A METHOD OF PREDICTING THE EFFICACY OF A DRUG

RELATED APPLICATION

This application claims for the priority based on U.S. Provisional Application No. 61/205,971, filed on 26 January, 2009, the entire disclosure of which application is incorporated herein.

TECHNICAL FIELD

The present invention relates to a method of predicting the efficacy of a drug. More specifically, the present invention relates to a method of predicting sensitivity or a cytotoxic effect of drugs on tumor and/or cancer.

BACKGROUND OF THE INVENTION

Lung cancer continues to be the leading cause of cancer death worldwide (1, 2), and non-small cell lung cancer (NSCLC) is the most common type of lung cancer. Despite many clinical trials of platinum-based chemotherapy in combination with various drugs, the median survival time of NSCLC patients remains poor. The overall 5-year survival rate is ~15%, and has improved only marginally over the last few decades despite the introduction of new therapeutic agents (3, 4). A recent milestone in this field has been the development of molecular-targeting drugs, among which gefitinib and erlotinib targeting the epidermal growth factor receptor (EGFR) have improved the efficacy of therapy for NSCLC. The absence or presence of mutations within the kinase domain of the EGFR gene in lung adenocarcinoma cells has a key role in determining the therapeutic efficacy of

EGFR-targeting drugs (5-7). The presence of EGFR mutations in lung cancer cells confers an EGF/TGFα-dependent growth capacity together with susceptibility to the cytotoxic effect of gefitinib in culture (8). About 80% of tumors possessing EGFR mutations respond to EFGR-tyrosine kinase inhibitors (EGFR-TKI). In NSCLC patients, more than 90% of EGFR mutations are located in exon 19 (delE746-A750) or 21 (L858R point mutation). Several factors have been reported to be associated with the frequency of EGFR mutations, including an adenocarcinoma phenotype, female gender, having never smoked, and East Asian ethnicity (5-7, 9), but the underlying mechanisms whereby gefitinib-sensitive mutations are induced remains unclear. A recent study by Mok et al. (2009) reported that gefitinib is superior to carboplatin-paclitaxel as an initial treatment for NSCLC adenocarcinoma among nonsmokers in East Asia (34).

Oxidative DNA damage and repair contribute to the development of various human pathologies, including cancer. In both nuclear and mitochondrial DNA, the oxygenated nucleotide 8-hydroxy-2'-deoxyguanosine (8-OHdG) has been implicated in the type of somatic mutations found in human cancers (35). The major pathway for oxidative DNA damage repair is base excision repair, which in humans involves the MutM human homo log 8-oxoguanine DNAglycosylase (OGGl), the MutY homo log MUTYH, and the Mthl homolog MTHl (36). Among these base excision repair-related genes, OGGl appears to be very important. Oka et al. have demonstrated two distinct pathways of cell death by oxidative damage to nuclei and mitochondria, and OGGl plays key role in the repair of oxidative DNA damage in both pathways (37). On the other hand, OGGl-null mice develop adenomas and carcinomas of the lungs with a marked increase of 8-OHdG (38), and base excision repair-defective mice (myh-/-, OGGl-/-) show significantly increased accumulation of 8-OHdG in the liver, small intestine and lung DNA in comparison with wild-type mice (39). Frequent loss of heterozygosity has been observed in the region of the OGGl gene in lung and kidney cancer (40), and some genetic polymorphisms of the OGGl gene are associated with an increased risk for various cancers (41). Elevated levels of urinary 8-OHdG have also been detected in patients with various cancers, including those of the breast, bladder and prostate (42). Furthermore, 8-OHdG and base excision repair modulation are expected to be risk factors for human cancers (43). 8-OHdG has also been highlighted as a marker of oxidative stress and damage related to occupational and environmental exposure (44, 45). On the other hand, the Y-box binding protein- 1 (YB-I) has been implicated in numerous functions, such as drug resistance, cell growth/proliferation, malignant transformation and DNA repair through its regulation of transcription and translation, and its suppression of oxidative stress (10, 46). Increased expression of the YB-I gene has been shown to induce both the development of breast cancers of many histological types and genome instability in an experimental animal model, suggesting that YB-I has oncogenic activity (47). A recent study by de Souza-Pinto et al. (2009) showed that YB-I depletion in human cancer cells increases mitochondrial DNA mutagenesis, suggesting YBl as a key candidate for mitochondrial mismatch-binding protein (48). YB-I binds specifically to DNA/RNA that has been damaged or modified by DNA-damaging agents, including hyperoxide (28, 29). YB-I may thus play a protective role against various types of genotoxic damage, including oxidative DNA damage. Furthermore, it has been shown that the nuclear expression of YB-I is associated with the favorable outcome of patients with NSCLC (20, 49).

SUMMARY OF THE INVENTION

The present invention provides a method of predicting the efficacy of a drug, said method comprising determining a nuclear YB-I of a biological sample from a subject and predicting the efficacy of a therapy for treating said subject. In a preferred embodiment of the invention, the subject is a cancer patient. In another embodiment of the invention, the drug is anti-cancer drug.

The present inventors investigated whether YB-I expression in the nucleus was correlated with EGFR mutation in NSCLC. The present inventors investigated the expression of YB- 1 in the nucleus or cytoplasm by immunohistochemical analysis, and the absence or presence of EGFR mutations, in tumor samples from NSCLC patients (n = 104) who had undergone surgery and received no neo-adjuvant or adjuvant chemotherapy.

Thirty patients harbored exon 19 deletion mutation (delE746-A750) and exon 21 point mutation (L858R) in the EGFR gene. The present inventors discovered that in these 30 patients, nuclear YB-I expression showed a significant inverse correlation with the EGFR mutations. Patients showing nuclear YB-I expression were refractory to gefitinib therapy as compared with those showing non-nuclear YB-I expression. YB-I in the nucleus might alleviate somatic mutations in the EGFR gene in

NSCLC.

The present inventors also examined whether 8-hydroxy-2'-deoxyguanosine (8-OHdG), a representative oxygen nucleotide of DNA, could play a role in activating mutations of the EGFR gene, and also whether Y-box binding protein- 1 (YB-I) and 8-oxoguanine DNA glycosylase (OGGl) that are involved in repair of oxidative stimuli-induced DNA damages could play any role in EGFR activating mutations. Immuno histochemistry was used to evaluate the expression of 8-OHdG, YB-I and OGGl in patients with NSCLC (n=170). The present inventors analyzed mutations of delE746-A750 and L858R in the EGFR gene using peptide nucleic acid- locked nucleic acid PCR clamping and discovered that in NSCLC patients nuclear 8-OHdG expression is strongly associated with these EGFR mutations. Furthermore, nuclear expression of YB-I was inversely associated with EGFR mutations, but nuclear expression of OGGl was not. Among 51 patients who were treated with gefitinib, progression- free survival was substantially better when 8-OHdG expression was positive, as well as when EGFR mutations were present, and when nuclear YB-I expression was negative. Thus, activating mutations of the EGFR gene in NSCLC were closely associated with a decrease in the damage repair process for 8-OHdG in oxidized DNA.

On the basis of the above findings, the present invention provides followings:

[ 1 ] A method of predicting the efficacy of an anti-cancer drug, said method comprising the steps of:

(a) detecting the expression of Y-box binding protein- 1 or presence of 8-hydroxy-2'-deoxyguanosine in the nucleus of a cell included in a biological sample from a subject; and

(b) predicting the efficacy of a therapy for treating the cancer of said subject based on the result of said detecting step of (a).

[2] The method of [1], wherein said anti-cancer drug is an EGFR targeting drug. [3] The method of [1], wherein said anti-cancer drug displays different efficacy depending on the presence of EGFR gene mutation in cancer.

[4] The method of [1], wherein said cancer is an EGFR-related cancer.

[5] A method of predicting the prognosis of a cancer patient upon administration of an anti-cancer drug, said method comprising the steps of: (a) detecting the expression of Y-box binding protein- 1 or presence of

8-hydroxy-2'-deoxyguanosine in the nucleus of a cell included in a biological sample from the patient; and

(b) predicting the prognosis of the cancer patient upon administration of the anti-cancer drug based on the result of said detecting step of (a). [6] The method of [5], wherein said anti-cancer drug is an EGFR targeting drug.

[7] The method of [5], wherein said anti-cancer drug displays different efficacy depending on the presence of EGFR gene mutation in cancer.

[8] The method of [5], wherein said cancer is an EGFR-related cancer.

[9] A method of predicting the sensitivity of a subject to an anti-cancer drug, said method comprising the steps of:

(a) detecting the expression of Y-box binding protein- 1 or presence of 8-hydroxy-2'-deoxyguanosine in the nucleus of a cell included in a biological sample from the subject; and

(b) predicting the sensitivity of the subject to the anti-cancer drug based on the result of said detecting step of (a).

[ 10] The method of [9] , wherein said anti-cancer drug is an EGFR targeting drug. [11] The method of [9], wherein said anti-cancer drug displays different efficacy depending on the presence of EGFR gene mutation in cancer. [12] A method of predicting the presence of an EGFR gene mutation in a cell, said method comprising the steps of:

(a) detecting the expression of Y-box binding protein- 1 or presence of 8-hydroxy-2'-deoxyguanosine in the nucleus of a cell included in the cell; and

(b) predicting the presence of the EGFR gene mutation based on the result of said detecting step of (a).

[13] The method of [12], wherein said EGFR gene mutation is the deletion of the amino acids in positions 746 to 750 of human EGFR peptide or substitution of leucine in position 858 of human EGFR peptide with arginine.

The present invention provides an effective tool to predict the presence of an

EGFR gene mutation in a cell obtained from a subject by detecting nuclear expression of YB-I or 8-OHdG The present invention also provides effective means to predict the efficacy of an anti-cancer drug, the prognosis of a cancer patient upon administration of said drug, and the sensitivity of a subject to said drug in advance of administration of said anti-cancer drug.

BRIEF DESCRIPTION OF THE DRAWINGS

Figure 1. Histologic findings and expression of YB-I in human non-small cell lung cancer. YB-I expression showed two patterns: nuclear positive or negative. Figure 2. Kaplan-Meier estimators for the first and second progression-free survival (PFS) periods according to nuclear YB-I expression and EGFR mutation in 26 patients with non-small cell lung cancer treated with gefitinib after the first PFS. The first PFS was in the absence of gefitinib and the second PFS was in its presence. Nuclear YB-I is a predictive marker in the absence of gefitinib (upper left panel), as well as in its presence (upper right panel). EGFR mutation is not a predictive marker in the absence of gefitinib (lower left panel), whereas it is a predictive marker in its presence (lower right panel.).

Figure 3. Typical histological findings of squamous cell carcinoma and adenocarcinoma. Immunohistochemistry with anti-EGFR antibody for two samples of each squamous cell carcinoma and adenocarcinoma (A and B). Two samples with or without activating EGFR mutation are presented

Figure 4. Examples of immunohistochemistry showing negative and positive expressions of 8-OHdG (A), YB-I (B) and OGGl (C) in NSCLC. (A) Expression of 8-OHdG in four clinical specimens: 8-OHdG negative (no or weak immuno staining) (a, b) and 8-OHdG positive (strong immuno staining) (c, d). (B) Nuclear expression of YB-I in two clinical samples: negative samples show no or weak immunostaining in the nucleus, and nuclear YB-I -positive samples show strong immunostaining in the nucleus, as indicated by arrowheads. (C) OGGl in two clinical specimens: defined as positive (a) and negative (b). Figure 5. Kaplan-Meier estimate of progression-free survival from the start of gefitinib therapy in relation to 8-OHdG expression (A) and nuclear YB-I expression (B) and EGFR mutation (C) in 51 patients who received the drug for NSCLC recurrence after surgical resection. HR, hazard ratio; CI, confidence interval.

DETAILED DESCRIPTION OF THE INVENTION

Hereinafter, the present invention will be described in details.

The present invention relates to a pre-administration examination of a pharmaceutical agent whereby the efficacy of an Epidermal Growth Factor Receptor (EGFR) targeting drug can be estimated based on the presence or absence of YB-I in the nucleus of a cell.

Gefitinib (Iressa ^® ) is a drug used for treating lung cancer which specifically inhibits the tyrosine-kinase activigy of EGFR. It is known that patients having mutations in EGFR gene are more sensitive to Gefitinib and that Gefitinib achieves 70-80% shrinkage of tumor in NSCLC patients with EGFR gene mutations. It is also known that Gefitinib shows higher tumor shrinkage ratio in Oriental ethnics, females, adenocarcinoma cells and non-smokers in which EGFR mutations are found with higher frequencies.

To date, the EGFR mutations are detected by PCR amplification, however, the sensitivity of the detection is not satisfying. Further, performing PCR at bedside is impractical. Indeed, such molecular examinations are assigned to external institutions. Therefore, PCR amplification does not provide an efficient tool to detect EGFR gene mutation with ease and high sensitivity at bedside. In view of the above circumstances, the present inventors extensively studied to develop methods of predicting efficacy of EGF receptor targeting drugs such as Gefitinib and the sensitivity of a patient to such drugs by detecting the presence of YB-I expression or 8-OHdG in the nucleus a cell included in a biological sample from a subject. As exemplified in the Examples below, the presence of YB-I nuclear expression in lung cancer patient is inversely-correlated with the presence of EGFR mutations. Patients with YB-I nuclear expression shows less favorable prognosis upon Gefitinib administration. The presence of YB-I protein in nuclei can easily be detected by subjecting a cancer cell obtained from the patient to immunostaining, thereby an easy and sensitive examination tool can be provided.

In one aspect, the present invention provides a method for predicting responsiveness to a drug in a subject, where subject has a disease or disorder amenable to treatment with the drug and the drug is, for example, an anti-cancer drug. The method generally includes the following steps: (1) isolating a biological sample that includes nuclear and/or nuclear proteins from the subject; and (2) analyzing the biological sample to determine the presence or absence of the YB-I protein; whereby the absence of YB-I or reduced level of YB-I is indicative of an increased responsiveness to the drug in the subject. In certain embodiment of the invention, the drug is an anti-cancer agent such as, e.g., Gefitinib or Erlotinib.

The Y-box binding protein 1 (YB-I), a cold shock domain-containing ancestronic protein, plays essential roles in DNA damage repair; and in the transcription and regulation of various genes in the nucleus (10, 11). In the nucleus, YB-I repairs DNA damage induced by cisplatin and radiation, and promotes the transcription of drug resistance-relevant genes such as those for MDR1/ABCB1, a multidrug resistance-related ATP-binding cassette transporter, and MVP/LRP, a drug-resistance-related vault protein, suggesting that YB-I could be a global drug resistance biomarker (10, 11). Nuclear translocation of YB-I in cancer cells is often stimulated by DNA-damaging agents or growth factors, and is dependent on protein kinase C (12) and Akt/PI3k (13, 14). Concerning the possible involvement of YB-I in tumor growth, blocking of the nuclear translocation of YB-I results in marked decreases of EGFR and HER2, and also the expression of EGFR family proteins in breast cancer cells in culture (15). YB-I knockdown also down-regulates the expression of EGFR and HER2 in breast cancer (16). In a clinical study of breast cancer patients, nuclear YB-I expression was shown to be significantly associated with the expression of HER2, but not that of EGFR (16, 17). Taken together, these findings suggest that nuclear YB-I plays a key role in the expression of not only global drug resistance genes but also some EGFR family protein genes.

Nuclear expression of YB-I significantly affects the survival of cancer patients with various malignancies, including ovarian cancer (18, 19), non-small cell lung cancer (NSCLC) (20), breast cancer (16) and pediatric glioblastoma (21). As most of these patients show a close association between nuclear YB-I expression and poor outcome, it might be expected that YB-I plays a role in tumor growth and proliferation, as well as the acquisition of global drug resistance (11). In the present study of patients with NSCLC, the present inventors further investigated whether nuclear YB-I expression could affect survival, and also the incidence of EGFR mutations. As a result of this study, the present inventors discovered that the absence of YB-I expression or presence of 8-OHdG in the nucleus of a cell, in particular a cell obtained from lung cancer, is highly correlated with the presence of EGFR gene mutations. The present inventors also discovered that the absence of YB-I expression or presence of 8-OHdG in the nucleus of a cell, in particular a cell obtained from lung cancer, is correlated with a better prognosis of cancer patients undergone anti-cancer drug administration.

Therefore in one aspect, the present invention relates to a method of predicting the efficacy of an anti-cancer drug by detecting the expression of YB-I or presence of 8-OHdG in the nucleus of a cell obtained from a subject, and predicting the efficacy of a therapy for treating the cancer of said subject based on the result of said detecting step. In this method, the efficacy of an anti-cancer drug may be estimated to be higher to a subject with the absence of YB- 1 expression or presence of 8-OHdG compared with another subject positive for YB-I expression or negative for 8-OHdG Vice versa, the efficacy may be estimated to be lower to a subject with the presence of YB-I expression or absence of 8-OHdG compared with another subject being negative for YB-I expression or positive for 8-0HdG

In another aspect, the present invention relates to a method of predicting the prognosis of a cancer patient upon administration of an anti-cancer drug, by detecting the expression of YB-I or presence of 8-OHdG in the nucleus of a cell included in a biological sample from the patient, and predicting the prognosis of the cancer patient upon administration of the anti-cancer drug based on the result of said detecting step. In this method, the prognosis of a patient upon administration of the anti-cancer drug may be estimated to be better with the absence of YB-I expression or presence of 8-OHdG compared with that of another patient being positive for YB-I expression or negative for 8-0HdG Vice versa, the prognosis of a patient may be estimated to be worse with the presence of YB- 1 expression or absence of 8-OHdG compared with that of another patient being negative for YB-I expression or positive for 8-OHdG

In still another aspect, the present invention relates to a method of predicting the sensitivity of a subject to an anti-cancer drug by detecting the expression of YB-I or presence of 8-OHdG in the nucleus of a cell included in a biological sample from the subject, and predicting the sensitivity of the subject to the anti-cancer drug based on the result of said detecting step. In the above method, the sensitivity of a subject to an anti-cancer drug may be estimated to be higher with the absence of YB-I expression or presence of 8-0HdG compared with that of another subject being positive for YB-I expression or negative for 8-OHdG Vice versa, the sensitivity of a subject to an anti-cancer drug may be estimated to be lower with the presence of YB-I expression or absence of 8-OHdG compared with that of another subject being negative for YB-I expression or positive for 8-OHdG

In a further aspect, the present invention relates to a method of predicting the presence of an EGFR gene mutation in a cell by detecting the expression of Y-box binding protein- 1 or presence of 8-hydroxy-2'-deoxyguanosine in the nucleus of the cell, and predicting the presence of the EGFR gene mutation based on the result of said detecting step. In this method, an EGFR gene mutation may be estimated to be present with the absence of YB-I expression or presence of 8-OHdG whereas an EGFR gene mutation may be estimated to be absent with the presence of YB-I expression or absence of 8-0HdG

As used herein, the "anti-cancer drug" is any of pharmaceutical agents used for treating cancer of cancer patients that have the functional property of inhibiting a development or progression of a neoplasm in animals including human, particularly a malignant (cancerous) lesion. Inhibition of metastasis is frequently a property of anti-cancer drug. In some embodiments, the anti-cancer drug is an EGFR targeting drug. The "EGFR targeting drug" may be any agent as long as being capable of specifically binding to EGFR proteins and modifies the signal transduction thereof. Preferably, the EGFR targeting drug displays different efficacy depending on the presence of EGFR gene mutation in cancer. As already mentioned above, it is known that patients with NSCLC harboring mutations in the EGFR gene, including delE746-A750 and L858R, are more sensitive to therapy with EGFR targeting drugs such as gefitinib and erlotinib, in comparison with those harboring wild-type EGFR. Non-limiting examples of the EGFR targeting drug include gefitinib

(Iressa ^® ), erlotinib (Tarceva ^® ), cetuximab (Erbitux ^® ) and panitumumab (Vectibix ^® ), among which gefinitib and erlotibib may be advantageously used in the method of the present invention. As used herein, the term "efficacy" of the anti-cancer drug is the degree or level of the ability of said drug to inhibit the cellular growth or metastasis of the cancer. In this regard, when the growth rate of the cancer in subject A decreases with a higher ratio compared with that of the same type of cancer in subject B upon administration of an anti-cancer drug, the efficacy of said drug to subject A is estimated to be higher compared with that to subject B. As such, the term "predicting the efficacy of an anti-cancer drug" means to forecast or estimate the degree or level of the above described ability of the drug, in advance of the administration thereof to the subject.

As used herein, "Y-box binding protein 1 (YB-I)" refers to a cold shock domain-containing ancestronic protein which plays critical roles in the repair of DNA damage and transcriptional regulation of various genes that are closely associated with global cancer drug resistance and cell proliferation.

As used herein, "8-hydroxy-2'-deoxyguanosine (8-OHdG)" is a substance produced through oxidization of deoxy guanosine (dG), a component of DNA molecule, by free radicals such as reactive oxygen. The presence of 8-OHdG in DNA has been implicated in the type of somatic mutations found in human cancers (35).

The terms "patient" and "subject" include human and animal subjects. The terms

"mammal" and "animal" for purposes of treatment refer to any animal classified as a mammal, including humans, domestic and farm animals, and zoo, sports, or pet animals, such as dogs, horses, cats, cows, etc. Preferably, the mammal is human.

As used herein, the term "biological sample" refers to any liquid or solid material that is obtained from a biological source, preferably from an animal, most preferably from a human. A biological sample may be a cell sample or the lysate thereof and examples of such biological samples include blood, serum, urine, saliva, mucosa, feces and resected tissues obtained from a subject or cancer patient. The term "expression" of YB-I refers to the presence of any expression products of the YB-I gene within the nucleus of a cell. As used herein, the term "expression products" of the YB-I gene refers to any transcriptional or translational products of the YB-I gene, including messenger RNA (mRNA), pre-mRNA such as hetero nuclear RNA (hnRNA), peptides or peptide fragments, proteins and metabolites thereof.

The expression of YB-I or presence of 8-OHdG can be detected through a variety of methods generally used in the field of molecular biology, including immunohistochemistry, Enzyme Linked Immuno Sorbent Assay (ELISA), Biacore method, Western blotting, Southern blotting, Northern blotting, Polymerase Chain Reaction (PCR), Reverse Transcriptional PCR (RT-PCR), microarrays and the like. Examples of such detection methods are described in detail in "Sambrook & Russell, Molecular Cloning: A Laboratory Manual, Cold Spring Harbor Laboratory Press 2001 ". Among these methods, immunohistochemistry may be advantageously employed.

An anti-YB-1 antibody may be used to detect YB-I protein or fragment peptide thereof. Such an anti-YB-1 antibody may be either commercially obtained or produced in accordance with the description of Basaki. Y, et al., (Oncogene 26: 2736-46, 2007).

An anti-8-OHdG antibody may be used to detect 8-OHdG protein or fragment peptide thereof. Such an anti-8-OHdG antibody may be purchased from Japan Institute for the Control of Aging (Shizuoka, Japan: Cat No. MOG-IOOP).

The term "predict" as used herein means to forecast or estimate the presence, degree or level of the objective characteristics such as the efficacy of a drug, prognosis of a subject, or the sensitivity of a subject to a drug in advance of the administration of an anti-cancer drug, or to forecast or expect the presence of EGFR gene mutation without directly detecting said mutation, based on the result of nuclear YB-I or 8-OHdG detection. The result of nuclear YB-I or 8-OHdG detection may be compared with the statistical results obtained in the following examples, or may be integrated to the population used in the statistical analyses in the following examples so as to obtain a statistical result with higher accuracy from the thus enlarged population. For statistical processing, Kaplan-Meier method, Cox poroprtional hazards models or the like can be employed. As exemplified in the following examples, the hazard ratio of nuclear YB-I positivity over negativity upon gefitinib treatment is estimated to be 4.07-4.80 (PO.005) using Cox poroprtional hazards model whereas that of nuclear 8-OHdG positivity over negativity is estimated to be 0.34 (p=0.001). These hazard ratios indicate that patients positive for nuclear YB-I have about four time-higher risks of cancer progression than those negative for nuclear YB- 1 (p<0.005), whereas patients positive for nuclear 8-OHdG have about one third time-highr risks of cancer progression than those negative for nuclear 8-OHdG (p=0.001).

The term "treat", "treating" or "treatment" refers to both therapeutic treatment and prophylactic or preventative measures, wherein the object is to prevent or slow down

(lessen) an undesired physiological change or disorder, such as the development or spread of cancer. For purposes of this invention, beneficial or desired clinical results of treatment include, but are not limited to, alleviation of symptoms, diminishment of extent of disease, stabilized (i.e., not worsening) state of disease, delay or slowing of disease progression, amelioration or palliation of the disease state, and remission (whether partial or total), whether detectable or undetectable. "Treatment" can also mean prolonging survival as compared to expected survival if not receiving treatment. Those in need of treatment include those already with the condition or disorder as well as those prone to have the condition or disorder or those in which the condition or disorder is to be prevented.

The cancer may be any cancer and examples thereof include, but not limited to, brain tumor, tongue cancer, pharynx cancer, lung cancer, breast cancer, esophageal cancer, gastric cancer, pancreas cancer, biliary cancer, gallbladder cancer, duodenal cancer, colon cancer, liver cancer, uterus cancer, ovary cancer, prostate cancer, kidney cancer, bladder cancer, rhabdomyosaroma, fibrosarcoma, osteosarcoma, chondrosarcoma, skin cancer and various types of leukemia (e.g., acute myeloid leukemia, acute lymphatic leukemia, chronic myeloid leukemia, chromic lymphatic leukemia, adult T-cell leukemia and malignant lymphoma). The cancers recited above may be primary, metastatic or complicated with other disease.

In one embodiment, the cancer may be an EGFR-related cancer. The term "EGFR-related cancer" refers to any cancer which is caused by an EGFR polypeptide, a cancer which is exacerbated by an EGFR polypeptide, a cancer which is contributed by an EGFR polypeptide, or a cancer which is associated with an EGFR polypeptide. Exemplary EGFR-related cancers include, but are not limited to, colorectal cancer, lung cancer (including, but not limited to, non-small cell lung cancer), breast cancer, kidney cancer, colon cancer, gastric cancer, brain cancer, bladder cancer, head and neck cancers, ovarian cancer, and prostate cancer. In certain embodiment, the EGFR-related cancer is lung cancer, in particular, non-small cell lung cancer.

The term "prognosis of cancer" as used herein refers to the likely course or progression of the cancer and the probable outcome of the treatment such as administration of an anti-cancer drug. It also includes the chances of recovery from the disease.

As used herein, the term "administration of an anti-cancer drug" refers to contacting the drug with a cancer cell in a patient.

The term "sensitivity of a subject to an anti-cancer drug" as used herein means the ability of the subject to show beneficial or desired clinical results of treatment in response to the administration of the anti-cancer drug. Therefore, the term "predicting the sensitivity of a subject to an anti-cancer drug" means to forecast or estimate the degree or level of the above described ability of the subject in advance of the actual administration of the drug. As used herein, the term "EGFR gene mutation" refers to any mutation included in the EGFR gene that can be found in cancer cells. Examples of the mutation include insertion, deletion, substitution and inversion of one or more nucleotides of the EGFR gene. These mutations may be any of silent, missense or nonsense mutations. In one embodiment, the EGFR gene mutation exists in either exon 19 or 21. In another embodiment, the EGFR gene mutation causes either the deletion of the amino acids in positions 746 to 750 (Glu-Leu-Arg-Glu-Ala) of human EGFR peptide (hereinafter referred to as "delE746-A750") or substitution of leucine in position 858 of the human EGFR peptide with arginine (hereinafter referred to as "L858R"). In this embodiment, the human EGFR peptide is preferably one of wild type. The sequences of SEQ ID NO: 1 and 2 each represent the nucleotide sequence of human wild type EGFR cDNA and the amino acid sequence of human wild type EGFR peptide. The presence of such mutations can be confirmed by peptide nucleic acid-locked nucleic acid (PNA-LNA) PCR clamp as described in Nagai Y. et al. (Cancer Res., 65: 7276 - 82, 2005).

As already mentioned above, the present invention is based on the finding that the absence of YB-I expression or presence of 8-OHdG in the nuclei of cells included in a biological sample from a cancer patient is highly associated with a favorable prognosis after administration of an anti-cancer drug such as gefitinib. In this regard, the present invention also provides a method of detecting the sensitivity of a subject to an anti-cancer drug, characterized in that detection of the expression of YB- 1 or presence of 8-OHdG in the nucleus of a cell obtained from the subject is performed, and the result of the detection is linked to the sensitivity of the subject to said anti-cancer drug. By linking the result of the detection to the sensitivity of the subject to an anti-cancer drug in a way that the absence of YB-I expression or presence 8-OHdG in the nucleus of the cell is linked to a higher sensitivity of the subject to the drug whereas the presence of YB-I expression or absence of 8-OHdG is linked to a lower sensitivity of the subject to the same drug, those who obtained the result of detection can readily determine whether the subject has a higher or lower sensitivity to the drug without using medical expertise. Also provided by the present invention is a kit for use in a method of predicting the efficacy of an anti-cancer, the prognosis of a cancer patient upon administration of an anti-cancer drug, the sensitivity of a subject to an anti-cancer drug, or the presence of an EGFR gene mutation in a cell, said kit comprising an anti-YB-1 or anti-8-OHdG antibody. The kit may also comprise one or more of agents necessary for visualizing the anti-YB-1 or anti-8-OHdG antibody, an instruction describing how to detect YB-I or 8-OHdG in cell nuclei and predict the efficacy of an anti-cancer, the prognosis of a cancer patient upon administration of an anti-cancer drug, the sensitivity of a subject to an anti-cancer drug, or the presence of an EGFR gene mutation in a cell based on the result of the detection.

EXAMPLES

Hereinafter, the present invention will be illustrated by way of specific examples, although the invention should not be limited thereto.

EXAMPLE 1:

Correlation Between the Absence of YB-I and the Presence of EGFR Gene Mutation

Patients and Methods Patients, tumor samples and treatment

The present inventors retrospectively examined with primary NSCLC whose tumors had been completely removed surgically at the Department of Surgery, Kurume University Hospital, between 1997 and 2004. Among these patients, 66 were diagnosed histologically as having adenocarcinoma, and the other 38 as having squamous cell carcinoma. The age of the NSCLC patients ranged from 41 to 82 years (median, 66), and 67 of them were men and 37 were women. None of the patients had received neo-adjuvant or adjuvant chemotherapy.

Among these patients, 26 received gefitinib therapy for recurrent disease after surgical resection between June 2003 and September 2008, with a median interval of 760 days (range: 225-3062 days) between surgery and gefϊtinib treatment. Five patients received gefitinib as the initial therapy, and the others as second- or third-line therapy (21 patients, platinum doublets as first line; 5 patients, monotherapy, non-platinum doublets, and platinum doublets as second line). Tumor response was evaluated after chemotherapy according to the RECIST (Response Evaluation Criteria for Solid Tumors) (22). This study was approved by the Institutional Review Board of Kurume University.

Antibodies and inimunohistochemistry (IHC)

Anti-YB- 1 antibody was generated as described previously (14). Paraffin-embedded tissue samples were cut at a thickness of 4 μm, mounted on coated glass slides, and with antibodies using the BenchMark XT (Ventana Automated Systems, Inc., Tucson, AZ, USA). IHC analysis was performed as described previously (16). The expression of YB-I protein was classified on the basis of strong expression in the nucleus or expression in the cytoplasm only. EGFR expression was classified into four categories: score 0, no staining at all, or membrane expression in <10% of cancer cells; score 1+, faint/barely perceptible partial membrane expression in >10% of cancer cells; score 2+, weak to moderate expression of the entire membrane in >10% of the cancer cells; score 3+, strong expression of the entire membrane in >10% of cancer cells. All IHC studies were evaluated by two IHC-experienced reviewers who were unaware of the conditions of the patients (M. Ka and A.K).

Peptide nucleic acid-locked nucleic acid (PNA-LNA) PCR clamp for EGFR

Mutations of the EGFR gene were examined in exons 19 (E746-A750del) and 21 (L858R) by peptide nucleic acid-locked nucleic acid (PNA-LNA) PCR clamp as described previously (23). In brief, genomic DNA was purified from paraffin-embedded tissues using a QIAamp DNA Micro kit (QIAGEN). The PCR primers employed were synthesized by Invitrogen Inc.

The PCR primers used for the detection of the mutation in exon 19 (delE746-A750) were as follows: 5'-GTGCATCGCTGGTAACATCC-S' (SEQ ID NO: 3) 5 '-TGAGGTTCAGAGCCATGGAC-S ' (SEQ ID NO: 4) The PCR primers used for the detection of the mutation in exon 21 (L858R) were as follows: 5'-GCATGAACTACTTGGAGGAC-S' (SEQ ID NO: 5)

5 '-ACCTAAAGCCACCTCCTTAC-3 ' (SEQ ID NO: 6)

PNA clamp primers and LNA mutant probes were purchased from FASMEC (Kanagawa, Japan) and IDT (Coralville, IA), respectively. The probe used for the detection of the mutation in exon 19 (delE746-A750) was as follows:

6-FAM / CTATCAAAACATCTCCGAAAGC (SEQ ID NO:7) / BHQl TET / CGCTATCAAGACATCTCCG (SEQ ID NO: 8) / BHQl The probe used for the detection of the mutation in exon 21 (L858R) was as follows:

6-FAM / TTTGGCCCGCCCAA (SEQ ID NO: 9) / BHQl

PNA-LNA PCR clamp was done using a SDS-7500 System (Applied Biosystems). PNA-LNA PCR clamp were carried out for 45 cycles between 95 °C for 3 sec and 56 °C for 30 sec.

Statistical analysis

Association of EGFR mutation and nuclear YB-I with histological type and clinicopathologic factors (age, sex, smoking status, and pathological stage) was tested by Fisher's exact test. Association between YB-I and EGFR mutation was also tested by

Fisher's exact test. In 26 patients who were treated with gefitinib after disease progression, the effects of nuclear YB-I and EGFR mutation on sensitivity to gefitinib were examined. Time until disease progression or death from the date of surgery was defined as the first progression- free survival (PFS). The second PFS was defined as the time until disease progression from the start of gefitinib treatment. Kaplan-Meier estimators for the first and second PFS were calculated according to nuclear YB-I and EGFR mutation, respectively. Log-rank tests were applied to examine the effects of nuclear YB-I and EGFR mutation on the first and second PFS. Hazard ratios for nuclear YB-I (or EGFR mutation)-positive patients over negative patients were estimated with the Cox proportional hazards models for the first and second PFS based on the same patients. The hazard ratio of nuclear YB-I (or EGFR mutation) for the second PFS was the effect of nuclear YB-I (or EGFR mutation) in the presence of gefitinib, while that for the first PFS was the effect in the absence of gefitinib. Thus by comparing the effects of nuclear YB-I (or EGFR mutation) on the first and second PFS, it was possible to examine the role of nuclear YB-I (or EGFR mutation) in gefitinib sensitivity. Statistical analysis was performed with SAS version 9.1 (SAS Institute Inc.) and R version 2.7.0.

Results In this study, 104 NSCLC patients were included, and their tumor histologies were classified as adenocarcinoma in 66 patients and squamous cell carcinoma in 38. Figure 1 shows representative examples of IHC staining for YB-I in both adenocarcinoma and squamous cell carcinoma, expression of YB-I being observed in both the nucleus and cytoplasm of the lung cancer cells. YB-I expression in the nucleus was evaluated as positive in 44 patients, and that in the cytoplasm as negative in 60 patients. In adenocarcinoma, YB-I was positive in 19 patients, and negative in 47. In squamous cell carcinoma, on the other hand, YB-I expression was positive in 25 patients and negative in 13. The clinical and pathological characteristics of the 104 patients analyzed in this study at the time of diagnosis are summarized in Table 1. In NSCLC patients, more than 90% of EGFR mutations are located in exons 19 (delE746-A750) and 21 (L858R point mutation) (9). hi our series of NSCLC patients overall, there was a significant correlation between EGFR mutation and histological type of cancer, patient age, sex, smoking status, and pathological stage (Table 1) in adenocarcinoma, but not in squamous cell carcinoma (data not shown). Nuclear YB-I expression also showed a significant inverse correlation with female gender (P=0.002), smoking (PO.001) and adenocarcinoma (PO.001) (Table 1).

Table 1. Characteristics of the 104 patients with non-small ceil lung cancer.

All patients EGFR status Nuclear YB-I

Wild Type Mutation Negative Positive n n n P value n n P value

Age (yr)

<66 48 28 20 P=0.009 30 18 P=0.428

> 66 56 46 10 30 26

Sex

Female 37 20 17 P=0.006 29 8 P=0.002

Male 67 54 13 31 36

Smoking Status

Never smoked 37 17 20 P<0.001 30 7 PO.001

Smoker 67 57 10 30 37

Histological Type

Adenocarcinoma 66 40 26 P=0.002 47 19 PO.001

Squamous cell carcinoma 38 34 4 13 25

Pathological Stage at Surgery

I 50 38 12 P=0.033 36 14 P=0.015

II 21 18 3 8 13 πi 33 18 15 16 17

On the basis of nuclear YB-I expression and EGFR mutation in the 104 patients, the present inventors observed that mutations of delE746-A750 and L858R was significantly less frequent in cancer cells with nuclear YB-I expression (positive) than in those with cytoplasmic YB-I expression (negative). Statistical analysis demonstrated a significant (pO.OOl) inverse correlation between EGFR mutation and nuclear YB-I expression in cancer cells from all NSCLC patients (Table 2).

Table 2 Correlation between nuclear YB-I expression and EGFR mutation status in patients with non-small cell lung cancer.

EGFR status Nuclear YB-I

Negative Positive n n P value

Wild type 35 3' <0.001*

Mutation 25 5

delE746-A750 17 4

L858R 8 1

P value is by Fisher's exact test for a 2 by 2 table

The characteristics of 26 patients who were treated with gefitinib after the first PFS are presented in Table 3, which shows the pathological stage at the start of gefitinib therapy, as well as that at the time of surgery. Twenty-four adenocarcinoma patients and 2 squamous cell carcinoma patients were administered gefitinib. At the start of gefitinib therapy, the pathological stage was Stage IV in 24 patients, while at the time of surgery all the patients had a pathological stage of less than Stage IV. One patient showed a complete response (CR), 8 showed a partial response (PR), 10 stable disease (SD), and 7 progressive disease (PD). The first PFS was observed for all patients, and 2 patients were censored for the second PFS. Kaplan-Meier estimators for the first and second PFS by nuclear YB-I and EGFR mutation are presented in Figure 2, and the p- values of log-rank tests and hazard ratios by the Cox proportional hazards model are summarized in Table 4. For the first PFS, the Kaplan-Meier estimator of patients who were positive for nuclear YB-I differed from that of patients who were negative (p=0.078, HR=2.19; 95% confidence interval (CI)

0.89-5.39), whereas that for mutant EGFR was close to that for wild-type EGFR (p=0.439, HR=0.72; 95% CI 0.31-1.66). Table 3. Characteristics of 26 patients with non-small cell lung cancer treated with gefitinib after the first progression-free survival

All patients EGFR status Nuclear YB-I

WUd Type Mutation Negative Positive n n n n n

Age (yr)

< 66 15 5 10 10 5

> 66 11 S 6 9 2

Sex

Female 18 7 11 14 4

Male 8 3 5 5 3

Smoking Status

Never smoked 19 6 13 16 3

Smoker 7 4 3 3 4

Histological Type

Adenocarcinoma 24 9 15 18 6

Squamous cell carcinoma 2 1 1 1 1

Pathological Stage at Surgery

I 13 5 8 11 2

II 3 2 1 1 2 in 10 3 7 7 3

Pathological Stage at Start of Gefitinib

UI 2 1 1 1 1

IV 24 9 15 18 6

Table 4 Log rank tests and estimate of hazard ratios by the Cox regression model for the first progression-free survival (PFS) and second PFS in 26 patients treated with gefitinib after the first PFS.

RretPFS Second PFS Pvalue HR(95%CI) P value HR(95%CD

Nuclear YB-I P=0.078 2.19(0.89-5.39) P=0.004 4.29(1.48-12.4) EGFRstatus P=0.439 0.72(0.31-1.66) P=0.066 0.45(0.19-1.07)

Abbreviation: 95% 0, 95% confidence interval.

P value is by the log rank test

Nuclear YB-I expression was thus expected to be a predictive marker even in the absence of gefitinib, whereas EGFR mutation was not. Both nuclear YB-I expression and EGFR mutation affected the second PFS (P=0.004, HR=4.29; 95% CI 1.48-12.4 for nuclear YB-I, and P=0.066, HR=0.45; 95% CI of 0.19-1.07 for EGFR mutation), indicating that both are predictive markers in the presence of gefitinib. The hazard ratio of nuclear YB-I positivity over negativity for the first PFS was estimated at 2.19 and that for the second PFS at 4.29. Thus, nuclear YB- 1 expression appeared to be a predictive marker even in the absence of gefitinib, and the larger HR for the second PFS compared to the first suggested that nuclear YB-I may affect sensitivity to gefitinib.

Discussion

In this study the present inventors demonstrated a close correlation between mutations in the EGFR gene and nuclear expression of YB- 1 in NSCLC. Mutations identified in this study included a small in- frame deletion (del E746-A750) in exon 19, and a missense mutation (L858R) in exon 21 of the EGFR gene, both of which are highly sensitive to the therapeutic effects of EGFR-targeting drugs such as gefitinib and erlotinib (5-7). These mutations in NSCLC are well known to be significantly associated with female gender, and never having smoked (24). Toyooka and colleagues examined the impact of gender and smoking status on the mutational spectrum of the EGFR gene in NSCL (n=1467). They found that, in females, mutations in exons 19 and 21 were significantly less frequent in ever smokers than in never smokers, whereas in males, mutations in exons 19, 21, and 18 were significantly less frequent in ever smokers than in never smokers, suggesting no gender-related difference among never smokers (25). Consistent with these findings, our present study showed that EGFR mutations were also significantly correlated with smoking status and female gender (Table 1). Furthermore, nuclear YB-I expression was significantly correlated with non-smoking status, female gender and adenocarcinoma.

Almost all previous studies have suggested it is unlikely that smoking plays a critical role in EGFR mutations that are susceptible to EGFR-targeting therapeutics (24, 25), or in nuclear YB-I expression. Bell and colleagues examined the association of sex-related factors in the increased prevalence of EGFR mutations in women, and concluded that nine polymorphisms in the 7 genes involved in the estrogen biosynthesis and metabolism pathway are unlikely to be major genetic modifiers of the prevalence of EGFR mutation in NSCLC (26). So far, it remains unclear which factor is specifically responsible for the increased prevalence of EGFR mutations in women, never-smokers and non-caucasians.

The present inventors found that EGFR mutations were inversely correlated with nuclear YB-I expression in NSCLC. With regard to how nuclear YB-I expression affects EGFR mutations in this cancer, the molecule plays a role in transcription of the ABCB1/MDR1 gene, encoding P-glycoprotein, and the MVP/LRP gene, in response to environmental stimuli including radiation, DNA-damaging agents, and anticancer agents (10, 11). Moreover, through its exo-nuclease activity, YB-I specifically binds to DNA or RNA that has been damaged by radiation or oxidative stress (27, 28, 29), and it also interacts specifically with DNA-repair-related proteins such as PCNA, p53 and HMGBl when cancer cells are stressed by DNA-damaging agents (30, 31). YB-I knockdown sensitizes cancer cells to DNA-damaging anticancer agents and hyperoxide, and inhibits malignant cell transformation (32, 33). Thus, YB-I in the nucleus is expected to scavenge damaged DNA/RNA, plausibly through its interaction with PCNA, HMGBl and p53. This anti-mutagenic effect of YB-I might play a role in reducing the extent of EGFR mutation. At present, however, this possibility must remain highly speculative until it is further clarified how YB-I alleviates EGFR mutation in lung cancer cells.

In conclusion, the present inventors have presented for the first time evidence that EGFR mutations are inversely correlated with nuclear expression of YB-I in NSCLC. Our study also demonstrated that nuclear YB- 1 expression desensitized NSCLC to the therapeutic effect of gefitinib when PFS was evaluated in the absence and presence of gefitinib after disease progression. Our results suggest that EGFR mutation is a predictive marker for PFS in the presence of gefitinib, but not in its absence, whereas nuclear expression of YB-I may be a predictive marker irrespective of the presence of gefitinib.

EXAMPLE 2:

Correlation Between the Absence of YB-I or Presence of 8-OHdG and the Presence of

EGFR Gene Mutation

Patients, tumor samples and treatment

The present inventors retrospectively examined 170 patients with primary NSCLC whose tumors had been completely removed surgically at the Department of Surgery,

Kurume University Hospital between 1995 and 2005. Among these patients, 102 were diagnosed histologically with adenocarcinoma, and the other 68 with squamous cell carcinoma (Table 5). The age ranged from 41 to 82 years with a median of 68 years. Two patients had received adjuvant chemotherapy, and the other patients had not received neo-adjuvant or adjuvant chemotherapy.

Among 170 patients, 51 had received gefitinib therapy for the recurrence after surgical resection between July 2002 and February 2009. Thirteen patients had received gefitinib as the initial therapy, and the others as second- or third-line therapy (33 patients, platinum doublets as first line; 5 patients, monotherapy, non-platinum doublets, and platinum doublets as second line). Tumor response was evaluated after chemotherapy according to the Response Evaluation Criteria for Solid Tumors (RECIST). This study was approved by the Institutional Review Board of Kurume University.

Antibodies and immunohistochemistry (IHC)

Paraffin-embedded tissue samples were cut at a thickness of 4 μm and examined on coated glass slides, and labeled with the following antibodies using the BenchMark XT (Ventana Automated Systems, Inc., Tucson, AZ, USA). Anti-8-OHdG antibody was obtained from the Japan Institute for the Control of Aging (Shizuoka, Japan: Cat No. MOG-IOOP). Anti-YB-1 polyclonal antibody was generated against a 15-aminoacid synthetic peptide in the COOH-terminal domain, and this antibody was used at a working dilution of 1 :2000. OGG 1 used the ChemMate ENVISION method (DakoCytomation,

Glostrup, Denmark). Endogenous peroxidase activity was inhibited by incubating the slides in 3% H ₂ O ₂ for 5 min. Each slide was heat-treated using Target Retrieval Solution, pH 9.0 (DAKO, Glostrup, Denmark) for 30 min, and incubated with the antibody at 4 °C overnight. IHC analysis was performed as described previously. Positive and negative controls were used for each section. Figure 3 shows representative images of H&E and IHC with anti-EGFR antibody for two samples of each squamous cell carcinoma and adenocarcinoma. Two samples with or without activating EGFR mutation are presented.

The intensity of nuclear-positive cancer cells of 8-OHdG was expressed as follows: none, weak or strong (Fig. 4A). the present inventors classified 8-OHdG nuclear expression of none and weak as negative (Fig. 4 A, a and b) and strong as positive (Fig. 4 A, c and d). YB-I was expressed in the cytoplasm alone, in both nuclei and cytoplasm or in cytoplasm in NSCLC (Fig. 4B). The extent of staining of nuclear YB-I and OGGl was classified based on cells with strongly stained nuclei; 5% or more tumor cells had nuclear-positive YB-I or OGGl (Fig. 4B, c and d, and 4C, a), and less than 5% had nuclear negative (Fig. 4B, a and b, and 2C, b) (29). All immunohistochemical studies were evaluated by two experienced observers who were blind to the condition of the patients (A.K., M.Ka).

Peptide nucleic acid-locked nucleic acid (PNA-LNA) PCR clamp for EGFR mutation

Mutations of the EGFR gene were examined in exons 19 (delE746-A750) and 21 (L858R) by PNA-LNAPCR clamp as described in Example 1.

Statistical analysis Histological type and clinicopatho logic factors (age, gender, smoking status, and pathological stage) by EGFR mutation and 8-OHdG expression were tested by Fisher's exact test. The association between EGFR mutations and 8-OHdG, and also between 8-OHdG and YB-I was tested by Fisher's exact test. It was also applied to examine whether they are associated with the nuclear expression of YB- 1 and OGGl . In 51 patients who were treated with gefitinib after recurrence, the effects of 8-OHdG, EGFR mutation and nuclear YB-I on progression- free survival (PFS) were examined. PFS was defined as the time until disease progression from the start of gefitinib treatment. Kaplan-Meier estimators for PFS were calculated according to 8-OHdG, EGFR mutation and nuclear YB-I, respectively. Log-rank tests were applied to examine the effects of 8-OHdG, EGFR mutation and nuclear YB- 1 , and hazard ratios and 95% confidence intervals were estimated with the Cox proportional hazards models. Statistical significance was declared if the two-sided P value was less than 0.05. Statistical analysis was performed with SAS version 9.1 (SAS Institute Inc., Cary, NC), R version 2.8.1 and StatXact 7 (Cytel Inc., Cambridge, MA).

Results

Nuclear expression of 8-OHdG was associated with activating mutations of the

EGFR in NSCLC

Figure 4 shows representative examples of IHC staining for 8-OHdG (Fig. 4A), YB-I (Fig. 4B), and OGGl (Fig. 4C) in NSCLC. IHC staining showed very clear differences between positive and negative expressions of 8-OHdG, YB-I and OGGl in the nucleus. The clinical and pathological characteristics at the time of diagnosis according to EGFR mutations are summarized in Table 5. Among the 170 patients, 48 (28.2%) harbored activating mutations of the EGFR; delE746-A750 and L858R mutations were observed in 26 and 22 patients, respectively. None showed simultaneous mutations at the two loci. There were higher proportions of younger patients, females, and nonsmokers among patients with the EGFR mutation. Adenocarcinoma was also more frequent in patients with the EGFR mutation. All of these differences were statistically significant. In 102 cases of adenocarcinoma, 24 cases of pure bronchioalveolar carcinomas and 78 cases of invasive adenocarcinomas were observed and there was no minimally invasive bronchioalveolar carcinoma. EGFR mutations of pure bronchioalveolar carcinomas and invasive adenocarcinomas were observed in 11/24 (45.8%) and 31/ 78 (39.7%), respectively. Of 170 patients, 97 patients were found to be positive for 8-OHdG expression (57.1% of the total). The predominance of younger patients, males, smokers, and adenocarcinomas was evident in patients positive for 8-OHdG expression (Table 5). On the other hand, more patients showing positivity for 8-OHdG were at the earliest stage. On the basis of 8-OHdG expression and EGFR mutation in the 170 patients, EGFR mutation showed significantly (.PO.001) higher frequently in patients positive for 8-OHdG expression than in those who were negative (Table 6).

Table S. Characteristics of 170 patients with NSCLC according to EGFR status and 8-OHdG expression.

EGFR status 8-OHdG

Wild type Mutation Positrve Negative

Characteristic P value P value (n=122) (n=48) (n=97) (n=73)

Age

≤65 44 30 0.002 49 25 0.042

>65 78 18 48 48

Gender

Female 31 30 O.001 44 17 0.004

Male 91 18 53 56

Smoking status

Never 29 33 <0.001 44 18 0.006

Ever 93 15 53 55

Histological type

Squamous cell carcinoma 62 6 <0.001 28 40 <0.001

Adenocarcinoma 60 42 69 33

BAC 13 11 0.640 17 7 0.806

Non-BAC 47 31 52 26

Pathological stage

I 57 16 0.257 49 24 0.010

II 29 13 16 26

III 36 19 32 23

Table 6. Correlation between EGFR mutation status and 8-OHdG expression.

8-OHdG

Positive Negative

EGFR status (n=97) (n=73) P value

Wild type (n=122) 58 (59.8%) 64 (87.7%) <0.001

Mutation (n=48) 39 (40.2%) 9 (12.3%) delE746-A750 22 4

L858R 17 5

Nuclear YB-I expression was associated with EGFR mutations, but not with 8-OHdG Since YB-I or OGGl is expected to be involved in the repair process for oxidized

DNA, the present inventors next examined whether the nuclear expression of YB-I or OGGl was associated with the 8-OHdG expression or EGFR mutations. The present inventors observed that those with delE746-A750 or L858R mutation in the EGFR gene had a significantly (PO.001) lower prevalence of the nuclear positivity for YB-I expression than those who were negative (Table 7). Concerning the possible correlation between 8-OHdG and nuclear YB- 1 expression, nuclear YB- 1 expression was found to be positive in 32 (33.0%) and negative in 65 (67.0%) of 8-OHdG-positive patients (n=97) while there was similar number of YB-I positive in 34 (46.6%) and YB-I negative in 39 (53.4%) of 8-OHdG-negative patients (n=73). However, the correlation between 8-OHdG expression and nuclear YB- 1 expression was just short of statistical significance (P=0.082). In contrast, OGGl expression was not statistically significantly associated with 8-0HdG expression or EGFR mutation (Table 7). Table 7. Correlation of 8OHdG expression and EGFR mutation with nuclear expression of YB l and OGGl.

8-OHdG EGFR status

Positive Negative Wild type Mutation (n=97) (n=73) P value (n=122) (n=48) P value

Nuclear YB-I

Positive (n=66) 32 (33.0%) 34 (46.6%) 0.082 58 (47.5%) 8 (16.7%) O.001 Negative (n=104) 65 (67.0%) 39 (53.4%) 64 (52.5%) 40 (83.3%)

OGGl

Positive (n=43) 28 (28.9%) 15 (20.5%) 0.285 30 (24.6%) 13 (27.1%) 0.845 Negative (n=127) 69 (71.1%) 58 (79.5%) 92 (75.4%) 35 (72.9%)

Correlation of 8-OHdG and EGFR mutations with progression-free survival in patients who received gefitinib

Among 51 patients who received gefitinib for recurrence after surgical resection, tumors in 46 were histologically diagnosed as adenocarcinoma, and the other 5 as squamous cell carcinoma. Fifteen patients were men and 36 were women. Fifteen were smokers and 36 were non-smokers. Six patients were classified as stage III, and the other 45 as stage IV at the start of gefitinib therapy. The response rate in patients with mutant EGFR was 10/27=37.0% (10 PR, 12 SD and 5 PD), whereas that in patients with wild-type EGFR was 6/24=25.0% (6 PR, 11 SD, 7 PD). The response rate in patients with nuclear YB-I positive was 4/11=36.4% (4 PR, 2 SD and 5 PD), whereas that in patients with nuclear YB-I negative was 12/40=30.0% (12 PR, 21 SD, 7 PD). The estimated product- limit survival functions of 8-OHdG and EGFR mutation with respect to the progression-free period from the start of gefitinib therapy are shown in Figure 5. Progression- free survival was distinctly better in patients who were 8-OHdG positive than in those who were negative (hazard ratio: 0.34, 95%CI: 0.18-0.65, PM).OO1) (Fig. 5A). Furthermore, patients who were nuclear YB-I positive had a shorter progression-free period than those nuclear YB-I negative (hazard ratio: 4.03, 95%CI: 1.86-8.77, PO.001) (Fig. 5B). Patients with EGFR mutation also showed a significantly longer progression- free period than those without the mutation (hazard ratio: 0.29, 95% CI: 0.15-0.57, PO.001) (Fig. 5C). By multivariate Cox regression analysis, even adjusting for possible confounding factors of age (=65, >65), gender and smoking status, 8-OHdG expression (hazard ratio: 0.34, 95%CI: 0.17-0.68, P=0.002), EGFR mutation (hazard ratio: 0.28, 95%CI: 0.14-0.57, P<0.001) and nuclear YB-I (hazard ratio: 4.80, 95%CI: 2.10-10.99, PO.001) were found to be independent prognostic factors with regard to the progression- free period.

Discussion

In this Example, the present inventors demonstrated a strong association between mutations in the EGFR gene and the presence of elevated levels of 8-OHdG in patients with NSCLC. Mutations identified in this study included a small in-frame deletion (delE746-A750) in exon 19 and a missense mutation (L858R) in exon 21 of the EGFR gene, both of which are highly sensitive to the therapeutic effects of EGFR-targeting drugs, such as gefitinib and erlotinib (5-8). These mutations in NSCLC are well known to be significantly associated with female gender and never having smoked (24). Toyooka et al. (2008) examined the impact of gender and smoking status on the mutational spectrum of the EGFR gene in NSCLC (n=1467) and found that, in females, mutations in exons 19 and 21 were significantly less frequent in ever smokers than in never smokers, whereas in males, mutations in exons 19, 21, and 18 were significantly less frequent in ever smokers than in never smokers (25). Concerning the induction of 8-OHdG, smoking has been identified as an important factor. Although some studies concluded that 8-OHdG is a biomarker of oxidative stress associated with chemical exposure, including smoking, benzene, and asbestos, various occupational studies did not reveal higher levels of 8-OHdG in smokers (44). It remains to be further studied how 8-OHdG is induced in response to oxidative stress in lung cancer, and also how 8-OHdG could affect mutations in the EGFR gene in lung cancer. One possible mechanism whereby 8-OHdG affects EGFR mutations in NSCLC could be failure of the base excision repair process for eliminating oxidized DNA, thus resulting in augmentation of EGFR mutations and promotion of lung carcinogenesis. Two large independent case-control studies of lung cancer demonstrated that the rate of base excision repair of 8-OHdG was decreased in blood leukocytes of cancer patients in comparison with controls (50, 51). Furthermore, Speina et al. (2003) reported that repair capacity was significantly lower in blood leukocytes of lung cancer patients than in those of controls (52). Our findings suggest that decreased efficacy of base excision repair to eliminate 8-OHdG in oxidized DNA lesions may enhance not only the development of lung cancers, but also mutations in EGFR genes.

Our present study demonstrated that YB-I expression was inversely associated with EGFR-activating mutations in NSCLC. A protective effect of YB-I against genotoxic damage may explain the inverse relationship between EGFR mutation and nuclear YB-I expression.

YB- 1 shows much higher affinity for DN A/RN A that has been damaged by oxidation or genotoxic drugs than for undamaged DN A/RN A (10, 11), and this molecule interacts with the repair-related proteins PCNA, p53 and HMGBl to promote the repair of genotoxic damage.

In particular, YB-I harboring endonuclease III is considered to mediate base excision repair and strand separation of damaged DNA (53, 54). On the other hand, DNA base excision repair by OGGl and relevant molecules is known to be a major pathway for repair of oxidative DNA damage. Some genetic changes of OGGl are associated with increased risks of various human malignancies (40, 41). However, in our present study, OGGl expression was not significantly correlated with 8-OHdG expression or EGFR mutation status. The reparative property of YB-I might play a rather niore important role in activating EGFR mutations than that of OGGl.

The present study further demonstrated a substantially better prognosis after gefitinib treatment among NSCLC patients with 8-OHdG expression as well as among those with EGFR mutation. No previous study has evaluated the effect of gefitinib in relation to oxidative DNA damage, and the present inventors provide for the first time evidence that EGFR mutations are positively correlated with 8-OHdG expression in NSCLC. Our study also demonstrated that 8-OHdG expression as well as EGFR mutation was associated with non-smoking status and female gender. Thus, it is considered that 8-OHdG is a biomarker for mutagenesis in the EGFR gene, and could be used to optimize anti-cancer therapeutics employing EGFR-targeting drugs.

REFERENCES

1. Parkin DM: Global cancer statics in the year 2000. Lancet Oncol 2: 533-43, 2001

2. Spiro SG, Silvestri GA: One hundred tears of lung cancer. AM J Respir Crit Care Med 172: 523-9, 2005

3. Ohe Y, Ohashi Y, Kubota K et al: Randomized phase III study of cisplatin plus irrinotecan versus carboplatin plus paclitaxel, cisplatin plus gemcitabine, and cisplatin plus vinorelbine for advanced non-small-cell lung cancer: Four- Arm Cooperative Study in Japan. Ann Oncol 18: 317-323, 2007

4. Schiller JH, Harrington D, Belani CP et al: Comparison of four chemotherapy regimens for advanced non-small-cell lung cancer. New Engl J Med 346: 92-8, 2002 5. Lynch TJ, Bell DW, Sordella R, et al: Activating mutations in the epidermal growth factor receptor underlying responsiveness of non-small-cell lung cancer to gefitinib. N

Engl J Med 350: 2129-39, 2004 6. Paez JG, Janne PA, Lee JC, et al: EGFR mutations in lung cancer: correlation with clinical response to gefitinib therapy. Science 304: 1497-500, 2004 7. Pao W, Miller V, Zakowski M, et al: EGF receptor gene mutations are common in lung cancers from "never smokers" and are associated with sensitivity of tumors to gefitinib and erlotinib. Proc Natl Acad Sci U S A 101: 13306-11, 2004 8. Ono M, and Kuwano M: Molecular mechanisms of epidermal growth factor receptor

(EGFR) activation and response to gefitinib and other EGFR-targeting drugs. CUn Cancer Res 12: 7242-51, 2006 9. Shigematsu H, Lin L, Takahashi T, et al: Clinical and biological features associated with epidermal growth factor receptor gene mutations in lung cancers. J Natl Cancer Inst 97:339-46, 2005 10. Kohno K, Izumi H, Uchiumi T, et al: The pleiotropic functions of the

Y-box-binding protein, YB-I. Bioessays 25: 691-8, 2003 ll.Kuwano M, Oda Y, Izumi H, et al: The role of nuclear Y-box binding protein 1 as a global marker in drug resistance. MoI Cancer Ther 3: 1485-92, 2004

12. Koike K, Uchiumi T, Ohga T, et al: Nuclear tlanslocation of the YB- 1 binding protein by ultraviolet irradiation FEBS letters 417: 390-394, 1997

13. Sutherland BW, Kucab J, Wu J, et al: Akt phosphorylates the Y-box binding protein 1 at Ser 102 located in the cold shock domain and affects the anchorage- independent growth of breast cancer cells. Oncogene 24: 4281-92, 2005

14. Basaki Y, Hosoi F, Oda Y, et al: Akt-dependent nuclear localization of Y-box binding protein 1 in acquisition of malignant characteristics by human ovarian cancer cells. Oncogene 26: 2736-46, 2007

15. Wu J, Lee C, Yokom D, et al: Disruption of the Y-box binding protein- 1 results in suppression of the epidermal growth factor receptor and HER-2. Cancer Res 66: 4872-9, 2006 16. Fujii T, Kawahara A, Basaki Y, et al: Expression of HER2 and estrogen receptor alpha depends upon nuclear localization of Y-box binding protein- 1 in human breast cancers. Cancer Res 68: 1504-12, 2008

17. Janz M, Harbeck N, Dettmar P, et al: Y-box factor YB- 1 predicts drug resistance and patient outcome in breast cancer independent of clinically relevant tumor biologic factors HER2, uP A and PAI- 1. Int J Cancer 97: 278-82, 2002

18. Kamura T, Yahata H, Amada S,et al: Is nuclear expression of Y-box binding protein- 1 a new prognostic factor in ovarian serous adenocarcinoma? Cancer 85: 2450-4, 1999

19. Oda Y, Ohishi Y, Basaki Y, et al: Prognostic implication of the nuclear localization of the Y-box binding protein- 1 and CXCR4 expression in ovarian cancer: their correlation with activated Akt, LRP/MVP and P-glycoprotein expression. Cancer Sci 98: 1020-6, 2007

20. Shibahara K, Sugio K, Osaki T, et al: Nuclear expression of the Y-box binding protein as a novel marker of disease progression in non-small cell lung cancer. CUn

Cancer Res 7: 3151-5, 2001

21. Faury D, Nantel A, Dunn SE, et al: Molecular profiling identifies prognostic subgroups of pediatric glioblastoma and shows increased YB-I expression in tumors. J Clin Oncol 25: 1196-208, 2007 22. Therasse P, Arbuck SG, Eisenhauer EA, Wanders J, Kaplan RS, Rubinstein L, et al: New guidelines to evaluate the response to treatment in solid tumors. J Natl Cancer Inst 92: 205-16, 2000

23. Nagai Y, Miyazawa H. Huqun et al: Genetic heterogeneity of the epidermal growth factor receptor in non-small cell lung cancer cell lines revealed by a rapid and sensitive detection system, the peptide nucleic acid-locked nucleic acid PCR clamp.

Cancer Res., 65: 7276 - 82, 2005

24. Subramanian J, Govindan R: Lung cancer in never smokers: A Review. J CUn Oncol 25: 561-570, 2007

25. Toyooka S, Takano T, Kosaka T, et al: Epidermal growth factor receptor mutation, but not sex and smoking, is independently associated with favorable prognosis of gefitinib-treated patients with lung adenocarcinoma. Cancer Sci 99: 303-308, 2008

26. BeU DW, Brannigan BW, Matsuo K, et al.: Increased prevalence of EGFR-mutant lung cancer in woman and east asian populations: analysis of estrogen-related polymorphisms. CUn Cancer Res 14: 4079-4084, 2008 27. Izumi H, Imamura T, Nagatani G, et al: Y box-binding protein- 1 binds preferentiaUy tonsingle-stranded nucleic acids and exhibits 3'-5' exonuclease activity. Nucleuic Acids Res. 29: 1200-1207, 2001

28. Ise T, Nagatani G, Imamura T, et al: Transcription factor Y-box binding protein 1 binds preferentially to cisplatin-modified DNA and interacts with proUferating ceU nuclear antigen. Cancer Res 59: 342-6, 1999

29. Hayakawa H, Uchiumi T, Fukuda T, et al : Binding capacity of human YB-I protein for RNA containing 8-oxoguanine. Biochemistry 41: 12739-44, 2002

30. Imamura T, Izumi H, Nagatani G, et al: Interaction with p53 enhances binding of cisplatin-modified DNA by high mobility group 1 protein. J Biol Chem 276: 7534-40,

2001

31. Okamoto T, Izumi H, Imamura T, et al: Direct interaction of p53 with the Y-box binding protein, YB-I : a mechanism for regulation of human gene expression. Oncogene 19: 6194-02, 2000 32. Lu ZH, Books JT, Ley TJ: YB- 1 is important for late-stage embryonic development, optimal cellular stress responses, and the prevention of premature senescence. MoI CeU Biol 25: 4625-37, 2005

33. Uchiumi T, Fotovati A, Sasaguri T: YB-I is important for an early stage embryonic development: Neural tube formation and cell proliferation. J Biol Chem 281 : 40440-9, 2006.

34. Mok TS, Wu YL, Thongprasert S, et al: Gefitinib or carboplatin-paclitaxel in pulmonary adenocarcinoma. New Engl J Med 361: 947-57, 2009.

35. Evans MD, Dizdaroglu M, Cooke MS: Oxidative DNA damage and disease: induction, repair and significance. Mutat Res: 567: 1-61, 2004. 36. Tsuzuki T, Nakatsu Y, Nakabeppu Y: Significance of error-avoiding mechanisms for oxidative DNA damage in carcinogenesis. Cancer Sci 98: 465-70, 2007. 37. Oka S, Ohno M, Tsuchimoto D, Sakumi K, Furuichi M, Nakabeppu Y: Two distinct pathways of cell death triggered by oxidative damage to nuclear and mitochondrial DNAs. EMBO J 27: 421-32, 2008. 38. Sakumi K, Tominaga Y, Furuichi M, et al: Oggl knockout-associated lung tumorigenesis and its suppression by Mthl gene disruption. Cancer Res 63: 902-5, 2003. 39. Russo MT, De Luca G, Degan P, et al: Accumulation of the oxidative base lesion 8-hydroxyguanine in DNA of tumor-prone mice defective in both the Myh and Oggl DNA glycosylases. Cancer Res 64: 4411-4, 2004. 40. Chevillard S, Radicella JP, Levalois C et al: Mutations in OGGl, a gene involved in the repair of oxidative DNA damage, are found in human lung and kidney tumours. Oncogene 16: 3083-6, 1998.

41. Hung RJ, Hall J, Brennan P, Boffetta P: Genetic polymorphisms in the base excision repair pathway and cancer risk: a HuGE review. Am J Epidemiol 162: 925-42, 2005.

42. Chiou CC, Chang PY, Chan EC, Wu TL, Tsao KC, Wu JT: Urinary 8-hydroxydeoxyguanosine and its analogs as DNA marker of oxidative stress: development of an ELISA and measurement in both bladder and prostate cancers. CUn Chim Acta 334: 87-94, 2003.

43. Tudek B: Base excision repair modulation as a risk factor for human cancers. MoI Aspects Med 28: 258-75, 2007.

44. Pilger A, Rϋdiger HW: 8-Hydroxy-2'-deoxyguanosine as a marker of oxidative DNA damage related to occupational and environmental exposures. Int Arch Occup Environ Health 80: 1-15, 2006.

45. De Vizcaya-Ruiz A, Barbier O, Ruiz-Ramos R, Cebrian ME: Biomarkers of oxidative stress and damage in human populations exposed to arsenic. Mutat Res 674: 85-92, 2009.

46. Matsumoto K, Wolffe AP: Gene regulation by Y-box proteins: coupling control of transcription and translation. Trends Cell Biol 8: 318-23, 1998.

47. Bergmann S, Royer-Pokara B, Fietze E, et al: YB-I provokes breast cancer through the induction of chromosomal instability that emerges from mitotic failure and centrosome amplification. Cancer Res 65: 4078-87, 2005.

48. de Souza-Pinto N, Mason P, Hashiguchi K, et al: Novel DNA mismatch-repair activity involving YB-I in human mitochondria. DNA Repair 8: 704-19, 2009.

49. Kashihara M, Azuma K, Kawahara A, et al: Nuclear Y-box binding protein- 1 YB-L, a predictive marker of prognosis, is correlated with expression of HER2/ErbB2 and HER3/ErbB3 in non-small cell lung cancer. J Thorac Oncol 4: 1066-74, 2009.

50. Gackowski D, Speina E, Zielinska M, et al: Products of oxidative DNA damage and repair as possible biomarkers of susceptibility to lung cancer. Cancer Res 63: 4899-902, 2003.

51. Paz-Elizur T, Krupsky M, Blumenstein S, Elinger D, Schechtman E, Livneh Z: DNA repair activity for oxidative damage and risk of lung cancer. Journal of National Cancer Institute 95 : 1312-9, 2003.

52. Speina E, Zielinska M, Barbin A, et al: Decreased repair activities of 1, N6-ethenoadenine and 3, N4-ethenocytosine in lung adenocarcinoma patients. Cancer Res 63: 4351-7, 2003.

53. Gaudreaut I, Guay D, Lebel M: YB-I promotes separation in vitro of duplex DNA containing either mispaired bases or cisplatin modifications, exhibits endonucleolytic activities and binds several DNA repair proteins. Nucleic Acids Res 32: 316-27, 2004.

54. Das S, Chattopadhyay R, Bhakat KK, et al: Stimulation of NEIL2-mediated oxidized base excision repair via YB-I interaction during oxidative stress. J Biol Chem 282: 28474-84, 2007.