WO2009007677A2 - Anti-tumour methods - Google Patents

Abstract

There is provided a use of a DSB repair-inhibitor in the manufacture of a medicament for use in the treatment of a cancer in an individual, wherein the cancer comprises cells which are/is deficient in a BER and/or a SSBR pathway.

Description

Anti-tumour methods

Technical field The invention relates to the selective induction of cellular lethality in cancer cells when impaired in repair of single stranded DNA base damage and single strand breaks (gaps and nicks) induced by cellular metabolism, oxidative damage, radiation, or as a consequence of treatment with chemotherapeutic agents.

Background

Radiotherapy is one of the major treatment modalities for cancer, being applied in approximately half of all cancer patients. Response is often limited by intrinsic radioresistance, mainly determined by the ability to repair radiation-induced damage to the DNA in cancer cells. Several types of DNA lesion are induced after ionizing radiation of which double strand breaks (DSB) are thought to be the most lethal. However, single strand breaks (SSB) and base damages are quantitatively in the majority. Proteins of the base excision repair (BER) and single strand break repair (SSBR) pathway repair these numerous lesions. Other repair pathways include homologous recombination (HR) dependent DNA double strand break repair, non-homologous end joining (NHEJ), nucleotide excision repair (NER) and mismatch repair (MMR).

Base excision repair (BER) recognizes and repairs DNA damage inflicted by base modifying agents and oxidative damage exposure such as that due to metabolic radical formation or ionizing radiation. Single strand break repair is often defined as part of BER and repairs single strand nicks and gaps in the DNA helix. BER is initiated by the excision of the damaged base by lesion specific glycosylases, leaving an abasic site. The sugar-phosphate backbone is then incised by AP endonuclease (APEl) activity. DNA polymerase β (polβ) is then thought to remove the resulting moiety (a 5'-deoxyribose phosphate) and to fill the gap by incorporation of one or several nucleotides. Repair is completed by the XRCCl/ligaselll complex or, when driven through the long patch repair pathway, by polβ or the proliferating cell nuclear antigen (PCNA)-loaded replicative DNA polymerases δ and ε. With the help of the endonuclease activity of flap endonuclease 1 (FENl), completion can be achieved by ligase I. In addition to these pathways, initiated mainly by monofunctional glycosylases, bifunctional glycosylases remove the base lesion and simultaneously generate a nick by their 3'-β-lyase activity. This results in a baseless sugar that can be removed by APEl and is proposed to predominantly initiate short patch, polβ-dependent BER. Most glycosylases directed against oxidized bases as induced by ionizing radiation are bifunctional. In addition to base damages, single strand breaks are formed after cellular exposure to ionizing radiation. These have been shown in vitro to be repaired by a process similar to BER, involving polβ and XRCCl. In addition, however, polynucleotide kinases are first thought to trim the ends of the single strand nick to render it a suitable substrate for polβ (for review: Slupphaug et al. (2003) Mutat. Res. 531 231-251; Caldecott, (2007) DNA Repair (Amst), 6:443-453; Hazra et al, (2007) DNA Repair (Amst), 6:470-480).

DNA polymerase β (polβ) has been identified as a crucial enzyme in the single strand repair processes of BER and SSB repair. It has been shown that DNA polymerase β plays a role in determining response after ionising radiation (Vens et al, (2002) Nucleic Acids Res., 30:2995- 3004; Vermeulen et al, (2007) DNA Repair (Amst), 6:202-212). A considerable proportion of tumours express a catalytically inactive polβ protein (Starcevic et al, (2004) Cell Cycle, 3:998- 1001), thereby exhibiting BER impairment after ionising radiation. BER and SSB repair intermediates, if left unrepaired, will result in additional secondary lethal DSBs (Bryant et al, (2005) Nature, 434:913-917; Saleh-Gohari et al, (2005) MoI Cell Biol, 25:7158-7169).

The DNA double strand break (in contrast to single strand breaks affecting both strands of the DNA) is generally regarded as the most toxic of all DNA lesions. DSBs are induced in a number of different ways including exposure to ionizing radiation or radiomimetic drugs, as a result of chemotherapeutic treatment and after replication fork collapse when the replication machinery encounters a single stranded break in the template DNA, or crosslinks inhibiting replication fork progression. DSBs when left unrepaired are potent triggers of cell cycle arrest and apoptosis. DSB recognition is characterized by the activity of ATM and phosphorylation of the chromatin component H2AX. DNA double strand breaks are then repaired mainly by two pathways: homologous recombination (HR) and non-homologous end joining (NHEJ).

The difference in HR or NHEJ is based on the use of DNA sequence homology for repair. NHEJ uses little homology in a process that therefore may be error prone, whereas HR uses long stretches of sequence homology mostly from the undamaged neighbouring sister chromatid DNA strand to repair the damage. As a result, HR-directed repair is error free. The relative contribution of each repair pathway may define outcome with respect to cell survival or mutation induction.

NHEJ at a DSB is initiated by the binding of KU proteins to the ends which are then modified (cleaned) if necessary by TDPl and PNKP. Recruitment of DNA-PKcs (DNA-dependent protein kinase catalytic sύbunit) assures approximation and juxtaposition of the ends, initiating signalling events characterized by autophosphorylation and phosphorylation of other components of the DNA repair and DNA damage response pathways. Polymerases (possibly POLM and POLL) allow gap filling and finally ligation of the ends.

Initiation of HR, in contrast, is characterised by the resection of the DNA catalysed by nucleases and the MRN complex, creating single stranded DNA overhangs. The stability of these single stranded regions is assured by RPA coating. The overhangs are essential in the search for undamaged homologous regions and is facilitated by BRCAl/2 and the RAD51and RAD51 paralog proteins. After sister chromatid pairing, initiated with the loading of RAD52 and directed by RAD54 proteins, polymerisation on the undamaged sister chromatid template allows the filling of the gaps. DSB repair is completed by the resolution of the Holliday junction structure that resulted from this sister chromatid invasion and finally by ligation of the ends.

WO2005/053662 relates to exploitation of tumours impaired in homologous recombination (a DSB repair pathway active only in replicating cells). In this scenario, BER/SSBR is inhibited by poly (ADP-ribose) polymerase (PARP) inhibitors to increase DSB accumulation and therefore increase DSB repair dependence. Homologous recombination deficient tumours are then exposed to an increased number of DSBs that are not sufficiently repaired in the tumours only, subsequently resulting in cellular death. This is summarised in Figure IA.

Neijenhuis et al. (Radiother. Oncol. (2005) 76 123-129) described the radiosensitive phenotype of cells expressing a truncated polymerase β protein. It also demonstrated that sensitisation was dependent on XRCCl, indicating that XRCCl acts by inhibiting BER. In the discussion the possibility was raised that tumours with such aberrations could be expected to be more sensitive to radiation (especially to radiation alone) than other cells. The paper, however, did not show/mention or speculate on the targeted treatment of those tumours. No mechanism to achieve a therapeutic benefit was discussed or suggested, this problem in fact being solved by the invention described in the present application.

Boudsocq et al. (MoI. Pharmacol. (2005) 67 1485-1492) described the use of an inhibitor of DNA polymerase β in order to enhance the cellular response to cisplatin, a DNA crosslinking agent. The paper dealt with the inhibition of DNA polymerase β as therapeutic target. The authors proposed a therapeutic benefit in the treatment against cancer by the use of an inhibitor against elements of the BER, not against elements of DSB repair. No specific tumour targeting was described or proposed in this publication. In contrast, the strategy described in the current application takes advantage of some tumours displaying impaired DNA polymerase beta or

BER activity. DNA polymerase β inhibitors are not expected to change survival or response to

DNA-damaging agents in these tumours, since the enzyme is already inactive.

Taverna et al. (Cancer Res. (2003) 63 838-846) dealt with the increase in IdUrd mediated radiosensitisation by the use of the BER inhibitor MX. The authors indicated synergism in particular in mismatch repair deficient cells (MMR^"). However, the presented concept targeted MMR deficient tumours by the use of BER inhibitors, in combination with radiation. There was no mention of inhibition of DSB repair.

In contrast to the disclosures of the prior art, the invention disclosed here relates to the exploitation of tumours impaired in BER and/or SSBR. The inventors have shown that this results in the accumulation of DSBs. Cells under these conditions, in particular after DNA- damaging treatment such as radiation, rely on homologous recombination DSB repair. In this scenario, inhibitors of DSB repair (HR or NHEJ) affect BER and/or SSBR impaired cells more than BER/SSBR proficient cells. Normal tissues will usually be proficient in BER/SSBR whereas tumours relatively frequently show deficiencies in these pathways. This has the advantage that tumour cell-specific killing can be provided. This is summarised in Figure IB.

Disclosure of Invention According to a first aspect of the invention, there is provided the use of a DSB repair-inhibitor in the manufacture of a medicament for use in the treatment or prophylaxis of a cancer in an individual, wherein the cancer comprises at least one cell which is deficient in a BER and/or a SSBR pathway. Alternatively, there is provided the use of a DSB repair-inhibitor in treatment or prophylaxis of a cancer in an individual, the cancer comprising at least one cell which is deficient in a BER and/or a SSBR pathway.

As summarised in the accompanying Figure IB, the invention provides the selective induction of cellular lethality in cancer cells when impaired in repair of single stranded DNA base damage and single strand breaks (gaps and nicks) induced by radiation, oxidative damage, or as a consequence of treatment with chemotherapeutic agents. It has been found that base excision and/or single strand break repair (BER/SSBR) impairment present in a subgroup of cancer cells will result in increased secondary double strand break formation due to replication attempts at un-repaired BER sites, or as a consequence of attempted concomitant and inefficient repair at clustered sites. Therefore, advantageously, these cells are more sensitive to radiation after having been treated with compounds that inhibit DSB repair or signalling, than their BER/SSBR proficient counterparts. The use of DSB repair-inhibitors will, therefore, allow discrimination of cancer cells when deficient in BER/SSBR, inducing increased cellular lethality in response to radiotherapy. An increased radiosensitisation by DSB repair-inhibitors, for example ATM kinase inhibitors, has been observed by the inventors in cells expressing truncated DNA polymerase β. This has important implications in the treatment of cancer, since a considerable proportion of tumours have been described to express mutated and truncated DNA polymerase β.

The term "deficient in a BER and/or a SSBR pathway", as used throughout this specification, indicates that a cell has reduced or absent ability to repair DNA damage via the BER and/or SSBR pathway, as assessed in comparison to a normal, non-cancerous cell.

The invention was made by progress from the disclosure of the prior art, which (as mentioned above) showed that radiosensitisation by expression of an aberrant DNA polymerase beta depends on XRCCl, suggesting BER deficiency (Neijenhuis et al. (2005) Radiother. Oncol. 76 123-129).

DSB repair pathways had previously been identified as a possible target for inhibitors as general radiosensitisers. Targeting the DSB repair pathway would not result in specific kill of tumour cells, since all cells would be negatively affected by eliminating the DSB repair pathway. In addition, it was known that human tumours exhibit DNA polymerase β aberrations (Starcevic et al, (2004) Cell Cycle, 3:998-1001). However, there was no information provided as to how to target such tumours.

Subsequently, the inventors have made the following observations: • Extracts from cells with DNA polymerase β aberrations show impairment in base excision repair (BER) after exposure to ionizing radiation.

• Expression of aberrant DNA polymerase β results in the formation of secondary DSBs after exposure to ionizing radiation.

• Secondary DSBs after ionizing radiation resulting from BER inhibition are repaired via homologous recombination dependent DSB repair pathways.

• Specific radiosensitisation of base excision repair-impaired cells only and not BER proficient cells is possible by use of DSB repair-inhibitors.

• Specific kill of base excision repair-impaired cells only and not BER proficient cells is possible by the use of DSB repair-inhibitors.

The new observations by the inventors, set out above, have now allowed them to exploit the BER impairment characteristic of some tumours by the use of DSB repair-inhibitors, alone or combined with radiation. Surprisingly and advantageously, this allows targeted killing of tumour cells, whilst minimising damage to healthy cells. This invention would not have been achieved in the absence of the novel observations by the inventors, which would not have been expected.

In a further advantage of the invention, the DSB repair-inhibitor can be administered in a dosage or formulation that, in the absence of radiation or other DNA-damaging factors, is not lethal to cells. This reduces the chance of causing damage to non-cancerous cells. Such a dosage will typically be lower than a normal therapeutic dose. The skilled person is familiar with methods for determining a normal dose of such an inhibitor, according to the individual to whom the dose is being administered, the tissue type, the cancer type and the inhibitor compound to be used.

Preferably, the cancer comprises one or more cancer cells having a reduced or absent ability to repair DNA by BER and/or SSBR, compared with a normal, non-cancerous, cell. The individual may be homozygous or heterozygous for a mutation or polymorphism (SNP) in a gene encoding a component of a BER and/or SSBR pathway. Alternatively or additionally, the mutation may have arisen during the cancer development process. The individual may be a mammal and is preferably human.

As mentioned above, the term "deficient in a BER and/or a SSBR pathway" indicates that a cell has reduced or absent ability to repair DNA damage via the BER and/or SSBR pathway, as assessed in comparison to a normal, non-cancerous cell. Reduced ability to repair could result from changes in components of BER and SSBR.

Components of BER and SSBR, as mentioned anywhere in the specification, are defined as cellular elements such as molecules, proteins, RNA, miRNAs and peptides that, when absent, changed or deregulated, alter BER and SSBR parameters (initiation, efficiency, specificity, function, pathway use and component composition). Components of BER and SSBR also comprise regulators of the BER and SSBR repair pathway. Preferably, the component of a BER and/or SSBR pathway is selected from: UNG, SMUGl, OGGl, ALKBHl, ALKBH2, ALKBH3, TDG, MYH, NHTLl, MPG, NEILl, NEIL2, NEIL3, APEX2, APEXl, LIG3, XRCCl, ADPRT (PARPl), ADPRTL (PARP2), RPA, FENl, PCNA, POLDl, POLE, POLB, POLL, POLI, MGC5306, or TP53. MGC5306 and TP53 are proteins that function as regulatory factors on BER.

The cancer may be selected from any cancer which is treated by means of radiotherapy, particularly glioblastoma, head and neck cancer, lung cancer, cervical cancer, colorectal cancer, breast cancer or prostate cancer. However, this list is non-limiting and the skilled person will understand that any cancer may be susceptible to treatment using the invention.

The term "DSB repair-inhibitor", as used throughout this specification, means any entity as further defined below which is capable, in any way, of causing a reduction or elimination of DSB repair in a cell. The skilled person is readily able to determine whether exposing a cell to a given entity results in the reduction or elimination of DSB repair in the cell.

The DSB repair-inhibitor, as referred to throughout this specification, may be a small chemical molecule, antibody or antibody fragment, RNA aptamer or a peptide fragment binding to or consisting of a component of a DSB repair pathway, or may be a nucleic acid encoding all or part of the amino acid sequence of a component of a DSB repair pathway. In addition, a DSB repair-inhibitor may be a small molecule drug, peptide fragment or nucleic acid binding or targeting modulator, or regulator of the DNA double strand break repair pathway.

Examples of suitable inhibitors are ATM inhibitors (such as Ku-55933) or other DSB repair- inhibitors that target the non-homologous end joining machinery, such as DNA-Pk (KUDOS Pharmaceuticals) or the Ku70/80 proteins. Alternatively, a small chemical molecule, drug, peptide fragment or nucleic acid may be targeted to cellular components other than those directly involved in the DNA double strand break repair pathway, instead indirectly influencing the repair capacity of DNA double strands in a specific or non-specific way. Examples of such drugs are Cetuximab and inhibitors of heat shock proteins, such as the hsp90 inhibitor 17AAG or 17-DM AG. These drugs have been shown to alter DNA double strand break repair, i.e. by degradation of components of the DSB repair pathways such as BRCA2 and Rad51 (Noguchi et al, (2006) Biochem. Biophys. Res. Commun. 351, 658-663). Influence on DNA double strand break repair may comprise changes in the relative contribution of the DNA double strand break repair pathways, such as the shift to nonhomologous end joining and less homologous recombination, or the reverse. This class of DNA double strand break repair modulators are also referred to as DSB repair-inhibitors.

In summary, the term "DSB repair-inhibitor" is intended to encompass any entity which is capable of reducing, altering or eliminating, whether directly or indirectly, DSB repair in a cell.

In a preferred embodiment, the use according to the first aspect of the invention is a use in a treatment which further comprises administration of one or more of an ionising, oxidative or chemotherapeutic DNA-damaging agent.

The ionising DNA-damaging agent may be ionising radiation. The radiation may be emitted from particles or implants used in brachytherapy. The chemotherapeutic DNA-damaging agent may be a drug used during chemotherapy.

Suitable chemotherapeutic agents, where referred to in any part of this specification, may include inhibitors of topoisomerase I and II activity (e.g., camptothecin), drugs such as irinotecan, topotecan and rubitecan, alkylating agents such as temozolomide and DTIC (dacarbazine), and platinum agents like cisplatin, cisplatin-doxorubicin-cyclophosphamide, carboplatin, and carboplatin-paclitaxel. Other suitable agents include doxorubicin- cyclophosphamide, capecitabine, cyclophosphamide-methotrexate-5-fluorouracil, docetaxel, 5-flouracil-epirubicin-cyclophosphamide, paclitaxel, vinorelbine, etoposide, pegylated liposomal doxorubicin, gemcitibine and topotecan. Administration of such agents can be a single dose, or continuous or separate administrations. Such treatment regimes are well known and the skilled person is able to manage any required variations as the result of differences between individuals and types of cancer.

According to a second aspect of the invention, there is provided a method of treatment or prophylaxis of a cancer in an individual comprising administering a DSB repair-inhibitor to the individual, wherein the cancer comprises at least one cell having a reduced or absent ability to repair DNA by BER and/or SSBR, compared with a normal cell.

Administration of an inhibitor of a DSB repair pathway may be scheduled before and during radiotherapy treatment similar to treatment schedules of current radiosensitisers. Cancer growth might be reduced by administration of the DSB repair-inhibitors alone, thereby increasing tumour control.

The method may comprise a step of determining that the individual has a cancer which comprises one or more cells having a reduced or absent ability to repair DNA by BER and/or SSBR, compared with a normal cell. It is within the routine ability of the skilled person to assess this.

The individual may be homozygous or heterozygous for a mutation or polymorphism (SNP) in a gene encoding a component of a BER and/or SSBR pathway. The individual may be a mammal and is preferably human.

According to a third aspect of the invention, there is provided an in vitro method of treatment of a cancer which comprises cells having a reduced or absent ability to repair DNA by BER and/or SSBR, compared with normal cells, the method comprising administering a DSB repair-inhibitor to the cells. The term "in vitro method of treatment of a cancer" indicates that the cancer is treated away from the body of the individual suffering from the cancer, e.g., following removal from the individual of the organ or tissue comprising the cancer. The organ or tissue may subsequently be returned to the individual, following the treatment. The individual may be a mammal and is preferably human.

As mentioned above, the term "deficient in a BER and/or a SSBR pathway" indicates that a cell has reduced or absent ability to repair DNA damage via the BER and/or SSBR pathway, as assessed in comparison to a normal, non-cancerous cell. Reduced ability to repair could result from changes in components of BER and SSBR, such components being defined as above.

In the second and third aspects of the invention, the cancer may be selected from any cancer which is treated by means of radiotherapy, particularly (but not limited to) glioblastoma, head and neck cancer, lung cancer, cervical cancer, colorectal cancer, breast cancer or prostate cancer.

In the second and third aspects of the invention, the method may further comprise administering one or more of an ionising, oxidative or chemotherapeutic DNA-damaging agent to the individual. The ionising DNA-damaging agent may be ionising radiation. The radiation may be emitted from particles or implants used in brachytherapy. The chemotherapeutic DNA-damaging agent may be a drug used during chemotherapy.

Mutations or deficiencies that render cancer cells specifically sensitive to ionizing radiation by the use of DSB repair-inhibitors may affect any one of the components of base excision repair and single strand break repair. Components of BER and SSBR are defined as above.

As outlined in the "Background" section above, DNA double strand break repair pathways repair double strand breaks (DSB) in DNA mainly via homologous recombination and non-homologous end joining. A DSB repair-inhibitor may inhibit any of the components of these pathways. In addition, a DSB repair-inhibitor may induce changes in the DSB repair parameters such as initiation, efficiency, function, specificity, pathway and sub-pathway use and repair component composition. The term "DSB repair-inhibitor" is clearly defined above.

The components of the HR dependent double strand break (homologous recombination) repair pathway include: ATM, ATR, RAD51 (XRCC3), RAD51L1, RAD51C, RAD51L3, DMCl, XRCC2, XRCC3, XRCC4, RAD52, RAD54L, RAD54B, BRCAl, BRCA2, RAD50, MREIlA, NBSl, RPA, RPA1, RPA2, RPA3.

Components of the non-homologous end joining (NHEJ) pathway include:

DNA-PK (PRKDC), XRCC6 and XRCC5 (KU70, KU80), PNK, LIG4, POLL, POLK, POLM, XLF, DCLRElC (Artemis), TDPl. XRCCl, LIGl and APEl are in addition to their role in BER and/or SSBR found to influence DSB repair. Proteins such as BRCAl, MREIlA and NBS might play a role in HR and NHEJ.

Regulatory factors modulating DSB repair efficiency include: EMSY, H2AX, TP53BP1, TP53, SMC5. Other modulators include PARP (ADPRT) and RAD18.

Any entity which alters the activity of any one or more of these components such that DSB repair is reduced or eliminated is included in the term "DSB repair-inhibitor", as used throughout this specification. It is within the routine ability of the skilled person to determine whether the activity of one or more of these components is altered.

Molecules, peptides or nucleotides which affect any of the above-mentioned factors may be used as a DSB repair-inhibitor in a use or a method according to the invention. As mentioned previously, throughout this specification, the term "DSB repair-inhibitor" is intended to encompass compounds which directly or indirectly affect the level or quality of DSB repair, e.g., by inhibiting the activity of HR-dependent DSB repair or NHEJ repair, or by inhibiting the activity of agents which promote the activity of these pathways, or by binding to and inhibition of proteins and molecules involved in the promotion of these pathways, or by inhibition of signalling to downstream effectors that cause cell cycle checkpoint blocks, or by changing the composition of repair components, or by altering the respective use of HR versus

NHEJ, or by changing the specificity of the repair pathways. Suitable binding or inhibiting molecules may include small chemical molecules, antibodies or functional binding fragments or derivatives thereof. Peptides, nucleotides and RNA aptamers may also be included.

Therefore, the term encompasses any entity which is capable of reducing or eliminating, whether directly or indirectly, DSB repair in a cell. For example, these compounds might belong to a class of ATM inhibitors (such as Ku-55933) as described in WO03/070726. Other DSB repair-inhibitors that target the non-homologous end joining machinery, such as DNA-Pk (KUDOS Pharmaceuticals) or the Ku70/80 proteins result in the successful use of the invention described herein, in particular when combined with radiation. These drugs have been proposed as radiation sensitisers but do not induce tumour-specific kill. They are, therefore, expected to be of limited use as a result of causing considerable normal tissue damage. By use of the invention, for tumours presenting with a BER impairment, it is possible to lower the dose of drug or radiation to the extent of achieving an acceptable normal tissue damage with good tumour response. The data presented below demonstrate the selective radiosensitisation in cells deficient in BER by the use of a specific ATM inhibitor.

Base excision repair (BER/SSBR) impairment after oxidative damage, radiotherapy or chemotherapy results in the accumulation of un-repaired BER intermediates which could occur at any stage of the repair process. Intermediates will be present such as un-excised base damages and excised bases resulting in abasic sites, nicks and gaps. These BER intermediates present significant obstacles during replication and, if not resolved, induce DSBs or permanent replication arrest. Therefore, inhibitors targeting components that resolve replication fork arrest increase tumour-specific radiosensitisation in tumours bearing BER/SSBR deficiencies. Genes involved in this process have been identified by their potential to determine sensitivity to DNA crosslinking agents and include: FANCA, FANCB, FANCC, FANCDl, FANCD2, FANCE, FANCF, FANCG, FANCC, FANCM, RAD9, RAD17, RAD18, XPD, ERCCl, XPF, XPG, WRN, BLM, CHKl, CHK2, RADl, REV3, REV 1, POLH, HUSl, as well as components of the homologous recombination dependent DSB repair pathway as described above.

The compounds that would allow the use of the invention presented here are of particular interest in a radiotherapeutic setting, increasing damage resulting from ionizing radiation in the tumour cells only when present with aberrant gene expression that will lead to impaired BER/SSBR.

They also allow cancer treatment of such BER defective tumours independent of radiation treatment due to DNA damage resulting from cellular metabolism mostly caused by the exposure of metabolic induced oxidative radicals. Furthermore, when combined with chemotherapeutic agents, these drugs increase the cell death rate specifically in tumour cells impaired in BER/SSBR.

A therapeutically effective amount of compound described herein is typically one which is sufficient to achieve the desired effect and may vary according to the BER/SSBR impairment found in the cancer cells and general BER proficiency in the normal tissue. Furthermore, the normal tissue, inevitably irradiated when treating the cancer, might dictate the dose of a potential therapeutic inhibitor (the BER/SSBR tumour specific radiosensitising agent). Normal tissue response depends on patient individual variation and tissue type. The skilled person will routinely be able to make adjustments to dosage according to the differences between individuals, tissue type and cancer type.

A defect in a gene that mediates base excision repair or single strand break repair may be due to a mutation in, the absence of, or defective or altered expression of a gene encoding a protein involved in BER/SSBR. Altered expression compared to non-impaired cells encompasses overexpression and/or defective regulation of expression such as the lack of down or up regulation after a DNA damage insult. Changes in expression could be result of transcriptional alteration due to relocation, mutations, alternative splicing or epigenetic factors. They might also be caused by alteration in the process of translation into proteins as a consequence of mutations, miRNA interaction or other translation regulating factors. In addition, the activity of BER protein might be altered due to the regulation of other regulating proteins being kinases, acetylases, ubiquitin ligases and methyl transferases.

Microarray data analysis on tumour specimens demonstrated variations in expression levels of glycosylases. Concomitant reduction in the expression of the genes SMUG and UNG results in significantly decreased BER (Wilson et al. (2005) EMBO J. 24 2205-13), with unrepaired damages to be resolved during replication by homologous recombination associated repair mechanism, thereby rendering these cells to be more sensitive to the addition of DSB repair- inhibitors. Similarly, overexpression of glycosylases results in an increased formation of BER intermediates after ionizing radiation depending on DSB repair once converted into DSBs during replication or at clustered damages.

Tumours have been demonstrated to express, to a high proportion, aberrant DNA polymerase β (Sweasy et al. (2006) Radiat. Res. 166 693-714; Sweasy et al. (2006) Cell Cycle 5 250-9; Starcevic et al. (2004) Cell Cycle 3 998-1001). A significant number of these aberrations affect the activity of the enzyme. In particular, mutations and alterations in the catalytic part of the enzyme, such as deletions, truncations and mutations that lead to amino acid sequence alterations, are expected to cause BER deficiencies of the cells. The data presented below demonstrate the dependence on DSB repair in such cells expressing a truncated variant of DNA polymerase β.

The DSB repair-inhibitor mentioned throughout this specification may be administered with a pharmaceutically suitable adjuvant, carrier or vehicle which may be selected from: ion exchangers, alumina, aluminium stearate, lecithin, serum proteins, such as human serum albumin, buffer substances such as phosphates, glycine, sorbic acid, potassium sorbate, partial glyceride mixtures of saturated vegetable fatty acids, water, salts or electrolytes, such as protamine sulfate, disodium hydrogen phosphate, potassium hydrogen phosphate, sodium chloride, zinc salts, colloidal silica, magnesium trisilicate, polyvinyl pyrrolidone, cellulose- based substances, polyethylene glycol, sodium carboxymethylcellulose, polyacrylates, waxes, polyethylene-polyoxypropylene-block polymers, polyethylene glycol and wool fat.

The DSB repair-inhibitor may be administered within a pharmaceutical composition which may be administered orally or parenterally, preferably orally.

Where the pharmaceutical composition is administered orally, it may be in the form of a capsule or a tablet, and may preferably comprise lactose and/or corn starch. The pharmaceutical composition may further comprise a lubricating agent, preferably magnesium stearate. The pharmaceutical composition may be in the form of an aqueous suspension or aqueous solution, and may further comprise an emulsifying agent and/or a suspending agent. The pharmaceutical composition may comprise sweetening, flavouring and/or colouring agents.

The DSB repair-inhibitor may be administered within a pharmaceutical composition which may be administered by injection, by use of a needle-free device, by inhalation spray, topically, rectally, nasally, buccally, vaginally or via an implanted reservoir.

Where the pharmaceutical composition is administered by injection or needle-free device, it may be in the form of a sterile injectable preparation or a form suitable for administration by needle-free device. The sterile injectable preparation or form suitable for administration by needle-free device may be an aqueous or an oleaginous suspension, or a suspension in a nontoxic parenterally-acceptable diluent or solvent. The aqueous suspension may be prepared in mannitol, water, Ringer's solution or isotonic sodium chloride solution. The oleaginous suspension may be prepared in a synthetic monoglyceride, a synthetic diglyceride, a fatty acid or a natural pharmaceutically-acceptable oil. The fatty acid may be an oleic acid or an oleic acid glyceride derivative. The natural pharmaceutically-acceptable oil may be an olive oil, a castor oil, or a polyoxyethylated olive oil or castor oil. The oleaginous suspension may contain a long-chain alcohol diluent or dispersant, preferably Ph. HeIv.

The DSB repair-inhibitor may be administered within a pharmaceutical composition which may be administered rectally and may be in the form of a suppository for rectal administration. The suppository may comprise a non-irritating excipient which is solid at room temperature and liquid at rectal temperature. The non-irritating excipient may be one of cocoa butter, beeswax or a polyethylene glycol.

The DSB repair-inhibitor may be administered within a pharmaceutical composition which may be administered topically and may be an ointment comprising a carrier selected from mineral oil, liquid petroleum, white petroleum, propylene glycol, polyoxyethylene- polyoxypropylene compounds, emulsifying wax and water. Alternatively, it may be a lotion or cream comprising a carrier selected from mineral oil, sorbitan monostearate, polysorbate 60, cetyl esters wax, cetearyl alcohol, 2-octyldodecanol, benzyl alcohol and water.

The DSB repair-inhibitor may be administered within a pharmaceutical composition which may be administered nasally, for example by nasal aerosol and/or inhalation.

According to a fourth aspect of the invention, there is provided a method of selecting a DSB repair inhibiting compound (the DSB repair-inhibitor) comprising exposing a sample of cells deficient in a BER and/or SSBR pathway to a test compound and selecting a test compound if the level of cell death in the sample is greater than the level in a sample not exposed to the compound. This level may readily be assessed, for example, by determining the percentage of cells in a sample which undergo cell death within a given time period. From such measurements, the rate of cell death in a sample can also be calculated. According to a fifth aspect of the invention, there is provided a method of selecting a compound which increases the sensitivity of a cell to a DNA-damaging factor, the cell being deficient in a BER and/or a SSBR pathway, the method comprising: a) exposing a sample of the cells to a test compound and to the DNA-damaging factor and measuring the rate and/or amount of cell death in the sample; b) exposing an identical sample to that used in (a) to the DNA-damaging factor alone and measuring the rate and/or amount of cell death in the sample; and c) selecting a test compound if the rate and/or amount of cell death in step (a) is greater than in step (b).

According to a sixth aspect of the invention, there is provided a method of selecting a compound which increases the sensitivity of a cell, deficient in a BER and/or a SSBR pathway, to a DNA-damaging factor, the method comprising: a) exposing a sample of cells not deficient in a BER and/or a SSBR pathway to a test compound and to the DNA-damaging factor and measuring the rate and/or amount of cell death in the sample; b) exposing a sample of cells deficient in a BER and/or a SSBR pathway to the test compound and to the DNA-damaging factor and measuring the rate and/or amount of cell death in the sample; and c) selecting a test compound if the rate and/or amount of cell death in step (b) is greater than in step (a).

The DNA-damaging factor may be an ionising, oxidative or chemotherapeutic DNA-damaging agent. The cancer may be selected from any cancer which is primarily treated by means of radiotherapy, particularly glioblastoma, head and neck cancer, lung cancer, cervical cancer or colorectal cancer. The cells may comprise a mutation in the gene encoding DNA polymerase β. The data presented below demonstrate the feasibility of a high through-put screening in combination with radiation in cells deficient in BER and/or SSBR compared to cells proficient in these pathways.

According to a seventh aspect of the invention, there is provided a method of selecting a compound which increases the likelihood of cell death in a DNA-darnaged cell which is deficient in a BER and/or a SSBR pathway, the method comprising: a) exposing a sample of cells having DNA damage and being not deficient in a BER and/or a SSBR pathway to a test compound and measuring the rate and/or amount of cell death in the sample; b) exposing a sample of cells having the same DNA damage but being deficient in a BER and/or a SSBR pathway to the test compound and measuring the rate and/or amount of cell death in the sample; and c) selecting a test compound if the rate and/or amount of the cell death in step (b) is greater than in step (a).

The DNA damage may be due to endogenous damage whilst the cells are growing in contrast to the exogenously applied damage which is involved in methods in accordance with the sixth aspect of the invention.

In the fifth, sixth and seventh aspects of the invention, the cells may be cancer cells or may include one or more cancer cells.

Brief Description of Figures

Embodiments of the invention will now be described, by way of example only, with reference to the further accompanying Figures lB-10 in which:

Figure IB shows the scheme according to the invention for treatment of cancer cells;

Figure 2 shows increased γH2AX foci occurrence after ionising radiation in cells impaired in BER (A549-polβDN) compared to BER proficient cells (A549-LZRS) as a function of both time (A) and dose (B). Increased numbers of γH2AX foci indicate the increased occurrence of DNA double strand breaks;

Figure 3 shows increased occurrence of chromosome aberrations (A and B) after ionising radiation as a consequence of a BER defect caused by the expression of a truncated DNA polymerase β as determined by FISH, principally due to increased fragments (C; representing unrepaired breaks) rather than translocation (D; representing repaired breaks);

Figure 4 shows an increase in chromosome and chromatid type aberration induction in BER deficient cells (A549-polβDN) when irradiated in the S-phase of the cell cycle; Figure 5 shows radiation specificity of polβDN induced kill (A) and chromosome aberration induction (B) and increased oxidative damage induced chromatid aberrations (C) in BER impaired cells (A549-polβDN);

Figure 6 shows increased use and dependence on the homologous recombination driven DSB repair pathway after ionizing radiation when cells are impaired in BER;

Figure 7 shows targeted radiosensitisation of BER-impaired cells by the addition of a general DSB repair inhibiting drug, caffeine;

Figure 8 shows targeted radiosensitisation of BER impaired human tumour cells (A549- polβDN) by the addition of a specific ATM inhibitor, Ku-55933, as a function of ATM inhibitor dose at a fixed radiation dose of 2Gy (A) or as a function of radiation dose with a fixed ATM inhibitor dose (B);

Figure 9 shows targeted drug sensitisation and radiosensitisation of BER impaired human tumour cells by the addition of an unspecific drug, 17- AAG; and

Figure 10 shows feasibility of a cell based targeted radiosensitising drug screen approach.

Examples

Materials and methods Cells and cell culture

Human A549 (lung adenocarcinoma) cells were transduced with a retroviral construct carrying a His-tagged truncated version of DNA polymerase β (polβDN) and single cell clones were established. A549 polβDN-IIGlO, expressing the DNA polymerase β dominant negative (further referred to as A549-ρolβDN) and A549 LZRS-IIE2, transduced with the empty vector (referred to as A549-LZRS), were selected for further analysis. Both cell lines have been characterized extensively with respect to radiosensitivity (Vens et al, (2002) Nucleic Acids Res, 30:2995-3004). Cells were grown as monolayers in DMEM (Gibco) supplemented with 10% foetal calf serum at 37°C and 5% CO₂. CLV4B rad51 -deficient hamster cells and rad51 -complemented homologous recombination proficient CLV4B+rad51 cells were obtained from Dr. Zdzienicka (Drexler et al, (2004) DNA Repair (Amst) 3:1335-1343) and grown as monolayers inF12 medium at 37⁰C and 5% CO₂.

CLV4B and CLV4B+ cells were transfected with the LZRS vector carrying a truncated version of DNA polymerase β (polβDN) (further designated CLV4B or CLV4B+rad51 - polβDN) or with the empty vector (referred to as respectively CLV4B or CLV4B+rad51 - LZRS) using Liptofectamine according to manufacturers instructions. Transfected cells were selected by the addition of puromycin containing medium (lOμg/ml). Cells were further maintained and cultured in puromycin containing medium.

Radiation sensitivity and colony formation assay

Radiation experiments were performed as previously described (Neijenhuis et al, (2005) Radiother. Oncol, 76:123-129; Vens et al, (2002) Nucleic Acids Res., 30:2995-3004; Vermeulen et al, (2007) DNA Repair (Amst) 6:202-212).

For radiation experiments in confluence, cells were plated, cultured for 4 days to reach confluence and maintained under this condition for another 3 days. Cells were then irradiated using a ¹³⁷Cs irradiation unit with a dose rate of 0.66 Gy/min at room temperature. After an additional culture period of 24h to allow repair, cells were plated for colony formation. Preliminary time course experiments were performed to establish changes in cell cycle phase distribution during confluent culture and to further define the optimal time point of radiation. For each experiment, independent identical flasks were prepared to determine S-phase content of the cells by BrdUrd labelling and flow cytometry at time of irradiation.

For measurement of sensitivity to chemical inhibitors, cell cultures in exponential growth were trypsinised and plated. Six to fourteen hours thereafter cells were, where appropriate, treated with inhibitors for Ih and irradiated. Incubation with inhibitors continued for another 23h after radiation. Cells were then re-fed with fresh medium without inhibitor. Alternatively, for continuous exposure, cells were kept in inhibitor containing medium until wash and fixation.

After 12 days for A549 based cell lines and 6 days for CL4VB and CLV4B+, cells were washed with PBS, fixed and stained with crystal violet. Colonies containing greater than approximately 50 cells were counted. Survival is expressed as colonies per plated cells treated / colonies per plated cells untreated.

Fluorescence in situ hybridisation (FISH)

Whole chromosome specific probes were prepared from plasmid libraries from Dr. J. Gray (University of San Francisco, CA). Probes were amplified by degenerate oligonucleotide- primed-polymerase chain reaction (DOP-PCR) and labelled in a second reaction with biotin labelled-dUTP. Slides with metaphase spreads were hybridized with specific probes to the human chromosome 1, 2, and 4 after RNase / pepsin treatment as described previously (Coco

Martin et al, (1999) Int. J. Radiat. Biol., 75:1161-1168). The targeted chromosomes were detected by FITC (fluorescein-isothiocyanate) labelled streptavidin with an additional signal amplification step using biotinylated goat anti-avidin.

Chromosome and chromatid type aberration analysis

Translocations and acentric fragments per cell were determined after fluorescence in situ hybridisation with whole chromosome probes on metaphase chromosomes of cells in their first post-irradiation mitosis. Asynchronous exponentially growing cells were irradiated using a ¹³⁷Cs source at a dose rate of approximately 0.9Gy min^"1. After approximately one cell cycle (22h), metaphase cells were collected by overnight colcemid treatment. Colcemid (0.2μg/ml, Gibco) was added to the culture medium for an additional 16h and mitotic cells were then shaken off and collected by centrifugation. Metaphase cells and spreads were prepared according to standard cytogenetic protocols. Briefly, cells were washed with PBS, subsequently treated in hypotonic KCl solution (0.075M, Sigma ) and fixed in methanol/acetic acid (3:1). Metaphase chromosomes were stained with chromosome specific probes to 1, 2 and 4 and analysed on an Olympus fluorescence microscope. Translocations between stained and unstained chromosomes (colour junctions) were counted. Two independent translocation events on the same chromosome were counted separately. In addition, the number of stained acentric fragments was determined. 100 metaphases were analysed for each experiment. A minimum of three independent radiation experiments on cells of early passages were performed.

Further, chromosome and chromatid type of aberrations were determined on metaphases resulting from a 2h colcemid treatment 6h after radiation with 2Gy. To allow determination of sister chromatid exchanges, cells were cultured with lOμM BrdUrd (Sigma) for a total of two cell cycles. Mitotic cells were shaken off and prepared as described above. Metaphase spreads were prepared according to standard protocols (Perry et al, (1974) Nature, 251:156-158). Briefly, spreads were stained with Hoechst 33258 (5μg/ml, Roche, Mannheim) for 12min, rinsed and exposed to black light. After wash, slides were then stained with Giemsa (3%, Sigma) and analysed under a Vanox microscope (Olympus).

Immunohistochemistry for yH2AXand rad.51

Cells were grown on coverslips for 2-3 days until 50-70% confluent for exponentially growing cells or for confluent samples after reaching full confluence for another 3 days, irradiated using a ¹³⁷Cs irradiation unit with a dose rate of 0.66 Gy/min at room temperature and cultured for several hours afterwards. At different time points after irradiation cells were fixed with 2% paraformaldehyde for 15 min at room temperature, washed and permeabilised in 0.2% Triton X-100 on ice. After blocking in PBS with 1% BSA, the coverslips were incubated with anti- phospho-H2AX antibody (clone JBW301, Upstate, NY) or rad51 antibody for Ih. Washes were performed with 1% BSA in PBS before incubation with the secondary FITC conjugated anti-mouse antibody (Sigma). Cells were washed with PBS/0.1% Tween20 (Sigma- Aldrich). Coverslips were mounted on slides in Vectashield with DAPI. Slides were viewed using a Zeiss fluorescence microscope (Axiovert 100M) equipped a 10Ox Neofluor objective and a CCD camera (MAC 200A, Photometries). Foci were counted in 100 cells per dose and time. Each irradiation experiment was repeated a minimum of three times.

Cell cycle phase distribution

Cells were pulse labelled with lμM BrdUrd (bromodeoxyuridine, Sigma) by incubating for 10 min. Cells were trypsinised, resuspended in PBS and fixed in 70% ethanol. BrdUrd detection was performed as described elsewhere (Begg et al, (1991) Cytometry, 12:445-454). Samples were measured using a FACScan flow cytometry (Becton Dickinson). Data were analysed with the FCS Express (DeNovo Software) software package and the percentage of labelled BrdUrd incorporating S-phase cells was calculated. For cell cycle block analyses, cells were pulse labelled with BrdUrd and washed. Irradiations were performed 20 min after labelling. Cells were trypsinised and fixed at different time points after irradiation. Cell cycle phase distributions were determined after removing doublets by gating on the dual parameter histogram of area versus width of the PI signal. Analysis of chromosome and replication associated chromatid aberrations Cells were irradiated and 6h later treated for 2h with colcemid. The metaphases collected at this time were derived from cells that were in mid S-phase at the time of irradiation. Aberrations were grouped into acentric chromosome deletions, dicentric chromosomes, rings and chromatid-type aberrations, including chromatid gaps (gaps), chromatid gaps with loss (fragments), tri and quadraradials.

BER deficient tumour-specific drug screen hi addition to known inhibitors of double strand break repair pathways, drug screens using BER impaired cell lines as targets reveal new classes of drugs that allow the targeting of a subclass of tumours with BER impairment. Such a screen can be used to assess the response to combined drug and radiation. Alternatively, drug screens can be performed which identify BER impairment specific growth inhibition and kill specific to the BER impairment (without the combination with radiation).

Read-outs of response are based on growth assay based techniques such as the previously described ratio assay, determining fluorescence ratios of two cell lines with and without BER/SSBR impairment measured by flow cytometry or ELISA readers. Either tumour or normal cells were genetically modified to express aberrant BER proteins or BER regulating proteins. Genetically modified and control cells were labelled with distinguishing markers, mixed and concomitantly treated. Labelling was achieved by the expression of fluorescent, luciferase active proteins or other enzymes, stains or genetic codes. Screens included the analysis of cell lines originally derived from tumours, with and without BER impairment. In particular, the expression of aberrant DNA polymerase β variants such as those observed in human tumour material is useful for screens identifying new tumour-targeted radiosensitising drugs.

This new class of drug may act on proteins of DSB repair pathways, or on other proteins which have an ultimate effect on DSB repair pathways, for example by modulation of key genes such as POLB, the gene encoding DNA polymerase β. Such proteins may also include inhibitors of cell cycle checkpoints or other yet unknown pathways involved in the response to BER deficiency. Results

PolβDN expression results in increased yH2AXfoci after ionizing radiation in cells expressing truncated DNA polymerase β.

Several cell lines have been shown to be radiosensitised by the expression of truncated DNA polymerase β (polβDN) (Neijenhuis et al, (2005) Radiother. Oncol, 76:123-129; Vens et al,

(2002) Nucleic Acids Res., 30:2995-3004). In addition, these cells have been found to be impaired in base excision repair / single strand break repair after ionizing radiation using a plasmid based in vitro assay (Vens et al, (2007) DNA Repair (Amst) 6 202-212) . It was proposed that increased cell kill after ionizing radiation resulted from accumulation of unrepaired BER intermediates leading to an additional burden of DSBs. To search for these breaks, γH2AX foci formation was analysed.

For these studies, the p53 wild type human adenocarcinoma cell line A549 was chosen, since radiosensitivities and growth characteristics of the polβDN and empty vector transduced single cell clones (A549-polβDN II Gl 0 and A549-LZRS II E2) have been reported previously (Vens et al, (2002) Nucleic Acids Res., 30:2995-3004). Elevated γH2AX foci numbers were found in polβDN expressing cells at 6 and 1Oh after radiation with 2Gy (Fig. 2A). In search of persistent unrepaired DSBs, foci were analysed 24h after irradiation. Survival after ionizing radiation correlates strongly with residual γH2AX foci values. The dose response of γH2AX foci formation at these late time points showed a significant increase in average γH2AX foci numbers per cell (Fig. 2B) in the polβDN expressing cells. However, this increase was less than expected when calculated from the observed difference in kill (dashed line in Fig. 2B).

In summary, the increase in γH2AX foci in polβDN expressing cells demonstrates an augmentation in DNA damage after ionizing radiation.

Cells expressing aberrant truncated DNA polymerase β show increased chromosome aberrations after ionizing radiation

To confirm the observed increase in residual DSBs indicated by the foci, chromosome aberrations were also studied. Chromosome aberrations were determined by FISH with whole chromosome probes against chromosome 1, 2 and 4, covering approximately 25% of the genome in this cell line. Chromosome aberrations increased after γ-radiation in both cell lines, being significantly increased in the cell line expressing the polβDN (A549-polβDN) compared to the empty vector (A549-LZRS) control (Fig. 3A). A549-polβDN cells showed a nearly 2-fold increase of chromosome aberrations/metaphase at 6Gy (DEF, dose enhancement factor of approximately 1.9). The number of aberrant cells increased similarly and to a greater extent in the polβDN expressing cells (Fig. 3B).

Further analysis of the types of aberration revealed no significant change in the induction of stable chromosome aberrations (translocations; Fig. 3C) either with increasing dose or due to the expression of the polβDN. However, for the induction of deletions, a significant difference (p = 0.008 at 3Gy and p = 0.14 at 6Gy) was observed between polβDN expressing cells and their vector controls (Fig. 3D). Since chromosome deletions are a strong indicator of unrepaired DSBs, these data confirmed the hypothesized DSB formation in polβDN expressing cells after ionizing radiation.

In summary, γH2AX foci and chromosome aberrations studies demonstrated additional DSB formation after radiation in cells with a defect in BER/SSBR, here caused, by way of example, by the expression of aberrant DNA polymerase β. The discrepancy displayed by the strong chromosome aberration induction by the polβDN and the relatively low increase of residual yH2AX foci indicates a lack of γH2AX phosphorylation / foci formation of polβDN induced secondary DSBs.

Replication-dependent double strand break formation and sensitization of the polβDN after ionizing radiation

One possible source of secondary DSBs is the conversion to DSBs at replication forks encountering a nick (SSB) in the template. During replication, unrepaired SSBs resulting from polβDN interference after ionizing radiation would lead to additional DSBs, observed by the γH2AX foci and chromosome analysis. As the other replication template remains intact, these additional DSBs, if unrepaired, will lead to an increase in chromatid aberrations. These are classical G2-type aberrations with one intact and one aberrant sister-chromatid. This question was, therefore, addressed by analysing chromosome and replication associated chromatid aberrations. Cells were irradiated and treated with colcemid. Aberrations were grouped into acentric chromosome deletions, dicentric chromosomes, rings and chromatid-type aberrations, including chromatid gaps (gaps), chromatid gaps with loss (fragments), tri and quadraradials. As shown in figure 4 , A549-polβDN exhibited a dramatic increase in chromosome aberrations (Fig. 4A) and chromatid aberrations (Fig. 4C). An increase in dicentrics was not observed (Fig. 4B). It was concluded that inhibition of BER/SSBR after ionizing radiation leads to secondary DSBs which are partly un-repaired or mis-repaired, leading mainly to Gl-type chromosome aberrations. In addition, the changes in chromatid aberrations indicated an additional burden of DSBs after ionizing radiation due to the conversion of unrepaired SSBs to DSB at the replication fork.

Radiation specificity: sensitivity to H₂O₂ is not affected by the polβDN expression

Persistent DSB formation after ionizing radiation was evident in the γH2AX and chromosome aberration studies. It was concluded that a small fraction of lesions remained un-repaired by the polβDN interaction, undetectable within the gross repair of oxidative lesions and leading to persistent and thereby highly cytotoxic type of DSBs. Since part of the additional lesions were apparently formed independent of replication, it was speculated that the additional DSBs resulted from interference in radiation-specific clustered lesions. The question of whether polβDN cytotoxicity was restricted to ionizing radiation was, therefore, tested. Growth after treatment with hydrogen peroxide was analysed, this being another agent resulting in oxidative DNA damage. A549-polβDN and A549-LZRS exhibited identical growth inhibition curves (Fig. 5A), indicating a lack of sensitization after hydrogen peroxide. This suggests involvement of the polβDN in repair of clustered damage, which is unique to ionizing radiation.

However, the chromatid aberration data indicated part of the secondary DSBs results from SSB to DSB conversion during replication. This was expected to happen similarly after H₂O₂ treatment but might be less toxic, since they only affected one sister chromatid. Chromosome and chromatid aberrations after H₂O₂ were, therefore, analysed and it was found that only chromatid aberrations significantly increases (Fig. 5B and C). These data further demonstrate the induction of secondary DSBs after exposure of other agents than radiation, in this case after oxidative damage from H₂O₂ exposure,

Increased dependence on homologous recombination DSB repair after ionizing radiation

The replication independent involvement of the aberrant DNA polymerase β protein, presumably at clustered damage, resulted in the formation of un-repairable lethal double strand breaks as evident in the chromosome aberration analysis, thereby causing increased cell death after ionizing radiation, as demonstrated by the survival analysis. It was further speculated that the expression of truncated DNA polymerase β, in addition to these "lethal" interactions, would independently cause the accumulation of BER intermediates that are repaired by backup repair pathways. In particular, BER intermediates encountering the replication fork would be converted to secondary DSBs that are repaired by the homologous recombination repair pathway.

Therefore, the dependence of polβDN expressing cells to homologous recombination repair (HR) was analysed. For this purpose, the polβDN was expressed in cells deficient or proficient in homologous recombination repair (CLV4B and CLV4B+rad51) caused by the deficiency of rad51C. As shown in Figure 6A, expression of polβDN in rad51C-complemented and HR- proficient cells resulted in a small radiosensitising effect, as observed previously with other cell lines (Neijenhuis et al, (2005) Radiother. Oncol, 76:123-129). A dramatic increase in radiosensitisation of the polβDN was observed in cells that are deficient in HR as shown in Figure 6B.

Rad51 is a crucial protein in homologous recombination directed DSB repair. The number of Rad51 foci in a cell are often used as surrogate markers for HR events in the cell. The formation of rad51 foci after ionizing radiation in the two cell lines was then analysed. Ionizing radiation induced increased numbers of rad51 foci in the polβDN expressing cells compared to the control cell line, confirming increased HR involvement in these cells (Fig. 6C). These results demonstrate the increased dependence on homologous recombination directed DSB repair after ionizing radiation of cells expressing truncated DNA polymerase β and impaired in BER.

Targeted radiosensitisation of cells expressing polβDN

The previous results showed that cells expressing truncated DNA polymerase β were highly sensitive to ionizing radiation when impaired in homologous recombination repair by genetic means. Blocking DSB repair by the addition of inhibitors, in an attempt to demonstrate general applicability in cancer cells impaired in BER/SSBR, was carried out. Human A549 tumour cells expressing the polβDN were treated with 2.5mM caffeine 1 hour prior to irradiation and up to 24 hours after irradiation. Survival was determined by colony forming assays. Caffeine has been demonstrated to inhibit kinases involved in DSB repair such as ATM. As shown by Wang et al (Wang et al, (2003) Radiat. Res., 159:420-425; Wang et al, (2003) Radiat. Res., 159:426-432), radiosensitisation to caffeine has been shown to depend on homologous recombination. The addition of caffeine resulted in an increased radiosensitisation of the polβDN expressing cells compared to empty vector controls (A549-LZRS) as expected (Figure 7A). Radiosensitisation was more pronounced at low doses achieving an additional two fold kill in A549-polβDN cells at a caffeine dose that did not radiosensitise empty vector cells and after a radiation dose of 2Gy that typically does not result in a radiosensitising effect of the polβDN alone (Figure 7B). The polβDN specific radiosensitising effect of the DSB repair-inhibitor caffeine can be observed over a large dose range as demonstrated in Figure 7C.

Targeted radiosensitisation of BER deficient cells with caffeine was confirmed by the analysis of caffeine radiosensitisation in EM9 cells, which are deficient in XRCCl, a crucial component in BER/SSBR. Addition of caffeine resulted in increased radiosensitisation in EM9 cells when compared to the parental AA8 control cells (Fig. 7D). Endogeneous (oxidative) damage will result in the accumulation of BER/SSBR intermediates in these BER deficient cells. These intermediates were expected to convert to secondary DSBs during replication. The concept depicted in Figure IB would predict increased sensitization to DSB inhibitors in BER impaired cells, even without radiation. Figure 7E demonstrates such an increased sensitivity. Toxicity to the DSB repair and checkpoint inhibitor caffeine is significantly increased in the BER deficient EM9 cells.

The potential use of specific inhibitors in combination with radiation was then analysed. The new ATM inhibitor (2-Morpholin-4-yl-6-thianthren-l-yl-pyran-4-one or Ku-55933, developed by Kudos/AstraZeneca) was tested for targeted radiosensitisation of BER deficient cells. As shown in Figure 8A, incubation for 24h with the ATM inhibitor Ku-55933 initiated Ih prior radiation with 2Gy resulted in increased sensitivity to ionizing radiation in the cells expressing the polβDN. When analysed at a fixed dose of the inhibitor (that results in radiosensitisation of the control cells) the radiation dose response (Fig. 8B) confirmed the increased radiosensitising potential of the inhibitor in the polβDN expressing cells.

As described in the "Background" section of this application, drugs targeting other cellular pathways but indirectly resulting in reduced HR might apply to this invention. 17 AAG, a Hsp90 inhibitor reported to result in deficiencies in DSB repair and DNA damage response (Dote et al. Cancer Res 2006; 66(18): 9211-20), was tested. As shown in Figure 9B, addition of 17AAG resulted in increased radiosensitisation of BER impaired cells. In addition, A549-polβDN cells are more sensitive to the exposure to 17AAG (without radiation) than controls (Fig. 9A), demonstrating the possibility of targeting BER deficient cells by single agents without the combination with radiation.

Targeted radiosensitisation drug screen

A method of screening drugs for their radiosensitising potential was tested. Cells either expressing the polβDN or control empty vector expressing cells were plated on 384 well plates, irradiated 24h later with 2Gy and analysed 7 days afterwards. Cell growth was determined by an automated microscopic cell count after fixation and DAPI staining. Survival was calculated as the fraction of average cell numbers in the irradiated plates versus untreated wells. As demonstrated in Fig. 10, cell survival as determined by this approach is similar to the radiation response analysed by conventional colony formation assays. Since kill by 2Gy and differential response in the polβDN cells versus control cells is detectable, this was concluded to be an approach to detect chances in radiosensitivities (or toxicities) equal or greater than the observed effect of the polβDN on cell survival after radiation.

Summary The data presented here emphasise that successful repair of SSBs and base damages after ionizing radiation is critical for cell survival. It has been shown that inhibition of SSBR and BER after ionizing radiation results in the formation of additional DSBs. These secondary DSBs are radiation specific and result in the death of cells expressing a truncated DNA polymerase β. In addition to these lethal radiation specific secondary DSBs, the data included in this specification demonstrate the formation of secondary DSBs (by replication dependent SSB to DSB conversion) which depend on repair by the homologous recombination repair pathway. These DSBs do not result in cell death when proficient in HR, but lead to increased death when deprived of the secondary backup repair pathway. Since a considerable portion of tumours present with aberrant DNA polymerase β that will result in BER impairment, this concept can be exploited for the current invention. Furthermore, any other deficiency or alteration in cancer cells that results in BER impairment will allow targeted radiosensitisation of the cancer cells only by the use of DSB repair-inhibitors. This can be achieved by choosing an inhibitor dose that does not increase radiosensitivity in non-impaired cells.

Claims

Claims

1. Use of a DSB repair-inhibitor in the manufacture of a medicament for the treatment or prophylaxis of a cancer in an individual, wherein the cancer comprises at least one cell which is deficient in a BER and/or a SSBR pathway.

2. Use of a DSB repair-inhibitor in treatment or prophylaxis of a cancer in an individual, the cancer comprising at least one cell which is deficient in a BER and/or a SSBR pathway.

3. Use according to claim 1 or 2 wherein the cancer comprises one or more cancer cells having a reduced or absent ability to repair DNA by BER and/or SSBR, compared with normal cells.

4. Method of treatment or prophylaxis of a cancer in an individual comprising administering a DSB repair-inhibitor to the individual, wherein the cancer comprises at least one cell having a reduced or absent ability to repair DNA by BER and/or SSBR, compared with a normal cell.

5. Method according to claim 4 comprising the step of determining that the individual has a cancer which comprises one or more cells having a reduced or absent ability to repair DNA by BER and/or SSBR, compared with a normal cell.

6. Use according to any one of claims 1-3 or method according to claim 4 or 5 wherein the individual is homozygous for a mutation in a gene encoding a component of a BER and/or SSBR pathway.

7. Use according to any one of claims 1-3 or method according to claim 4 or 5 wherein the individual is heterozygous for a mutation in a gene encoding a component of a BER and/or SSBR pathway.

8. Use or method according to claim 6 or 7 wherein the component of a BER and/or

SSBR pathway is selected from UNG, SMUGl, OGGl, ALKBHl, ALKBH2, ALKBH3, TDG, MYH, NHTLl, MPG, NEILl, NEIL2, NEIL3, APEX2, APEXl, LIG3, XRCCl, ADPRT (PARPl), ADPRTL (PARP2), RPA, FENl, PCNA, POLDl, POLE₅ POLB₅ POLL, POLI, MGC5306 and TP53.

9. Use according to any one of claims 1-3 or 6-8 or method according to any one of claims 4-8 wherein the cancer is selected from any cancer which is treated by means of radiotherapy.

10. Use or method according to claim 9 wherein the cancer is glioblastoma, head and neck cancer, lung cancer, cervical cancer, colorectal cancer, breast cancer or prostate cancer.

11. Use according to any one of claims 1-3 or 6-10 or method according to any one of claims 4-10 wherein the DSB repair-inhibitor is a peptide fragment of a component of a DSB repair pathway.

12. Use according to any one of claims 1-3 or 6-10 or method according to any one of claims 4-10 wherein the DSB repair-inhibitor is a nucleic acid encoding all or part of the amino acid sequence of a component of a DSB repair pathway.

13. Use according to any one of claims 1-3 or 6-10 or method according to any one of claims 4-10 wherein the DSB repair-inhibitor is a small chemical molecule.

14. Use according to any one of claims 1-3 or 6-13 or method according to any one of claims 4-13 wherein the DSB repair-inhibitor is a Hsp90 inhibitor.

15. Use or method according to claim 14 wherein the DSB repair-inhibitor is a small chemical molecule which is 17AAG or 17-DM AG.

16. Use according to any one of claims 1-3 or 6-10 or method according to any one of claims 4-10 wherein the DSB repair-inhibitor is an ATM inhibitor.

17. Use or method according to claim 16 wherein the ATM inhibitor is Ku-55933.

18. Use according to any one of claims 1-3 or 6-17 or method according to any one of claims 4-17 further comprising administering an ionising, oxidative or chemotherapeutic DNA-damaging agent to the individual.

19. Use or method according to claim 18 wherein the ionising DNA-damaging agent is ionising radiation.

20. Use or method according to claim 18 wherein the chemotherapeutic DNA-damaging agent is a drug used during chemotherapy.

21. In vitro method of treatment of a cancer which comprises cells having a reduced or absent ability to repair DNA by BER and/or SSBR, compared with normal cells, the method comprising administering a DSB repair-inhibitor to the cancer cells.

22. Method of selecting a DSB repair-inhibiting compound comprising exposing a sample of cells deficient in a BER and/or SSBR pathway to a test compound and selecting a test compound if the level of cell death in the sample is greater than the level in a sample not exposed to the compound.

23. Method of selecting a compound which increases the sensitivity of a cell to a

DNA-damaging factor, the cell being deficient in a BER and/or a SSBR pathway, the method comprising: a) exposing a sample of the cells to the compound and to the DNA-damaging factor and measuring the rate and/or amount of cell death in the sample; b) exposing an identical sample to that used in (a) to the DNA-damaging factor alone and measuring the rate and/or amount of cell death in the sample; and c) selecting the compound if the rate and/or amount of cell death in step (a) is greater than in step (b).

24. Method of selecting a compound which increases the sensitivity of a cell, deficient in a

BER and/or a SSBR pathway, to a DNA-damaging factor, the method comprising: a) exposing a sample of cells not deficient in a BER and/or a SSBR pathway to a test compound and to the DNA-damaging factor and measuring the rate and/or amount of cell death in the sample; b) exposing a sample of cells deficient in a BER and/or a SSBR pathway to the test compound and to the DNA-damaging factor and measuring the rate and/or amount of cell death in the sample; and c) selecting a test compound if the rate and/or amount of cell death in step (b) is greater than in step (a).

25. Method according to claim 23 or 24 wherein the DNA-damaging factor is an ionising, oxidative or chemotherapeutic DNA-damaging agent.

26. Method according to any one of claims 21-25 wherein the cancer is selected from any cancer which is treated by means of radiotherapy.

27. Method according to claim 26 wherein the cancer is glioblastoma, head and neck cancer, lung cancer, cervical cancer, colorectal cancer, breast cancer or prostate cancer.

28. Method according to any of claims 21-27 wherein the cells comprise a mutation in the gene encoding DNA polymerase β.

29. Method of selecting a compound which increases the likelihood of cell death in a DNA-damaged cell which is deficient in a BER and/or a SSBR pathway, the method comprising: a) exposing a sample of cells having DNA damage and being not deficient in a BER and/or a SSBR pathway to a test compound and measuring the rate and/or amount of cell death in the sample; b) exposing a sample of cells having the same DNA damage but being deficient in a BER and/or a SSBR pathway to the test compound and measuring the rate and/or amount of cell death in the sample; and c) selecting a test compound if the rate and/or amount of the cell death in step (b) is greater than in step (a).