOP3

sgs1

Gene function Drug-resistant mutation

Monotherapy Mutation Monotherapymre11 mre11SAE2

sgs1

inherited breast and ovarian cancers that lack wild-type copies of the BRCA1 or BRCA2 genes78,79. BRCA1- and BRCA2-mutated cells are defective in Hr repair80,81 and show extensive replication-associated lesions82,83. These recombination-defective cells are 100–1,000-fold more sensitive to PArP inhibitors than are the heterozygote or the wild-type cell lines, indicating their potential to be exploited as specific treatments of BRCA1- or BRCA2-defective tumours78,79. one explana-tion for this sensitivity is that PArP inhibitors induce single-strand breaks that can result in DSBs as a result of stalled replication forks. Such lesions would normally be repaired by Hr, but this is prohibited in BRCA1- or BRCA2-deficient cancer cells79,84–86. PArP activity is also required for the actions of CHFr (checkpoint protein with forkhead-associated and ring finger domains)58 and the reactivation of stalled replication forks57. These func-tions might also explain the hypersensitivity to PArP inhibitors of recombination-defective cells. Translation of these observations has led to phase II clinical trials of monotherapy using the PArP inhibitor AZD2281 (AstraZeneca) currently recruiting patients with breast and ovarian cancer who harbour mutations in BRCA1 or BRCA2 genes. A separate phase II trial with the PArP1 inhibitor AG014699 (Pfizer GrD) is due to open to recruitment of known carriers of BRCA1 or BRCA2 mutations with locally advanced or metastatic cancers of the breast or ovaries. It should be noted however that not all patients with mutations in BRCA1 or BRCA2 respond to PArP inhibitors as a monotherapy. The reasons for this are currently under investigation. Cells that are defective in recombination-related proteins other than BrCA1 or BrCA2, such as rAD51, rAD54, XrCC2,

XrCC3, DSS1 (also known as SHFM1), replication protein A1 (rPA1), ATM, ATr, CHK1, CHK2, NBS1 (also known as NBN) and components of the Fanconi anaemia repair pathway, also show increased sensitivity to PArP inhibition79,87,88. This suggests that PArP inhibi-tors might also be suitable in treating several types of tumours with defects in Hr.

Another synthetic lethal interaction has recently been discovered between the Fanconi anaemia repair pathway and ATM. Two pancreatic tumour lines defec-tive in the Fanconi anaemia pathway were more sensitive to the ATM inhibitor KU-55933 than isogenic control cells89. This finding provides a rationale for studying ATM inhibitors in the treatment of Fanconi anaemia repair-defective pancreatic cancer.

Finding new synthetic lethal DNA repair-based partner-ships. As mutations in checkpoint and DNA repair pathways are associated with cancer (TABlE 2), it should be straightforward to exploit DNA repair inhibitors for the treatment of tumours carrying specific defects in DNA repair or damage signalling (FIG. 2). Genome-wide screens in model organisms, such as the bud-ding yeast Saccharomyces cerevisiae, have helped to identify protein interaction networks, which serve to elucidate protein functions within highly complex cel-lular processes. Such knowledge can prove extremely useful in identifying suitable targets for monotherapy, but perhaps more significantly those that could be used in combination therapies. we have compiled a list of reported mutations in DNA damage response genes found in human tumours and homologous synthetic lethal interactions that have been demonstrated in

Figure 2 |SyntheticlethalinteractionstoidentifymoleculartargetsforinhibitorsofDnarepair.A mutation in a single tumour suppressor gene can result in genetic instability that can accelerate tumour progression90. Here, we use synthetic lethal interactions in Saccharomyces cerevisiae to illustrate a hypothetical scenario. a | We start with an Mre11 mutation that is found in mismatch repair-defective tumours108. b | The yeast mre11 is synthetic lethal with topoisomerase III (Top��)109. If the synthetic lethal interaction is conserved from yeast to man, inhibitors of Top3 would, in theory, kill Mre11-mutated tumour cells. Top�-mutated yeast cells are viable, albeit with genetic instability110, and thus Top3 inhibitors may specifically kill Mre11-mutated cancer cells while being tolerated in normal cells. c | Replication repair networks are complex and an additional mutation might override synthetic lethal interactions. In yeast, an additional mutation in sgs1, upstream of the hypothetically targeted TOP� pathway, reverses the synthetic lethal phenotype109. Thus, in this scenario, an additional tumour mutation in the human homologue of SGS1, BLM, might result in resistance to Top3 inhibitors. d | sae2 is in turn synthetic lethal with sgs1 in yeast111. If this synthetic lethal interaction is conserved, the human SAE2 homologue CtIP (also known as retinoblastoma binding protein 8 (RBBP8)) might, in theory, then be targeted as a second-line therapy in Top3 inhibitor-resistant tumours that gained resistance owing to a BLM mutation. In summary, understanding of DNA repair networks is important to identify pathways to target and to circumvent resistance mechanisms.

R E V I E W S

200 | MArCH 2008 | volUMe 8 www.nature.com/reviews/cancer

© 2008 Nature Publishing Group

BiomarkersA molecule or substance whose detection indicates a particular disease state or treatment response.

HypoxiaA subnormal concentration of oxygen. In cancer tissue, hypoxia is often the result of abnormal vasculature.

S. cerevisiae (Supplementary information S1 (table)). This information could be used to identify proteins encoded by the human homologues of genes showing synthetic lethal interactions that may represent good targets for specific treatment of cancers carrying muta-tions in DNA repair genes. Unfortunately, the poorly defined biochemical properties of these proteins are still a significant barrier to the exploitation of these proteins as potential targets for drug intervention in the immediate future.

A further limitation to exploiting these pathways is the lack of reliable biomarkers to aid the selection of patients that might respond to such treatments. This is particularly important as cancers in patients that have no

DNA repair defect will not respond. The most reliable biomarkers will probably be those that identify loss of specific post-translational modifications present in the DNA damage response and repair pathways, or those that indicate increased activity of the targeted pathway.

Future directionsCurrent chemotherapy regimens demonstrate that pro-duction of excessive replication lesions is a successful means of killing cancer cells. It has been observed that tumour cells themselves exhibit a high level of endogenous replication lesions that result in genetic instability32,33,90. In theory, DNA repair inhibitors could be used to impair the repair of replication lesions that are present in tumour cells and convert them into fatal replication lesions that specifically kill cancer cells. For example, there is evidence that the increased expression or activity of oncogenes can induce replication stress32,33,91,92. In such tumours it might be possible to use DNA repair inhibitors to make existing cancer-specific replication lesions more toxic, resulting in fatal replication lesions selectively killing oncogene-expressing cancer cells32,33,91–972 (BOX 3).

More advanced cancers are exposed to another source of replication stress owing to the tumour microenviron-ment. Tumours are often hypoxic, which has been shown to disrupt DNA synthesis98. These conditions cause repli-cation lesions that activate the ATM- and ATr-mediated checkpoint response99–101. Furthermore, DNA repair is downregulated in hypoxic cells102, which cumulatively contributes to the genetic instability observed in these cells103,104. In hypoxic cancer cells, therefore, inhibitors of the checkpoint response could prove to be more efficient than inhibitors of DNA repair105.

ConclusionsThe potential of inhibitors of DNA repair in the future of cancer therapy is starting to become apparent. Although selective inhibition of DNA repair pathways can be used to enhance current chemotherapy, the most attractive use of DNA repair inhibitors might be in using cancer defects for selective cell killing. DNA

Box 2 | Advantages and limitations using DNA repair inhibitors as single agents in treatment of cancers

•DNArepairinhibitorscanexploittumour-specificdefectsincheckpointsignallingandDNArepairtoconvertendogenousDNAlesionsintofatalreplicationlesionsthatselectivelykilltumourcells.

•Ageneralproblemfornovelcancertherapiesisthattheyarenotsufficientlyefficientatkillingcancercellstoreplacecurrentcytotoxictherapies.Asaresult,someenzymeinhibitors(thatdonottargetDNArepair)havefailedinlateclinicaltrialdevelopmentowingtoagenerallackofanti-tumourefficacy.InhibitionofDNArepairamplifiestoxicreplication-associatedDNAlesionsthatdirectlyresultincelldeath.DNArepairinhibitorsmightthereforebehighlyefficientatkillingtumours.

•Cross-talkbetweenDNArepairpathwaysinnormalcellsminimizessideeffectsfromtheinhibitionofasingleDNArepairpathway.

•TumourinactivationofDNAdamagesignallingandDNArepairareoftenrelativelyearlyeventsduringcarcinogenesis,suggestingthatnon-toxicDNArepairinhibitorsmaybeconsideredinthetreatmentofpatientswithpre-malignantorearlyneoplasticlesions,suchasthosearisinginpatientswithinheritedmutationsinBRCA1orBRCA2genes,orintestinallesionsinpatientswithdefectsinMLH1andMSH2genesortheirmethylationstatus.

•Cross-talkbetweenDNArepairpathways,suchasthecross-talkbetweenhomologousrecombinationandnon-homologousendjoiningintherepairofDNAdouble-strandbreaksorthecross-talkbetweenbase-excisionrepair,alkyltransferasesandDNAdioxygenasesintherepairofalkylationdamage,islikelytoresultinacquisitionofresistancemechanismsintumours,whichisalimitationforkillingmoreadvancedtumours.

Box 3 | Targetting oncogene-induced replication stress

Thetransformationofnormalcellstoacancerousstateisofteninitiatedbytheactivationofoncogenes,whichprovideexcessivegrowthsignals106.Itwasrecentlyshownthatsuchoncogeneactivationcaninducereplication-associatedDNAlesions32,33,91,92,includingincreasedreplicationoriginfiring,re-replicationandimpairedforkprogression32,107.Thesereplication-associatedlesionstriggeracell-cyclecheckpointresponsethatstopstheproliferationofpre-cancerouscellsearlyduringneoplastictransformation32,33,91–93.Thisinvolvescheckpointproteins,suchasp53andCHK2,thatcaninduceapoptosisorsenescence94,95,

Genesencodingproteinsinthecheckpointpathwaysareoftenmutatedduringcancerdevelopment96,allowingcellstoevadecheckpointsandcontinuetoproliferate.Takentogether,theseobservationssuggestthatakeyfeatureofcancercellsthatoverexpressoncogenesmightbehigherlevelsofendogenousreplication-associatedlesionsthanthosepresentinnormalcells.Thisinturnwouldcontributetogeneticinstability97thatwouldassistthetumourtoinducethegeneticchangesrequiredforcontinuingtransformationtomalignantphenotypes90.Moreimportantly,thereplicationlesionscausedbyoncogeneactivationhavebeenfoundtoresemblethoseproducedbyanticancertreatments32,and,likethelatter,theselesionswouldrequirerepairforthecancercellstosurvive.WethereforeproposethatfutureDNArepairinhibitorsshouldbeusedtomakeexistingcancer-specificreplicationlesionsmoretoxic,resultinginfatalreplicationlesionsselectivelykillingoncogene-expressingcancercells.Thenatureofthereplicationlesionsthatareproducedfollowingchemotherapyoroncogene-inducedstressarepoorlyunderstood.Althoughseveralreplicationrepairpathwayshavebeenidentified,wecurrentlyhavelittleinformationregardingtheircomplexinterplay.Indeed,moreintensebasicresearchisrequiredinthisareatoidentifynovelanticancertargets.

R E V I E W S

NATUre revIewS | cancer volUMe 8 | MArCH 2008 | 201

© 2008 Nature Publishing Group

1.  Hsiang, Y. H., Lihou, M. G. & Liu, L. F. Arrest of replication forks by drug-stabilized topoisomerase I — DNA cleavable complexes as a mechanism of cell killing by camptothecin. Cancer Res. 49, 5077–5082 (1989).

2.  Markovits, J. et al. Topoisomerase II-mediated DNA breaks and cytotoxicity in relation to cell proliferation and the cell cycle in NIH 3T3 fibroblasts and L1210 leukemia cells. Cancer Res. 47, 2050–2055 (1987).

3.  Ikegami, S. et al. Aphidicolin prevents mitotic cell division by interfering with the activity of DNA polymerase-α. Nature 275, 458–460 (1978).

4.  Bianchi, V., Pontis, E. & Reichard, P. Changes of deoxyribonucleoside triphosphate pools induced by hydroxyurea and their relation to DNA synthesis. J. Biol. Chem. 261, 16037–16042 (1986).

5.  Lundin, C. et al. Different roles for nonhomologous end joining and homologous recombination following replication arrest in mammalian cells. Mol. Cell Biol. 22, 5869–5878 (2002).

6.  Saintigny, Y. et al. Characterization of homologous recombination induced by replication inhibition in mammalian cells. EMBO J. 20, 3861–3870 (2001).

7.  Swann, P. F. et al. Role of postreplicative DNA mismatch repair in the cytotoxic action of thioguanine. Science 273, 1109–1111 (1996).

8.  Wang, J. C. Cellular roles of DNA topoisomerases: a molecular perspective. Nature Rev. Mol. Cell Biol. 3, 430–440 (2002).

9.  Painter, R. B. & Cleaver, J. E. Repair replication in HeLa cells after large doses of x-irradiation. Nature 216, 369–370 (1967).

10.  Canman, C. E. et al. Activation of the ATM kinase by ionizing radiation and phosphorylation of p53. Science 281, 1677–1679 (1998).

11.  Falck, J., Mailand, N., Syljuasen, R. G., Bartek, J. & Lukas, J. The ATM–Chk2–Cdc25A checkpoint pathway guards against radioresistant DNA synthesis. Nature 410, 842–847 (2001).This paper describes the molecular mechanism by which ATM rapidly inactivates DNA synthesis following ionizing radiation.

12.  Cliby, W. A. et al. Overexpression of a kinase-inactive ATR protein causes sensitivity to DNA-damaging agents and defects in cell cycle checkpoints. EMBO J. 17, 159–169 (1998).

13.  Taylor, A. M. et al. Ataxia telangiectasia: a human mutation with abnormal radiation sensitivity. Nature 258, 427–429 (1975).

14.  Sargent, R. G., Brenneman, M. A. & Wilson, J. H. Repair of site-specific double-strand breaks in a mammalian chromosome by homologous and illegitimate recombination. Mol. Cell Biol. 17, 267–277 (1997).

15.  Arnaudeau, C., Lundin, C. & Helleday, T. DNA double-strand breaks associated with replication forks are predominantly repaired by homologous recombination involving an exchange mechanism in mammalian cells. J. Mol. Biol. 307, 1235–1245 (2001).

16.  Sharma, R. A. & Dianov, G. L. Targeting base excision repair to improve cancer therapies. Mol. Aspects Med. 28, 345–374 (2007).

17.  Huang, J. C., Svoboda, D. L., Reardon, J. T. & Sancar, A. Human nucleotide excision nuclease removes thymine dimers from DNA by incising the 22nd phosphodiester bond 5′ and the 6th phosphodiester bond 3′ to the photodimer. Proc. Natl Acad. Sci. USA 89, 3664–3668 (1992).

18.  Sugasawa, K. et al. A multistep damage recognition mechanism for global genomic nucleotide excision repair. Genes Dev. 15, 507–521 (2001).

19.  Sedgwick, B. Repairing DNA-methylation damage. Nature Rev. Mol. Cell Biol. 5, 148–157 (2004).

20.  Lindahl, T., Demple, B. & Robins, P. Suicide inactivation of the E. coli O6-methylguanine-DNA methyltransferase. EMBO J. 1, 1359–1363 (1982).

21.  Duncan, T. et al. Reversal of DNA alkylation damage by two human dioxygenases. Proc. Natl Acad. Sci. USA 99, 16660–16665 (2002).

22.  Karran, P. & Marinus, M. G. Mismatch correction at O6-methylguanine residues in E. coli DNA. Nature 296, 868–869 (1982).

23.  Yoshioka, K., Yoshioka, Y. & Hsieh, P. ATR kinase activation mediated by MutSα and MutLα in response to cytotoxic O6-methylguanine adducts. Mol. Cell 22, 501–510 (2006).

24.  Fram, R. J., Cusick, P. S., Wilson, J. M. & Marinus, M. G. Mismatch repair of cis-diamminedichloroplatinum(II)-induced DNA damage. Mol. Pharmacol. 28, 51–55 (1985).

25.  Masutani, C., Kusumoto, R., Iwai, S. & Hanaoka, F. Mechanisms of accurate translesion synthesis by human DNA polymerase η. EMBO J. 19, 3100–3109 (2000).

26.  Vaisman, A., Masutani, C., Hanaoka, F. & Chaney, S. G. Efficient translesion replication past oxaliplatin and cisplatin GpG adducts by human DNA polymerase η. Biochemistry 39, 4575–4580 (2000).

27.  Fukui, T. et al. Distinct roles of DNA polymerases δ and ε at the replication fork in Xenopus egg extracts. Genes Cells 9, 179–191 (2004).

28.  Pursell, Z. F., Isoz, I., Lundstrom, E. B., Johansson, E. & Kunkel, T. A. Yeast DNA polymerase ε participates in leading-strand DNA replication. Science 317, 127–130 (2007).

29.  Lehmann, A. R. Translesion synthesis in mammalian cells. Exp. Cell Res. 312, 2673–2676 (2006).

30.  Sorensen, C. S. et al. The cell-cycle checkpoint kinase Chk1 is required for mammalian homologous recombination repair. Nature Cell Biol. 7, 195–201 (2005).

31.  Kastan, M. B. & Bartek, J. Cell-cycle checkpoints and cancer. Nature 432, 316–323 (2004).

32.  Bartkova, J. et al. Oncogene-induced senescence is part of the tumorigenesis barrier imposed by DNA damage checkpoints. Nature 444, 633–637 (2006).

33.  Di Micco, R. et al. Oncogene-induced senescence is a DNA damage response triggered by DNA hyper-replication. Nature 444, 638–642 (2006).

References 32 and 33 show that oncogenes induce replication lesions in premalignant cancer cells that in turn activate senescence as a tumour barrier.

34.  Arnaudeau, C., Tenorio Miranda, E., Jenssen, D. & Helleday, T. Inhibition of DNA synthesis is a potent mechanism by which cytostatic drugs induce homologous recombination in mammalian cells. Mutat. Res. DNA Repair 461, 221–228 (2000).

35.  Helleday, T., Lo, J., van Gent, D. C. & Engelward, B. P. DNA double-strand break repair: From mechanistic understanding to cancer treatment. DNA Repair (Amst.) 6, 923–935 (2007).

36.  Patel, K. J. & Joenje, H. Fanconi anemia and DNA replication repair. DNA Repair (Amst.) 6, 885–890 (2007).

37.  Hanada, K. et al. The structure-specific endonuclease Mus81 contributes to replication restart by generating double-strand DNA breaks. Nature Struct. Mol. Biol. 14, 1096–1104 (2007).

38.  Karow, J. K., Constantinou, A., Li, J. L., West, S. C. & Hickson, I. D. The Bloom’s syndrome gene product promotes branch migration of holliday junctions. Proc. Natl Acad. Sci. USA 97, 6504–6508 (2000).

39.  Lebel, M., Spillare, E. A., Harris, C. C. & Leder, P. The Werner syndrome gene product co-purifies with the DNA replication complex and interacts with PCNA and topoisomerase I. J. Biol. Chem. 274, 37795–37799 (1999).

40.  Constantinou, A. et al. Werner’s syndrome protein (WRN) migrates Holliday junctions and co-localizes with RPA upon replication arrest. EMBO Rep. 1, 80–84 (2000).

41.  Wu, L. & Hickson, I. D. DNA helicases required for homologous recombination and repair of damaged replication forks. Annu. Rev. Genet. 40, 279–306 (2006).

42.  Niedzwiedz, W. et al. The Fanconi anaemia gene FANCC promotes homologous recombination and error-prone DNA repair. Mol. Cell 15, 607–620 (2004).

43.  Wu, L. & Hickson, I. D. The Bloom’s syndrome helicase suppresses crossing over during homologous recombination. Nature 426, 870–874 (2003).

44.  Chen, X. B. et al. Human Mus81-associated endonuclease cleaves Holliday junctions in vitro. Mol. Cell 8, 1117–1127 (2001).

45.  Hinz, J. M., Nham, P. B., Urbin, S. S., Jones, I. M. & Thompson, L. H. Disparate contributions of the Fanconi anemia pathway and homologous recombination in preventing spontaneous mutagenesis. Nucleic Acids Res. 35, 3733–3740 (2007).

46.  Thompson, L. H. Strategies for cloning mammalian DNA repair genes. Methods Mol. Biol. 113, 57–85 (1999).

47.  Chabner, B. A. & Roberts, T. G. Jr. Chemotherapy and the war on cancer. Nature Rev. Cancer 5, 65–72 (2005).

48.  Stevens, M. F. et al. Antitumor activity and pharmacokinetics in mice of 8-carbamoyl-3-methyl-

repair inhibitors that exploit tumour mutations in DNA repair pathways to convert spontaneous DNA lesions into fatal replication lesions might represent the most direct means of finding selective treatments for certain cancers. This type of therapy is highly advan-tageous when compared with current chemotherapy as it is likely to produce minimal side effects while resulting in highly toxic lesions that should actively trigger cell death in cancer cells. A potential limita-tion of this approach is that it is likely to be confined to DNA repair-defective tumours and that resistance mechanisms might develop.

A more challenging treatment strategy is the inhibi-tion of the repair of tumour-specific replication lesions and conversion of these into fatal lesions. replication stress appears to be present in a majority of tumours during at least one stage of carcinogenesis. Thus, the conversion of replication stress into fatal replication

lesions could potentially be used to target a wide range of tumours. As we are still unaware of the exact nature of the replication lesions that are induced by many traditional chemotherapies, there is still con-siderable work to be done in characterizing tumour-specific lesions to target cancers. Basic research into understanding the nature of toxic replication lesions, as well as obtaining a more complete picture of all DNA repair pathways and their interplay, is crucial for the future of DNA repair inhibitors as single agents in cancer therapy.

In summary, cancer cells are potentially exposed to unusually high levels of replication stress and endo-genous DNA damage during cancer development. A future challenge will be to identify and characterize forms of replication lesions occurring during carcino-genesis and neoplastic progression that may be exploited for selective therapy.

R E V I E W S

202 | MArCH 2008 | volUMe 8 www.nature.com/reviews/cancer

© 2008 Nature Publishing Group

imidazo[5,1-d]-1,2,3,5-tetrazin-4(3H)-one (CCRG 81045; M & B 39831), a novel drug with potential as an alternative to dacarbazine. Cancer Res. 47, 5846–5852 (1987).

49.  Stupp, R. et al. Radiotherapy plus concomitant and adjuvant temozolomide for glioblastoma. N. Engl. J. Med. 352, 987–996 (2005).References 48 and 49 demonstrate the timescale required from drug discovery in the laboratory to the demonstration of improved survival in patients in a randomized phase III clinical trial, in this case for temozolomide as a radiosensitizer.

50.  Dolan, M. E. & Pegg, A. E. O6-Benzylguanine and its role in chemotherapy. Clin. Cancer Res. 3, 837–847 (1997).

51.  Gerson, S. L., Berger, N. A., Arce, C., Petzold, S. J. & Willson, J. K. Modulation of nitrosourea resistance in human colon cancer by O6-methylguanine. Biochem. Pharmacol. 43, 1101–1107 (1992).

52.  Middleton, M. R. & Margison, G. P. Improvement of chemotherapy efficacy by inactivation of a DNA-repair pathway. Lancet Oncol. 4, 37–44 (2003).

53.  Ranson, M. et al. Lomeguatrib, a potent inhibitor of O6-alkylguanine-DNA-alkyltransferase: phase I safety, pharmacodynamic, and pharmacokinetic trial and evaluation in combination with temozolomide in patients with advanced solid tumors. Clin. Cancer Res. 12, 1577–1584 (2006).

54.  Quinn, J. A. et al. Phase I trial of temozolomide plus O6-benzylguanine for patients with recurrent or progressive malignant glioma. J. Clin. Oncol. 23, 7178–7187 (2005).

55.  Khan, O. & Middleton, M. R. The therapeutic potential of O6-alkylguanine DNA alkyltransferase inhibitors. Expert Opin. Investig. Drugs 16, 1573–1584 (2007).

56.  Durkacz, B. W., Omidiji, O., Gray, D. A. & Shall, S. (ADP-ribose)n participates in DNA excision repair. Nature 283, 593–596 (1980).This study was first to demonstrate that PARP inhibitors can be used to increase the toxicity of DNA-damaging agents.

57.  Yang, Y. G., Cortes, U., Patnaik, S., Jasin, M. & Wang, Z. Q. Ablation of PARP-1 does not interfere with the repair of DNA double-strand breaks, but compromises the reactivation of stalled replication forks. Oncogene 23, 3872–3882 (2004).

58.  Ahel, I. et al. Poly(ADP-ribose)-binding zinc finger motifs in DNA repair/checkpoint proteins. Nature 451, 81–85 (2008).

59.  Rosenberg, B., VanCamp, L., Trosko, J. E. & Mansour, V. H. Platinum compounds: a new class of potent antitumour agents. Nature 222, 385–386 (1969).

60.  Kubo, S. et al. Participation of poly(ADP-ribose) polymerase in the drug sensitivity in human lung cancer cell lines. J. Cancer Res. Clin. Oncol. 118, 244–248 (1992).

61.  Miknyoczki, S. J. et al. Chemopotentiation of temozolomide, irinotecan, and cisplatin activity by CEP-6800, a poly(ADP-ribose) polymerase inhibitor. Mol. Cancer Ther. 2, 371–382 (2003).

62.  Robins, H. I. et al. Phase I trial of intravenous thymidine and carboplatin in patients with advanced cancer. J. Clin. Oncol. 17, 2922–2931 (1999).

63.  Donawho, C. K. et al. ABT-888, an orally active poly(ADP-ribose) polymerase inhibitor that potentiates DNA-damaging agents in preclinical tumor models. Clin. Cancer Res. 13, 2728–2737 (2007).

64.  Gifford, G., Paul, J., Vasey, P. A., Kaye, S. B. & Brown, R. The acquisition of hMLH1 methylation in plasma DNA after chemotherapy predicts poor survival for ovarian cancer patients. Clin. Cancer Res. 10, 4420–4426 (2004).

65.  Plumb, J. A., Strathdee, G., Sludden, J., Kaye, S. B. & Brown, R. Reversal of drug resistance in human tumor xenografts by 2′-deoxy-5-azacytidine-induced demethylation of the hMLH1 gene promoter. Cancer Res. 60, 6039–6044 (2000).

66.  Rabik, C. A. & Dolan, M. E. Molecular mechanisms of resistance and toxicity associated with platinating agents. Cancer Treat. Rev. 33, 9–23 (2007).

67.  Olaussen, K. A. et al. DNA repair by ERCC1 in non-small-cell lung cancer and cisplatin-based adjuvant chemotherapy. N. Engl. J. Med. 355, 983–991 (2006).

68.  Tsao, M. S. et al. Prognostic and predictive importance of p53 and RAS for adjuvant chemotherapy in non small-cell lung cancer. J. Clin. Oncol. 25, 5240–5247 (2007).

69.  Jiang, H. & Yang, L. Y. Cell cycle checkpoint abrogator UCN-01 inhibits DNA repair: association with

attenuation of the interaction of XPA and ERCC1 nucleotide excision repair proteins. Cancer Res. 59, 4529–4534 (1999).

70.  Matthews, D. J. et al. Pharmacological abrogation of S-phase checkpoint enhances the anti-tumor activity of gemcitabine in vivo. Cell Cycle 6, 104–110 (2007).

71.  Syljuasen, R. G. et al. Inhibition of human Chk1 causes increased initiation of DNA replication, phosphorylation of ATR targets, and DNA breakage. Mol. Cell Biol. 25, 3553–3562 (2005).

72.  Hartley, K. O. et al. DNA-dependent protein kinase catalytic subunit: a relative of phosphatidylinositol 3-kinase and the ataxia telangiectasia gene product. Cell 82, 849–856 (1995).

73.  Blunt, T. et al. Defective DNA-dependent protein kinase activity is linked to VDJ recombination and DNA repair defects associated with the murine scid mutation. Cell 80, 813–823 (1995).This paper identifies DNAPK and shows that it can be used as a target to increase toxicity following ionizing radiation.

74.  Monfar, M. et al. Activation of pp70/85 S6 kinases in interleukin-2-responsive lymphoid cells is mediated by phosphatidylinositol 3-kinase and inhibited by cyclic AMP. Mol. Cell Biol. 15, 326–337 (1995).

75.  Wipf, P. & Halter, R. J. Chemistry and biology of wortmannin. Org. Biomol. Chem. 3, 2053–2061 (2005).

76.  Leahy, J. J. et al. Identification of a highly potent and selective DNA-dependent protein kinase (DNA-PK) inhibitor (NU7441) by screening of chromenone libraries. Bioorg. Med. Chem. Lett. 14, 6083–6087 (2004).

77.  Zhao, Y. et al. Preclinical evaluation of a potent novel DNA-dependent protein kinase inhibitor NU7441. Cancer Res. 66, 5354–5362 (2006).

78.  Farmer, H. et al. Targeting the DNA repair defect in BRCA mutant cells as a therapeutic strategy. Nature 434, 917–921 (2005).

79.  Bryant, H. E. et al. Specific killing of BRCA2-deficient tumours with inhibitors of poly(ADP-ribose)polymerase. Nature 434, 913–917 (2005).References 78 and 79 show that PARP inhibitors can be used to selectively kill BRCA1- or BRCA2-defective tumours.

80.  Moynahan, M. E., Chiu, J. W., Koller, B. H. & Jasin, M. Brca1 controls homology-directed DNA repair. Mol. Cell 4, 511–518 (1999).

81.  Moynahan, M. E., Pierce, A. J. & Jasin, M. BRCA2 is required for homology-directed repair of chromosomal breaks. Mol. Cell 7, 263–272 (2001).

82.  Patel, K. J. et al. Involvement of Brca2 in DNA repair. Mol. Cell 1, 347–357 (1998).

83.  Lomonosov, M., Anand, S., Sangrithi, M., Davies, R. & Venkitaraman, A. R. Stabilization of stalled DNA replication forks by the BRCA2 breast cancer susceptibility protein. Genes Dev. 17, 3017–3022 (2003).

84.  Helleday, T., Bryant, H. E. & Schultz, N. Poly(ADP-ribose) polymerase (PARP-1) in homologous recombination and as a target for cancer therapy. Cell Cycle 4, 1176–1178 (2005).

85.  Schultz, N., Lopez, E., Saleh-Gohari, N. & Helleday, T. Poly(ADP-ribose) polymerase (PARP-1) has a controlling role in homologous recombination. Nucleic Acids Res. 31, 4959–4964 (2003).

86.  Fisher, A., Hochegger, H., Takeda, S. & Caldecott, K. W. Poly (ADP-ribose) polymerase-1 accelerates single-strand break repair in concert with poly (ADP-ribose) glycohydrolase. Mol. Cell Biol. 27, 5597–5605 (2007).

87.  McCabe, N. et al. Deficiency in the repair of DNA damage by homologous recombination and sensitivity to poly(ADP-ribose) polymerase inhibition. Cancer Res. 66, 8109–8115 (2006).

88.  Bryant, H. E. & Helleday, T. Inhibition of poly (ADP-ribose) polymerase activates ATM which is required for subsequent homologous recombination repair. Nucleic Acids Res. 34, 1685–1691 (2006).

89.  Kennedy, R. D. et al. Fanconi anemia pathway-deficient tumor cells are hypersensitive to inhibition of ataxia telangiectasia mutated. J. Clin. Invest. 117, 1440–1449 (2007).

90.  Lengauer, C., Kinzler, K. W. & Vogelstein, B. Genetic instabilities in human cancers. Nature 396, 643–649 (1998).

91.  Gorgoulis, V. G. et al. Activation of the DNA damage checkpoint and genomic instability in human precancerous lesions. Nature 434, 907–913 (2005).

92.  Bartkova, J. et al. DNA damage response as a candidate anti-cancer barrier in early human tumorigenesis. Nature 434, 864–870 (2005).

93.  Kinzler, K. W. & Vogelstein, B. Cancer-susceptibility genes. Gatekeepers and caretakers. Nature 386, 761–763 (1997).

94.  Schmitt, C. A. Senescence, apoptosis and therapy — cutting the lifelines of cancer. Nature Rev. Cancer 3, 286–295 (2003).

95.  Braig, M. et al. Oncogene-induced senescence as an initial barrier in lymphoma development. Nature 436, 660–665 (2005).

96.  Vousden, K. H. & Lane, D. P. p53 in health and disease. Nature Rev. Mol. Cell Biol. 8, 275–283 (2007).

97.  Mariani, B. D. & Schimke, R. T. Gene amplification in a single cell cycle in Chinese hamster ovary cells. J. Biol. Chem. 259, 1901–1910 (1984).

98.  Rambach, W. A., Cooper, J. A. & Alt, H. L. Effect of hypoxia on DNA synthesis in the bone marrow and spleen of the rat. Science 119, 380–381 (1954).

99.  Hammond, E. M., Denko, N. C., Dorie, M. J., Abraham, R. T. & Giaccia, A. J. Hypoxia links ATR and p53 through replication arrest. Mol. Cell Biol. 22, 1834–1843 (2002).

100. Hammond, E. M., Dorie, M. J. & Giaccia, A. J. ATR/ATM targets are phosphorylated by ATR in response to hypoxia and ATM in response to reoxygenation. J. Biol. Chem. 278, 12207–12213 (2003).

101. Gibson, S. L., Bindra, R. S. & Glazer, P. M. Hypoxia-induced phosphorylation of Chk2 in an ataxia telangiectasia mutated-dependent manner. Cancer Res. 65, 10734–10741 (2005).

102. Bindra, R. S., Crosby, M. E. & Glazer, P. M. Regulation of DNA repair in hypoxic cancer cells. Cancer Metastasis Rev. 26, 249–260 (2007).

103. Rockwell, S., Yuan, J., Peretz, S. & Glazer, P. M. Genomic instability in cancer. Novartis Found. Symp. 240, 133–142; discussion 142–151 (2001).

104. Reynolds, T. Y., Rockwell, S. & Glazer, P. M. Genetic instability induced by the tumor microenvironment. Cancer Res. 56, 5754–5757 (1996).

105. Hammond, E. M., Dorie, M. J. & Giaccia, A. J. Inhibition of ATR leads to increased sensitivity to hypoxia/reoxygenation. Cancer Res. 64, 6556–6562 (2004).

106. Hanahan, D. & Weinberg, R. A. The hallmarks of cancer. Cell 100, 57–70 (2000).

107. Di Micco, R., Fumagalli, M. & di Fagagna, F. Breaking news: high-speed race ends in arrest — how oncogenes induce senescence. Trends Cell Biol. 17, 529–536 (2007).

108. Giannini, G. et al. Human MRE11 is inactivated in mismatch repair-deficient cancers. EMBO Rep. 3, 248–254 (2002).

109. Shor, E. et al. Mutations in homologous recombination genes rescue top3 slow growth in Saccharomyces cerevisiae. Genetics 162, 647–662 (2002).

110.  Wallis, J. W., Chrebet, G., Brodsky, G., Rolfe, M. & Rothstein, R. A hyper-recombination mutation in S. cerevisiae identifies a novel eukaryotic topoisomerase. Cell 58, 409–419 (1989).

111.  Tong, A. H. et al. Systematic genetic analysis with ordered arrays of yeast deletion mutants. Science 294, 2364–2368 (2001).This study show how systematic genetic analysis can be performed to produce a global map of gene function.

112. Miki, Y. et al. A strong candidate for the breast and ovarian cancer susceptibility gene BRCA1. Science 266, 66–71 (1994).

113. Hiramoto, T. et al. Mutations of a novel human RAD54 homologue, RAD54B, in primary cancer. Oncogene 18, 3422–3426 (1999).

114. Schoenmakers, E. F., Huysmans, C. & Van de Ven, W. J. Allelic knockout of novel splice variants of human recombination repair gene RAD51B in t(12;14) uterine leiomyomas. Cancer Res. 59, 19–23 (1999).

115. Wong, A. K. et al. Characterization of a carboxy-terminal BRCA1 interacting protein. Oncogene 17, 2279–2285 (1998).

116. Riballo, E. et al. Identification of a defect in DNA ligase IV in a radiosensitive leukaemia patient. Curr. Biol. 9, 699–702 (1999).

117. Moshous, D. et al. Partial T and B lymphocyte immunodeficiency and predisposition to lymphoma in patients with hypomorphic mutations in Artemis. J. Clin. Invest. 111, 381–387 (2003).

118. Nicolaides, N. C. et al. Mutations of two PMS homologues in hereditary nonpolyposis colon cancer. Nature 371, 75–80 (1994).

R E V I E W S

NATUre revIewS | cancer volUMe 8 | MArCH 2008 | 203

© 2008 Nature Publishing Group

119. Lipkin, S. M. et al. Germline and somatic mutation analyses in the DNA mismatch repair gene MLH3: Evidence for somatic mutation in colorectal cancers. Hum. Mutat. 17, 389–396 (2001).

120. de la Chapelle, A. Genetic predisposition to colorectal cancer. Nature Rev. Cancer 4, 769–780 (2004).

121. German, J., Bloom, D. & Passarge, E. Bloom’s syndrome. V. Surveillance for cancer in affected families. Clin. Genet. 12, 162–168 (1977).

122. Mohaghegh, P. & Hickson, I. D. DNA helicase deficiencies associated with cancer predisposition and premature ageing disorders. Hum. Mol. Genet. 10, 741–746 (2001).

123. Vorechovsky, I. et al. Clustering of missense mutations in the ataxia-telangiectasia gene in a sporadic T-cell leukaemia. Nature Genet. 17, 96–99 (1997).

124. Menisser-de Murcia, J., Mark, M., Wendling, O., Wynshaw-Boris, A. & de Murcia, G. Early embryonic lethality in PARP-1 Atm double-mutant mice suggests a functional synergy in cell proliferation during development. Mol. Cell Biol. 21, 1828–1832 (2001).

125. Bryant, H. E., Ying, S. & Helleday, T. Homologous recombination is involved in repair of chromium-induced DNA damage in mammalian cells. Mutat. Res. 599, 116–123 (2006).

126. Matsuura, S. et al. Positional cloning of the gene for Nijmegen breakage syndrome. Nature Genet. 19, 179–181 (1998).

127. Levine, A. J., Momand, J. & Finlay, C. A. The p53 tumour suppressor gene. Nature 351, 453–456 (1991).

128. Bell, D. W. et al. Heterozygous germ line hCHK2 mutations in Li–Fraumeni syndrome. Science 286, 2528–2531 (1999).

129. Cleaver, J. E. Cancer in xeroderma pigmentosum and related disorders of DNA repair. Nature Rev. Cancer 5, 564–573 (2005).

130. Cheng, L., Sturgis, E. M., Eicher, S. A., Spitz, M. R. & Wei, Q. Expression of nucleotide excision repair genes and the risk for squamous cell carcinoma of the head and neck. Cancer 94, 393–397 (2002).

131. Taniguchi, T. & D’Andrea, A. D. Molecular pathogenesis of Fanconi anemia: recent progress. Blood 107, 4223–4233 (2006).

132. Wang, L., Patel, U., Ghosh, L. & Banerjee, S. DNA polymerase β mutations in human colorectal cancer. Cancer Res. 52, 4824–4827 (1992).

133. Zheng, L. et al. Fen1 mutations result in autoimmunity, chronic inflammation and cancers. Nature Med. 13, 812–819 (2007).

134. Koster, D. A., Palle, K., Bot, E. S. M., Bjornsti, M.-A. and Dekker, N. H. Antitumour drugs impede DNA uncoiling by topoisomerase I. Nature 448, 213–217 (2007).

135. Wooster, R. et al. Localization of a breast cancer susceptibility gene, BRCA2, to chromosome 13q12-13. Science 265, 2088–2090 (1994).

AcknowledgementsWe would like to thank the Medical Research Council, The Swedish Research Council, The Swedish Cancer Society, The Swedish Children’s Cancer Foundation, The Swedish Pain Relief Foundation and Cancer Research UK for financial sup-port. We recognize that we were unable to cover all aspects of DNA repair in cancer in this Review. We apologize to those whom we have been unable to cite owing to space constraints.

Competing interests statementThe author declares competing financial interests: see web version for details.

DATABASESEntrez Gene: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=geneALKBH2 | ALKBH3 | ATM | ATR | BLM | BRCA1 | BRCA2 | BRIP1 | CHK1 | CHK2 | DCLRE1C| DDB1 | ERCC1 | ERCC2 | ERCC3 | ERCC4 | ERCC5 | FANCA | FANCB | FANCC | FANCD2 | FANCE | FANCF | FANCG | FANCL | FANCM | FEN1 | LIG4 | MGMT | MLH1 | MLH3 | Mre11 | MSH2 | MSH6 | MUS81 | NBN | p53 | PARP1 | PARP2 | PMS1 | PMS2 | POLB | POLH | PRKDC | RAD51 | RAD51L1 | RBBP8 | RECQL4 | RPA1 | sgs1 | SHFM1 | top� | WRN | XPA | XPC | XRCC2 | XRCC3ClinicalTrials.gov: http://clinicaltrials.gov/NCT00475917 | NCT00526617National Cancer Institute: http://www.cancer.gov/breast cancer | ovarian cancerNational Cancer Institute Drug Dictionary: http://www.cancer.gov/drugdictionary/5-fluorouracil | 6-thioguanine | bleomycin | carboplatin | chlorambucil | cisplatin | cyclophosphamide | dacarbazine | decitabine | etoposide | gemcitabine | hydroxyurea | melphalan | mitomycin C | oxaliplatin | temozolomide | UCN-01 | XL844

FURTHER INFORMATIONThomas Helleday’s homepage: http://www.rob.ox.ac.uk/research/researchgroups/thomas-helleday/Decitabine and Carboplatin in Relapsed Ovarian Cancer: http://pfsearch.ukcrn.org.uk/StudyDetail.aspx?TopicID=1&StudyID=2192

SUPPLEMENTARY INFORMATIONSee online article:S1 (table)

aLLLInkSareacTIveInTheOnLInePDf

R E V I E W S

204 | MArCH 2008 | volUMe 8 www.nature.com/reviews/cancer

© 2008 Nature Publishing Group