WO2009023172A2 - Predictions of responsiveness to egfr inhibitors - Google Patents

Abstract

Coordinated over-expression of the EGFR pathway predicts susceptibility to EGFR inhibitors in cancer. This suggests that a phenomenon of pathway- rather than oncogene-addiction predicts vulnerability to EGFR inhibition in cancer. This demonstrates the powerful value of unbiased systems biology approaches in drug development and have important implications for the management of use of these agents in cancers.

Description

PREDICTIONS OF RESPONSIVENESS

TO EGFR INHIBITORS

[01] This application claims the benefit of provisional application U.S. Ser. No. 60/964,099 filed August 9, 2007, the contents of which are expressly incorporated herein.

TECHNICAL FIELD OF THE INVENTION

[02] This invention is related to the area of cancer therapy and management. In particular, it relates to providing therapies that are tailored to a patient's own tumor.

BACKGROUND OF THE INVENTION

[03] The epidermal growth factor receptor (EGFR) inhibitor erlotinib is approved for treatment of pancreatic cancer but the overall therapeutic efficacy is minimal (1). There is an unmet need to identify the individual factors predicting such susceptibility. Cancer is a genetic disease (2), and accumulating data suggest that the factors determining the sensitivity to anticancer agents have also a genetic basis. The presence of acquired mutations in the catalytic domain of the EGFR gene increase sensitivity to anti-EGFR small-molecule inhibitors in non-small cell lung cancer (3, 4). Likewise, increased EGFR and HER2 gene copy number detected by FISH was associated with improved gefitinib efficacy in patients with NSCLC (5, 6). In addition, there is evidence that KRAS mutations confer resistance to EGFR inhibition (7).

[04] Because the EGFR is a validated target in pancreatic cancer but with limited clinical activity, the identification of factors predicting drug response is a relevant question. However several studies investigating known predictive factors for EGFR inhibition such as EGFR mutations or amplifications in pancreatic cancer have failed to document a meaningful prevalence of such alterations (8, 9), and HER2 amplification assessment has provided conflicting results (10, 11). Inversely, the almost universal presence of KRAS mutations in pancreatic cancer argues against a collective relevance of this trait in determining anti-EGFR therapy sensitivity.

[05] Because of existing data on other tumor types, the hypothesis driving our work was that vulnerability to EGFR-targeting agents is related to dependence on the EGFR pathway. However the array of available negative data led us to propose that factors other than those single oncogene alterations may be relevant in terms of determining anti-EGFR effect. In addition, the level of complexity of common cancers may be higher than expected (12), and probably more sophisticated, integrative approaches to gathering information will be needed in order to meaningfully interrogate a tumor. Gene expression analysis has shown promise to characterize cancer (13), and recently a platform derived from global unbiased testing received regulatory approval for risk prognostication for breast cancer (14). From a biologic perspective it allows for an unbiased evaluation of what is considered the dynamic language controlling cell processes, both normal and altered. In addition this information is not circumscribed to a single target, but allows for integrative approaches, now known as systems biology. There is a continuing need in the art to identify factors that can be used to predict responsiveness to therapeutic agents.

SUMMARY OF THE INVENTION

[06] According to one aspect of the invention a method is provided for predicting sensitivity or resistance of a tumor to an EGFR inhibitory drug or biological. Expression of at least 2 genes of the EGFR signaling pathway in a tumor sample of a subject is compared to expression of control samples of a plurality of human rumors resistant to an EGFR inhibitory drug, wherein the genes are selected from the group consisting of ABL2, AKT2, AKT3, BTC, CAMK2A, EGF, EREG, FRAPl, HBEGF, HRAS, MAPKlO, MAPK8, MAPK9, NRAS, NRG2, NRG4, PAK6, PIK3CA, PIK3R3, PRKCBl, SHC3, SHC4, SOSl, SRC, TGFA, ABLl, AKTl, ARAF, AREG, BAD, BRAF, CAMK2B, CAMK2D, CAMK2G, CBL, CBLB, CBLC, CDKNlA, CDKNlB, CRK, CRKL, EGFR, EIF4EBP1, ELKl, ERBB2, ERBB3, ERBB4, GABl, GRB2, GSK3B, JUN, KRAS, MAP2K1, MAP2K2, MAP2K4, MAP2K7, MAPKl, MAPK3, MYC, NCKl, NCK2, NRGl, NRG3, PAKl, PAK2, PAK3, PAK4, PAK7, PIK3CB, PIK3CD, PIK3CG, PIK3R1, PIK3R2, PIK3R5, PLCGl, PLCG2, PRKCA, PRKCG, PTK2, RAFl, RPS6KB1, RPS6KB2, SHCl, SHC2, SOS2, STAT5A, and STAT5B. If the average of the differences of expression is higher in the tumor sample than in the control samples, one predicts that the tumor of the subject will be sensitive to EGFR inhibitory drugs. Alternatively, if the average expression is lower in the tumor sample than in the control samples, one predicts that the rumor of the subject will be resistant to EGFR inhibitory drugs.

[07] According to another aspect of the invention a method is provided for predicting resistance of a tumor to an EGFR inhibitory drug. Expression of a gene in a tumor sample of a subject is compared to expression of a plurality of control samples of tumors sensitive to an EGFR inhibitory drug. The gene is selected from the group consisting of ATP2A3, DDC, HLXB9, KIAA0282, MAOA, MAPK7, PIP5K1B, PLCD3, PRSS3, TAC3, THRA, THRB and VIPRl . If the expression is higher in the tumor sample than in the control sample, then one predicts that the tumor of the subject is resistant to EGFR inhibitory drugs.

[08] Still another aspect of the invention is a kit for treating a tumor resistant to EGFR inhibitory drugs. The kit comprises an EGFR inhibitory drug or biological and an inhibitor of ATP2A3, DDC, HLXB9, KIAA0282, MAOA, MAPK7, PIP5K1B, PLCD3, PRSS3, TAC3, THRA, THRB or VIPRl . The EGFR inhibitory drug or biological and the inhibitor of ATP2A3, DDC, HLXB9, KIAA0282, MAOA, MAPK7, PIP5K1B, PLCD3, PRS S3, TAC3, THRA, THRB or VIPRl are in a single or separate containers within the kit container.

[09] Also provided as an aspect of the invention is a diagnostic reagent for assessing susceptibility or resistance to EGFR inhibitory drugs. A solid support comprises oligonucleotide probes which are complementary to at least 2 genes selected from the group consisting of ABL2, AKT2, AKT3, BTC, CAMK2A, EGF, EREG, FRAPl, HBEGF, HRAS, MAPKlO, MAPK8, MAPK9, NRAS, NRG2, NRG4, PAK6, PIK3CA, PIK3R3, PRKCBl, SHC3, SHC4, SOSl, SRC, TGFA, ABLl, AKTl, ARAF, AREG, BAD, BRAF, CAMK2B, CAMK2D, CAMK2G, CBL, CBLB, CBLC, CDKNlA, CDKNlB, CRK, CRKL, EGFR, EIF4EBP1, ELKl, ERBB2, ERBB3, ERBB4, GABl, GRB2, GSK3B, JUN, KRAS, MAP2K1, MAP2K2, MAP2K4, MAP2K7, MAPKl , MAPK3, MYC, NCKl, NCK2, NRGl, NRG3, PAKl, PAK2, PAK3, PAK4, PAK7, PIK3CB, PIK3CD, PIK3CG, PIK3R1, PIK3R2, PIK3R5, PLCGl, PLCG2, PRKCA, PRKCG, PTK2, RAFl, RPS6KB1 , RPS6KB2, SHCl, SHC2, SOS2, STAT5A, STAT5B; ATP2A3, DDC, HLXB9, KIAA0282, MAOA, MAPK7, PIP5K1B, PLCD3, PRSS3, TAC3, THRA, THRB and VIPRl . The at least 2 genes selected from the group comprise at least 50 % of the genes for which the solid support contains oligonucleotide probes.

[10] Yet another aspect of the invention is a method of providing a prediction of sensitivity or resistance of a tumor to an EGFR inhibitory drug or biological. Expression data for at least 2 genes of the EGFR signaling pathway in a tumor sample of a subject is obtained. The genes are selected from the group consisting of ABL2, AKT2, AKT3, BTC, CAMK2A, EGF, EREG, FRAPl, HBEGF, HRAS, MAPKlO, MAPK8, MAPK9, NRAS, NRG2, NRG4, PAK6, PIK3CA, PIK3R3, PRKCBl, SHC3, SHC4, SOSl, SRC, TGFA, ABLl, AKTl, ARAF, AREG, BAD, BRAF, CAMK2B, CAMK2D, CAMK2G, CBL, CBLB, CBLC, CDKNlA, CDKNlB, CRK, CRKL, EGFR, EIF4EBP1, ELKl, ERBB2, ERBB3, ERBB4, GABl, GRB2, GSK3B, JUN, KRAS, MAP2K1, MAP2K2, MAP2K4, MAP2K7, MAPKl, MAPK3, MYC, NCKl, NCK2, NRGl, NRG3, PAKl, PAK2, PAK3, PAK4, PAK7, PIK3CB, PIK3CD, PIK3CG, PIK3R1, PIK3R2, PIK3R5, PLCGl, PLCG2, PRKCA, PRKCG, PTK2, RAFl, RPS6KB1, RPS6KB2, SHCl, SHC2, S0S2, STAT5A, and STAT5B. The data are analyzed by comparing expression of the at least 2 genes in the tumor sample to expression of control samples of a plurality of human tumors resistant to an EGFR inhibitory drug. The EGFR signaling pathway expression is deemed to be increased if the average differences in expression of the at least 2 genes is increased. A prediction that the tumor of the subject will be sensitive to EGFR inhibitory drugs is transmitted to the subject or the subject's physician or a diagnostic laboratory which generated the expression data if the average differences in expression is higher in the tumor sample than in the control samples. A prediction that the tumor of the subject will be resistant to EGFR inhibitory drugs is transmitted if the average differences in expression is lower in the tumor sample than in the control samples.

[11] Another aspect of the invention is a computer that accesses a database of expression data of EGFR pathway genes in EGFR inhibitor-resistant tumor samples and compares the database values to test sample values and provides a prediction of EGFR inhibitor- sensitivity or EGFR inhibitor-resistance of the test sample based on average of difference of expression values of the genes compared to the database values. The expression data of EGFR pathway genes which are compared and differences averaged comprise at least 25 of ABL2, AKT2, AKT3, BTC, CAMK2A, EGF, EREG, FRAPl, HBEGF, HRAS, MAPKlO, MAPK8, MAPK9, NRAS, NRG2, NRG4, PAK6, PIK3CA, PIK3R3, PRKCBl, SHC3, SHC4, SOSl, SRC, TGFA, ABLl, AKTl, ARAF, AREG, BAD, BRAF, CAMK2B, CAMK2D, CAMK2G, CBL, CBLB, CBLC, CDKNlA, CDKNlB, CRK, CRKL, EGFR, EIF4EBP1, ELKl, ERBB2, ERBB3, ERBB4, GABl, GRB2, GSK3B, JUN, KRAS, MAP2K1, MAP2K2, MAP2K4, MAP2K7, MAPKl, MAPK3, MYC, NCKl, NCK2, NRGl, NRG3, PAKl, PAK2, PAK3, PAK4, PAK7, PIK3CB, PIK3CD, PIK3CG, PIK3R1, PIK3R2, PIK3R5, PLCGl, PLCG2, PRKCA, PRKCG, PTK2, RAFl, RPS6KB1, RPS6KB2, SHCl, SHC2, S0S2, STAT5A, and STAT5B.

[12] Also provided as an aspect of the invention is a method for treating a patient with a tumor predicted to be resistant to EGFR inhibitory drugs. An EGFR inhibitory drug or biological is administered to the patient. An inhibitor of ATP2A3, DDC, HLXB9, KIAA0282, MAOA, MAPK7, PIP5K1B, PLCD3, PRSS3, TAC3, THRA, THRB or VIPRl is also administered to the patient. The two types of inhibitors can be coadministered or separately administered within a hours, days, or weeks. [13] These and other embodiments which will be apparent to those of skill in the art upon reading the specification provide the art with tools, methods, and treatments for increasing the efficacy of anti-cancer drugs.

BRIEF DESCRIPTION OF THE DRAWINGS

[14] Fig. 1 shows efficacy of erlotinib (a small molecule inhibitor of the tyrosine kinase activity of EGFR), cetuximab (a monoclonal antibody targeting the extracelular domain) and the combination of both in pancreatic cancer xenografts. A. Bar graph of all ten cases. 198 and 410 were highly sensitive to either treatment modality. Overall erlotinib showed marginally higher potency compared with cetuximab, with an average T/C of 54% versus 65% when the indexes of all 10 cases were pooled together. The combined therapy had an average 45% T/C. Within each case the growth is normalized to the growth of the control. Cases with a T/C of less than 20% were considered sensitive. Bars represent standard deviation (SD). B. Graph of selected cases. 198 and 410 were the tumors that showed the highest sensitivity to EGFR inhibition. 215 was resistant to all three modalities.

[15] Fig. 2A-2D. Fig 2A shows a list of the top eight gene sets enriched in the erlotinib sensitive cases with nominal p-value < 0.01. The gene list is sorted descending with NES score. The EGFR signaling pathway is stippled. According to the KEGG database annotation, the EGFR signaling pathway consists of 87 genes, and 25 of these genes that contribute most to the enrichment result were defined as the core enrichment genes. (Size, number of genes in the gene set; ES, enrichment score; NES, normalized enrichment score; NOM p-val, nominal p-value; FDR q-val, false discovery rate). Fig. 2B shows enrichment plot for the EGFR signaling pathway. The top section of the plot shows the running ES for the gene set as the analysis walks down the ranked list. The score at the peak of the plot is the ES for the gene set. The middle section of the plot shows where the members of the gene set appear in the ranked list of genes. The bottom section of the plot shows the value of the ranking metric along the list of ranked genes. Fig. 2C shows a heatmap of the core enrichment genes. Each row corresponds to a gene and each column corresponds to a sample array. The expression level for each gene is normalized across the samples such that the mean is 0 and the standard deviation is 1. Genes with expression levels greater than the mean are denoted with the smallest stippling and those below the mean are denoted with vertical stripes. The other symbols are intermediate between the most highly expressed and the least highly expressed relative to the mean. Fig. 2D. shows EGFR signaling pathway. The core enrichment genes are stippled according to the KEGG map annotation.

[16] Fig. 3A-3B. Fig. 3A. EGFR and HER2 FISH of selected cases. Fig. 3B. Multiplex ligation-dependent probe amplification (MLPA) analysis. Both sensitive tumors had a very similar distribution of dosage by MLPA with the exception of HER2, and were the cases with the highest similitude. Half or more of the cases had gains of EGFR, PIK3CA, and Aktl. There was a poor correlation between EGFR FISH and MLPA. Cases with low number of HER2 copies by FISH tended to have either no change (286, 194) or a loss (198) by MLPA, but there was not a good correlation between HER2 by both techniques in cases scoring 4 or more by FISH. There was not a pattern between KRAS mutations and MLPA NRAS profile. Diagonal shading and smallest stippling indicate gain and loss, respectively, of the chromosomal area where the gene resides; heavy stippling denotes no change.

[17] Fig. 4 (Table 1.) Mutation patterns identified for the EGFR, KRAS, and PBKCA genes, and gene amplifications in EGFR and HER2 in pancreatic cancer tumors. No correlation was found between KRAS mutation profile and EGFR and/or HER2 gene amplification profiles. There was no correlation between mutational status and EGFR or HER2 amplification status and sensitivity. Of the three sensitive cases (in bold), 198 and 219 had normal EGFR copy number and 410 had high polisomy. Both low and high EGFR copy numbers were documented in the two cases with a more resistant pattern, 265 and 215 respectively. [18] Fig. 5. (Table 2.) IHC baseline patterns identified for the panel of 10 cases where efficacy data was obtained. Sensitive cases are shown in bold.

DETAILED DESCRIPTION OF THE INVENTION

[19] The inventors have discovered predictive associations between global pathway expression and drug responsiveness of tumors. Predictive associations have also been found between subsets of the pathway expression and drug responsiveness of tumors. Predictive associations have also been discovered between particular gene expression and drug non-responsiveness of tumors. These predictors can be used to guide treatment decisions, both the decision whether to treat and the decision with what agents to treat.

[20] The therapeutic agents to which the invention relates are EGFR inhibitory drugs. These drugs include small molecule drugs and biologicals, such as antibodies and antibody derivatives and peptides. The class of EGFR inhibitory drugs includes, without limitation erlotinib, cetuximab, and gefitinib.

[21] The EGFR pathway according to the KEGG database has 87 genes. These are ABL2, AKT2, AKT3, BTC, CAMK2A, EGF, EREG, FRAPl, HBEGF, HRAS, MAPKlO, MAPK8, MAPK9, NRAS, NRG2, NRG4, PAK6, PIK3CA, PIK3R3, PRKCBl, SHC3, SHC4, SOSl, SRC, TGFA, ABLl, AKTl, ARAF, AREG, BAD, BRAF, CAMK2B, CAMK2D, CAMK2G, CBL, CBLB, CBLC, CDKNlA, CDKNlB, CRK, CRKL, EGFR, EIF4EBP1, ELKl, ERBB2, ERBB3, ERBB4, GABl, GRB2, GSK3B, JUN, KRAS, MAP2K1, MAP2K2, MAP2K4, MAP2K7, MAPKl, MAPK3, MYC, NCKl, NCK2, NRGl, NRG3, PAKl, PAK2, PAK3, PAK4, PAK7, PIK3CB, PIK3CD, PIK3CG, PIK3R1, PIK3R2, PIK3R5, PLCGl, PLCG2, PRKCA, PRKCG, PTK2, RAFl, RPS6KB1, RPS6KB2, SHCl, SHC2, S0S2, STAT5A, and STAT5B. Other database definitions of the EGFR pathway may also be useful, such as Biocarta, GenMapp, NCI- Nature PID, Science STKE, Reactome, aMAZE, UCSD-Nature Signaling Gateway, Cancer CeIlMAP, Cell Snapshots collection, PharmaGKB, and Ingentuity Pathway DB. A particularly useful subset of these genes appear to be important to the determination of increased expression. This subset of "core" genes includes ABL2, AKT2, AKT3, BTC, CAMK2A, EGF, EREG, FRAPl, HBEGF, HRAS, MAPKlO, MAPK8, MAPK9, NRAS, NRG2, NRG4, PAK6, PIK3CA, PIK3R3, PRKCBl, SHC3, SHC4, SOSl, SRC, and TGFA. Individual genes which seem particularly relevant to determining resistance include ATP2A3, DDC, HLXB9, KIAA0282, MAOA, MAPK7, PIP5K1B, PLCD3, PRSS3, TAC3, THRA, THRB and VIPRl .

[22] The gene names are standard in the art. The full names of the 25 core genes are as follows:

ABL2 v-abl Abelson murine leukemia viral oncogene homolog 2 (arg,

Abelson-related gene) AKT2 v-akt murine thymoma viral oncogene homolog 2 AKT3 v-akt murine thymoma viral oncogene homolog 3 (protein kinase B, gamma)

BTC betacellulin

CAMK2A calcium/calmodulin-dependent protein kinase (CaM kinase) Il alpha EGF epidermal growth factor (beta-urogastrone)

EREG epiregulin

FRAP1 FK506 binding protein 12-rapamycin associated protein 1

HBEGF heparin-binding EGF-like growth factor

HRAS v-Ha-ras Harvey rat sarcoma viral oncogene homolog

MAPK8 mitogen-activated protein kinase 8

MAPK9 mitogen-activated protein kinase 9

MAPK10 mitogen-activated protein kinase

10

NRAS neuroblastoma RAS viral (v-ras) oncogene homolog

NRG2 neuregulin

2 NRG4 neuregulin

4

PAK6 p21 (CDKN1 A)-activated kinase 6

PIK3CA phosphoinositide-3-kinase, catalytic, alpha polypeptide PIK3R3 phosphoinositide-3-kinase, regulatory subunit 3 (p55, gamma)

PRKCB1 protein kinase C, beta

1

SHC3 SHC (Src homology 2 domain containing) transforming protein 3

SHC4 SHC (Src homology 2 domain containing) family, member 4

SOS1 son of sevenless homolog 1 (Drosophila)

SRC v-src sarcoma (Schmidt-Ruppin A-2) viral oncogene homolog (avian)

TGFA transforming growth factor, alpha [23] Although the global or cumulative expression of the EGFR pathway has been found to be a powerful and accurate predictor of EGFR inhibitory drug sensitivity, fewer genes than the whole pathway may also be useful. Thus at least 2, 3, 4, 5, 6, 7, 8, 9, 10, 12, 14, 16, 18, 20, 25, 30, 35, 40, 45, 50, 55, 60, 65, 70, 75, 80, 85, or 87 genes may be tested for expression and the differences in expression values between the test sample and pools of resistant tumors averaged. We have used the mean expression value to determine whether the EGFR pathway is increased in expression. Other statistical treatments may also provide meaningful results, including the mode or median. Because of the small levels of increased expression which are observed and because of the great tumor-to- tumor variations seen even among tumors which are classified similarly by cytological features, using the entire pathway or a large and significant portion of it may provide benefits of increased predictiveness. When average expression differences are higher than the expression "difference" in the resistant pool, the average is a positive number. When the average expression differences are lower than in the resistant pool, the average is a negative number.

[24] The predictive methods taught here can be used for any tumors for which EGFR inhibitory drugs are used or are being considered. These include, lung cancer, breast cancer, pancreatic cancer, colon cancer, prostate cancer, brain cancers, head and neck cancers (including squamous cell cancers), kidney cancer, gastric cancer, cervical cancer, ovarian cancer, liver cancer, bladder cancer, cancer of the urinary tract, thyroid cancer, melanoma, and others.

[25] Comparisons of patient samples to populations of either resistant tumors or sensitive tumors (non-responders and responders) can be done using a computer or manually. The population or pooled expression data may be obtained from xenografts or from clinical tumor samples. The populations may be homogeneous or heterogeneous with respect to type of cancer, type of EGFR inhibitory drug to which they are resistant or sensitive, and type of sample which is tested, xenograft or clinical tumor sample. Populations which are homogeneous for type of drug and type of cancer may provide better predictiveness. The expression data which are collected for the populations can be stored in a database which is accessed by the computer to perform the comparisons to the test subject expression.

[26] To determine sensitivity to an EGFR inhibitory drug or biological, an individual gene's expression is compared to expression data for that gene from a pool of resistant tumors. The fold-increase or fold-decrease can be determined, i.e., a difference is determined between the test sample and the pool. Each of these fold increases or decreases (i.e., ratios or differences) can be summed over a desired set of genes to provide an indication of cumulative pathway activation. As discussed above, an entire pathway's gene expression can be summed (or averaged) or core genes' expression can be summed (or averaged) or significant subsets' expression can be summed (or averaged). Changes are considered significant if the p-value is <0.05.

[27] A prediction of sensitivity can be recorded on a hard drive or on other magnetic storage device, or on a fixed medium such as paper, stone, or parchment. The prediction may be stored in a patient's chart (medical record). The prediction may be communicated to a testing lab, to a treating physician, or directly to a patient. The communication means may be any, including but not limited to telephone, facsimile, cable, world-wide web, post, human-to-human speech. Typically recommendations for drug treatments are made by a treating physician, but recommendations can also be made by the computer or personnel who determine sensitivity or resistance. Recommendations may or may not result in a prescription or actual treatment.

[28] If a tumor is determined to be resistant to EGFR inhibitory drugs, then a different class of drug or biological can be recommended. Further, if a particular tumor is determined not to be sensitive to EGFR inhibitory drugs, and if particular resistance genes are noted as overexpressed relative to EGFR inhibitory drug-sensitive tumor populations, then a combination treatment may be recommended. If a resistance gene is overexpressed that is the target of inhibitory drugs, then combinations of resistant gene-inhibitory drug and EGFR inhibitory drug or biologicals can be administered to a patient in conjunction. Inhibiting the former will make the latter more efficacious. Combination therapeutic regimens may be accomplished by mixtures of drugs, i.e., cocktails of two or more drugs, or separate administrations at the same time, or separate administrations that are close in time, for example within hours, days, or weeks. Possibly, cocktails and combined treatments with more than two inhibitors are possible, depending on the expression profile. Increased expression of any of the resistance genes can be antagonized using an antibody which specifically binds to the protein encoded by the resistance gene. Antibodies may be monoclonal, single chain, humanized, chimeric, or other derivative antibody type. siRNAs directed to the resistance genes which are overexpressed may optionally be used.

[29] Kits comprising combinations of drugs can be supplied, either as cocktails (mixtures) or as separate compositions. If each drug of a combination is in a separate vessel or container, then the separate vessels or containers are themselves in a single package. The kit can further comprise additional items, including delivery devices, such as syringes, capsules, etc., and literature such as administration instructions and safety warnings.

[30] Any means for assessing increased expression of mRNA for particular genes can be used. Any commercial microarray type platform can be used. These are typically geographically addressed oligonucleotide probes. Because analysis of EGFR pathway and sensitivity and resistance genes can be performed in a manner that ranks gene expression of one gene against another, the precise platform that is used is not critical. In deed, any method for assessing quantitatively mRNA expression can be used, including but not limited to Serial Analysis of Gene Expression (SAGE; Velculescu et al., Science 270: 484-487 (1995); and Velculescu et al., Cell 88 : 243-51 (1997)), gene expression analysis by Massively Parallel Signature Sequencing (MPSS; Brenner et al., Nature Biotechnology 18: 630-634 (2000)), Agilent microarrays, and Affymetrix microarrays. Special purpose microarrays can also be used which measure EGFR pathway and/or EGFR inhibitory drug resistance genes. These special purpose microarrays contain less than 50 %, less than 40 %, less than 30 %, less than 20 %, or less than 10 % oligonucleotide probes that do not relate to the EGFR pathway or EGFR inhibitory drug resistance genes. Any solid support material and means of depositing oligonucleotide probes on the solid support material can be used. Expression of genes in tumors can be measured in xenografts or clinical samples, whether fresh, paraffin-embedded tumor tissue, or otherwise treated or preserved. mRNA can be isolated from tumors using standard methods known in the art.

[31] Alternatively, means of quantitatively assessing protein expression of the same genes may be used. Any may be used, including but not limited to antibody microarrays, ELISA, Western blot, etc.

[32] The term "oligonucleotide" refers to a relatively short polynucleotide, including, without limitation, single-stranded deoxyribonucleotides oligomers, single- or double-stranded ribonucleotides oligomers, RNA: DNA hybrids and double-stranded deoxyribonucleotides oliogmers. Oligonucleotides, such as single-stranded DNA probe oligonucleotides, are often synthesized by chemical methods, for example using automated oligonucleotide synthesizers that are commercially available. However, oligonucleotides can be made by other methods, including in vitro recombinant DNA- mediated techniques and by expression or production in cells and organisms.

[33] The term "over-expression" or "increased expression" with regard to an RNA transcript of a gene is used to refer the level of the transcript determined by comparison to the level of reference mRNAs for the same gene in a pool of tumors that are sensitive or resistant to EGFR inhibitory drugs or biologicals. Average differences among genes between a test sample and a pool of control sample values can be determined. Larger increases Tn expression of a pathway will lead to larger positive values of the average differences. Larger decreases in expression of a pathway will lead to larger negative values of the average differences. Smaller average differences suggest that the pathway is not globally or coordinately up or down regulated.

[34] The term "prediction" is used to refer to the determination of a likelihood that a patient's tumor will respond to an EGFR inhibitor or the class of EGFR inhibitors. The predictive methods of the present invention can be used clinically to make treatment decisions by choosing the most appropriate treatment for a patient.

[35] The term "resistance" to a particular drug or class of drugs, means little or absence of response of a tumor to a standard dose of the drug or to a standard treatment protocol. The term "sensitivity" to a particular drug or treatment option, means response to a standard dose of the drug or to a class of drugs. Responses of tumors may be slowing of growth of the tumor, regression of the tumor, increased necrosis of the tumor, decreased risk of metastasis and invasiveness to adjacent tissues, increased time until cancer recurrence after resection or remission. Responses may also be observed in the whole body of the patient, such as decreased pain, cachexia, wasting, or other associated symptoms. Any standard measurements of cancer patient well-being can be used to assess responsiveness to an anti-tumor treatment.

[36] A computer system can be used for determining the similarity of the level of mRNA (or cDNA derived from the mRNA) in a sample to that in an EGFR inhibitor sensitive or resistant pool of tumors. The computer system may comprise a processor, and a memory encoding one or more programs coupled to the processor, wherein the one or more programs cause the processor to perform a method comprising computing the cumulative differences in expression of each marker between the sample and the pool. A computer readable medium may be used which has recorded on it one or more executable programs for determining the similarity of the level of nucleic acids expressed from individual genes of the EGFR signaling pathway in a sample to that in a pool of samples. One or more programs cause a computer to perform a method comprising computing the differences in expression of each gene between the sample and the pool and computing the cumulative differences in expression of the relevant group of genes between the sample and the pool. Computer programs may be used to store and access data, in particular pool data, e.g., in a database, and test patient data.

[37] Resistance genes which are expressed in resistant tumors more than in sensitive tumors include: ATP2A3, DDC, HLXB9, KIAA0282, MAOA, MAPK7, PIP5K1B, PLCD3, PRSS3, TAC3, THRA, THRB and VIPRl . In cases where a patient tumor has increased expression of MAOA, inhibitors can be administered in combination or in conjunction with inhibitors of EGFR. Inhibitors of MAOA which can be used in conjunction with an EGFR inhibitor in order to overcome a tumor's resistance include inhibitors which are specific for MAOA or ones that are less specific, including but not limited to isocarboxazid, moclobemide, phenelzine, tranylcypromine, rasagiline, nialamide, iproniazid, iproclozide, toloxatone, linezolid, selegiline, and dextroamphetamine. Inhibitors of ATP2A3 can be used in conjunction with an EGFR inhibitor as well. These are useful in cases where the patient's tumor has increased expression of ATP2A3. Such inhibitors include artimesin and antibodies to ATP2A3. If a patient's tumor has increased expression of DDC, a DDC inhibitor can be used in conjunction with an EGFR inhibitor. Such DDC inhibitors include, without limitation, benserazide and carbidopa. Increased expression of any of the resistance genes can be antagonized using an antibody which specifically binds to the protein encoded by the resistance gene. Antibodies may be monoclonal, single chain, humanized, chimeric, or other derivative antibody type. Use of such antibodies in conjunction with an EGFR inhibitor will increase the sensitivity of the tumor to the EGFR inhibitor. siRNAs and antisense RNAs which are directed to the resitance genes which are overexpressed can also be used.

[38] The above disclosure generally describes the present invention. All references disclosed herein are expressly incorporated by reference. A more complete understanding can be obtained by reference to the following specific examples which are provided herein for purposes of illustration only, and are not intended to limit the scope of the invention.

EXAMPLE 1

Methods Drugs

[39] Erlotinib (OSI Pharmaceuticals, Melville, NY) and cetuximab (Imclone Systems, New York, NY), were obtained from commercially available sources.

In vivo growth inhibition studies

[40] Six-week-old female athymic nude mice (Harlan, IN, US) were used. The research protocol was approved by the Johns Hopkins University Animal Care and Use Committee and animals were maintained in accordance to guidelines of the American Association of Laboratory Animal Care. The xenografts were generated according to methodology published elsewhere (15). Briefly, surgical non-diagnostic specimens of patients operated at the Johns Hopkins Hospital were reimplanted subcutaneously to 1-2 mice for each patient, with 2 small pieces per mouse (Fl generation). Tumors were let to grow to a size of 1.5 cm³ at which point were harvested, divided, and transplanted to another 5 mice (F2 generation). After a second growth passage tumors were excised and propagated to cohorts of 20 mice or more, that constituted the treatment cohort (F3 generation). Tumors from this treatment cohort were allowed to grow until reaching -200 mm³, at which time mice were randomized in the following three treatment groups, with 5-6 mice (10 evaluable tumors) in each group: 1) Control; 2) Erlotinib 50 mg/Kg/day ip; 3) Cetuximab 40 mg/Kg 2 times a week ip; and 4) Erlotinib plus cetuximab at the above doses. Treatment was given for 28 days. Gemcitabine and CIl 040 were administered in prior experiments twice weekly for 4 weeks at 100 mg/kg Ip and twice daily for 28 days at 150 mg/kg ip, respectively. Mice were monitored daily for signs of toxicity and were weighed three times per week. Tumor size was evaluated two times per week by caliper measurements using the following formula: tumor volume = [length X width²]/2. Relative tumor growth inhibition was calculated by relative tumor growth of treated mice divided by relative tumor growth of control mice since the initiation of therapy (T/C). Tumors with a T/C of less than 20% were considered sensitive. Microarray gene expression

[41] Baseline, untreated tumors were profiled using Affymetrix Ul 33 Plus 2.0 gene arrays in duplicate. Sample preparation and processing procedure was performed as described in the Affymetrix GeneChip® Expression Analysis Manual (Affymetrix Inc., Santa Clara, CA). Raw files are available as supplemental data.

Gene set enrichment analysis

[42] Gene expression levels were converted to a rank-based matrix and standardized (mean = 0, standard deviation = 1) for each microarray. Gene Set analysis was performed using the GSEA software (16) Version 2.0.1 obtained from the Broad Institute (available at its website). Genes represented by more than one probe were collapsed using the Collapse Probes utility to the probe with the maximum value. The gene sets database was compiled from the KEGG database (May 29, 2007 version) (17). The KEGG gene sets database contains 197 human pathways that include metabolism, genetic information processing, environmental information processing, cellular processes and human diseases. 165 gene sets passed the gene set size filter criteria (min = 10, max = 500). P-values for the gene sets were computed by permuting the genes 1000 times in this study.

Core gene expression classifier

[43] The core gene expression classifier was build by the logistic regression model using LogitBoost implemented in the WEKA machine learning package version 3.4 (18). The default parameters were used in this study.

DNA mutation analysis [44] Mutations of the KRAS oncogene were determined as previously described (19). PCR amplifications of exons 18, 19 and 21 of EGFR, exon 11 of BRAF, exons 9, 10, and 20 of PBKCA were performed as described (3, 20, 21). The primers used are available upon request. Sequencing in the forward and reverse direction was performed using an ABI 3730XL Sequencer in the Genetics Resource Core Facility, Johns Hopkins University School of Medicine.

Fluorescence in situ hybridization (FISH) assessment

[45] Paraffin-embedded sections were submitted to dual-color FISH assays using the EGFR SO/CEP7 SG probe set and the PathVysion DNA Kit (HER2 SO/CEP 17 SG; Vysis/ Abbott Laboratories, North Chicago, IL). Initially the slides were incubated for 2 hours at 60°C, deparafinized in Citro-Solv (Fisher, Liberty Lane Hampton, NH) and washed in 100% ethanol for 5 min. The slides were incubated in 2XSSC at 75°C for 10- 18 min and digested in 0.25mg/ml Proteinase K/2XSSC at 45°C for 11-18 min. Then, the slides were washed in 2XSSC for 5 min and dehydrated in ethanol. Probes were applied according to the manufacturer's instructions to the selected hybridization areas. DNA denaturation was performed for 15 min at 80⁰C and the slides were incubated at 37°C for 20 hours. Post-hybridization washes were performed with 1.5 Urea/0. IXSSC at 45°C for 35 min. Then, the slides were washed in 2XSSC for 2 min and dehydrated in ethanol. Chromatin was count erstained with DAPI (0.3 μg/ml in Vectashield; Vector Laboratories). Analysis was performed on epifluorescence microscope using single interference filters sets for green (FITC), red (Texas red) and blue (DAPI) as well as dual (red/green) and triple (blue, red, green) band pass filters and was done in the areas correspondent to the areas previously microdissected.

[46] According to the frequency of cells with specific number of copies of the EGFR gene and chromosome 7 centromere, the areas were classified into six FISH categories with ascending number of copies of the EGFR gene per cell: (1) Disomy (<2 copies in >90% of cells); (2) Low trisomy (<2 copies in >40% of cells, 3 copies in 10-40% of cells, >4 copies in <10% of cells); (3) High trisomy (<2 copies in ≥40% of cells, 3 copies in >40% of cells, >4 copies in <10 % of cells); (4) Low polysomy (> 4 copies in 10-40% of cells); (5) High polysomy (>4 copies in >40% of cells); and (6) Gene Amplification, defined by the presence of tight EGFR gene clusters, a ratio gene/chromosome per cell >2 or >15 copies of EGFR per cell in >10% of analyzed cells. FISH scores 1 to 4 classify the specimen as FISH negative (FISH-), scores 5 and 6 classify the specimen as FISH positive (FISH+).

Multiplex ligation-dependent probe amplification (MLPA)

[47] For MLPA analysis of DNA copy number changes, a specific probe mixture with 48 sub- telomeric probe-sets for all chromosomes was used according to the manufacturer's recommendations (Salsa P036, MRC-Holland b.v., Amsterdam, The Netherlands). In short, approximately 100 ng of DNA in 5 μl was denaturated at 98 °C for 5 min and subsequently hybridized overnight with a mix of sub-telomeric probe-pairs, each consisting of two oligonucleotides (hemiprobes) that recognize adjacent DNA sequences. On day 2, the adjacently hybridized hemiprobes were ligated. After denaturation, PCR was performed with two universal PCR primers, amplifying all probe-pairs in one reaction! Experiments for both test and reference samples were carried out in triplicate. Analysis of the MLPA PCR products was performed on an ABI model 3100 16-capillary sequencer (Applied Biosystems, Warrington, UK).

Immunohisto chemical analysis

[48] Five-micron sections were used for Ki67 staining that was performed following the manufacturer's instructions (DAKO, Carpinteria, CA), and scored as percentage staining nuclei. EGFR, phospho-EGFR, phospho-MAPK, and phospho-Akt (Cell Signaling Technology, Beverly, MA) staining was performed using citrate-steam recovery, followed by Catalyzed Signal Amplification (DAKO, Carpinteria, CA).

EXAMPLE 2

Efficacy of erlotinib, cetuximab and the combination of erlotinib plus cetuximab

[49] Initially, ten patient derived tumors from our colony were tested for drug efficacy. Two tumors (198 and 410) were highly sensitive to EGFR targeting including tumor regressions (Figure 1). All other tumors were resistent with best treatment resulting only in modest growth inhibition. Overall erlotinib showed marginally higher potency compared with cetuximab, with an average T/C of 54% versus 65% when the indexes of all 10 cases were pooled together. The combined therapy had an average 45% T/C.

EXAMPLE 3

Gene expression analysis

[50] We approached gene expression analysis by seeking a tool that would enable group interrogation. In order to rationally explore this hypothesis in a reproducible fashion, we used gene set enrichment analysis (GSEA) and pathway analysis, an approach that offers an unbiased global search for genes that are coordinately regulated in pre-defined pathways (in this case per the KEGG database (17)) rather than interrogating expression differences of single genes. Overall 98 gene sets were enriched in the sensitive cases, but only eight gene sets had a nominal p-value < 0.01 (Figure 2A). Out of these eight gene sets, four of them have a false-discovery rate (FDR) < 0.10. One of this four was the EGFR signaling pathway that according to the KEGG database annotation consists of 87 genes. Of these, the 25 genes that contributed most to the enrichment result were defined as the core enrichment genes (enrichment plot illustrated in Figure 2B). These include seven ligands (EGF, HB-EGF, TGFa, BTC, EPR, NRG2, and NRG4), and pathway genes such as MAPK8-10, Akt3, NRAS, PlKSCA, STATS and p27 were upregulated in the sensitive tumors. The heatmap of these core enrichment genes is shown in Figure 2C and Figure 2D illustrates the location of these core enrichment genes in the EGFR signaling pathway. These results suggest that global increase in the expression and activation of pathway-related genes is linked to drug susceptibility.

EXAMPLE 4

Prospective prediction in the validation set

[51] Thus, we hypothesized that by querying the EGFR pathway we would predict the response of EGFR inhibitors, and used an independent set of eight tumors of which no efficacy results were known. We tested whether the expression profiles of the core gene members in the EGFR signaling pathway could be used as discriminative features for prediction. We built a logistic regression classifier from these core gene features based on these ten cases (learning set). Next, we collected gene expression profiles of eight independent cases (159, 185, 219, 247, 281, 294, 354 and 420) and employed the core gene classifier to predict their drug response to EGFR inhibition. The classifier identified 219 as sensitive and the rest to be resistant to EGFR inhibition. To test this prediction, we conducted drug efficacy testing on the eight tumors with erlotinib, the EGFR inhibitor that is approved for use in pancreatic cancer patients. The tumor 219 was sensitive (T/C of 3%) and the other seven predictions were also accurate (global GSEA prediction Chi square P < 0.001), as those cases were uniformly resistant to erlotinib.

EXAMPLE 5

Specificity of the signature

[52] To exclude the possibility that these tumors were inherently sensitive/labile to any treatment, erlotinib efficacy was correlated with the response in these cases to gemcitabine and CI 1040, a cytotoxic agent and a signal transduction inhibitor with similar level of efficacy (3 of 15), respectively. No correlation existed between the responses to these three treatments, indicating that each tumor's response depends on inherent features. The EGFR core signature is not indicative of response to these drugs, and by GSEA the EGFR pathway is not differentially up-regulated in the gemcitabine or CI 1040 responsive tumors.

EXAMPLE 6

Gene mutation, FISH and multiplex ligation-dependent probe amplification (MLPA) analysis

[53] Next, we explored factors that have been related to EGFR sensitivity in other disease types. EGFR, KRAS, BRAF, and PIK3CA mutation and amplification profile of EGFR and HER2 by FISH was explored in these tumors (Fig. 4 (Table I)). No mutations or deletions were found in exons 18, 19 or 21 of the EGFR or in exon 11 of BRAF. One sensitive tumor had a mutation in the PIKiCA gene. KRAS mutations were prevalent but both mutated and wild-type cases were sensitive to EGFR inhibition. No specific mutation or combination of mutations was associated with sensitivity to EGFR inhibitors. EGFR gene amplification was not detected, but in seven tumors there was high polysomy (Figure 3A). HER2 gene amplification was found in two specimens. There was no correlation between EGFR or HER2 amplification status and sensitivity. Thus these best candidates to explain differential response to the drugs were not informative.

[54] We then conducted a MLPA analysis of selected genes in the EGFR pathway to investigate whether changes in gene dosage could explain pathway activation by using a novel, high-throughput quantitative method (22) (Figure 3B). Sensitive tumors had copy gains of key EGFR pathway genes such as AKTl, NRAS, PIKiCA, SRC and no change in HER4, but multiple resistant tumors (247, 420) showed similar profiles. EXAMPLE 7

Analysis of pathways activated by underlying genetic abnormalities.

[55] To investigate whether the presence of the EGFR signature was related to any of the above individual genetic factors, we determined the GSEA signatures of the cohort of 18 tumors stratifying by each of the parameters. The strata was mutated versus non-mutated for KRAS and PIK3CA, a score of 5-6 versus 1-4 for EGFR and HER2 FISH, and increased copy number versus no increase in each of the eight genes of the MLPA analysis. The EGFR pathway was not present in the top scoring pathways of the cases with neither mutations nor increased copy number/dosage compared to the normal state tumors, indicating that none of these individual features was causing per se the EGFR pathway over-expression.

EXAMPLE 8 lmmunohistochemistry (IHC) assessment

[56] Finally, to determine the impact of EGFR pathway gene over-expression at the protein level, we determined the baseline expression of selected elements in the EGFR by IHC. As shown in Fig. 5 (Table 2) sensitive cases had a globally activated EGFR pathway profile (high EGFR, phospho-MAPK and phospho-Akt positivity), but resistant cases (215, 265, 185) did stain too for these individual markers. So it can be concluded that pathway activation by IHC is necessary but not sufficient to confer sensitivity to anti- EGFR therapy.

EXAMPLE 9 Conclusions

[57] We used GSEA, an unbiased analysis that concentrates in the detection of modest but coordinated changes in expression of genes involved in a common pathway or biological function, and that is a tool that has helped identify growth-driving pathways in cancer (30). By GSEA analysis the EGFR pathway was among the highest expressing out of the 197 pathways on which the 55,000 transcripts are distributed. The gene classifier was capable of correctly identifying prospectively 8 cases and then the whole cohort of 18 cases (3 as sensitive, 15 as resistant; P < 0.001). Interestingly, the MAPK pathway was also among the top scoring sets. This highlights the plausibility of the findings as both pathways are interconnected. It is relevant to note that EGFR pathway components are present in some of the other differentially upregulated sets, such as the glioma pathway. The core gene components that drove EGFR pathway activation were ligands and positive effectors, indicating an activating effect.

[58] Products from some of these genes (Akt, MAPK) were shown to have increased activation by protein analysis. Higher pathway activation by IHC was linked with higher activity, but the presence of such did not necessarily predict an antitumor effect. This dichotomy versus EGFR pathway over-expression at the mRNA and protein levels may imply that whereas EGFR pathway proteins can also be trans-activated by other transducers (and thus it is a "secondary" or "passenger" activation), only when it is driven by an increase in gene expression does it indicate a "primary" or "driver" alteration that will be effectively tackled by a pharmacological intervention.

[59] It is of interest to note that the EGFR pathway specifically predicted the response to EGFR inhibitors as it was not predictive of response to gemcitabine and CI 1040, a cytotoxic drug commonly used in pancreatic cancer treatment and a signal transduction (MEK) inhibitor. The observation that a stronger pathway association existed in targeted versus cytotoxic agents is relevant as supports the notion that sensitivity to the former is related to pathway expression patterns.

[60] Gene expression analysis has shown promise to characterize cancer, as primary genetic alterations prompting or maintaining a cancer phenotype ultimately manifest by differential expression of genes required to sustain such state. Whereas proteomic assessment may be considered the ultimate step in these processes, current technology has not produced proteomic tools ready for use in a clinical setting. Recently a gene expression platform derived from global unbiased testing received regulatory approval for risk prognostication for breast cancer (13, 14). The potential applicability of the presented findings is that either using commercial global gene expression-based kits or customizing a platform similar to the MammaPrint™, the core EGFR signature could be readily incorporated into a clinical trial. If successful this would avoid unnecessary toxicities from inefficacious treatments and an overall increase in the quality of delivered anticancer care.

[61] Several reports have explored alterations in the EGFR pathway that are known to determine the sensitivity to EGFR inhibitors in other tumor types, but have been uniformly negative. In a series of 43 pancreatic cancers no EGFR mutations were documented (8). In a second report in 66 patients one subject had an EGFR mutation, another had a true amplification, and 26 had low-level EGFR amplification (9). It is relevant to note the similar incidence of these events in clinical series and in our model. However, after analyzing those EGFR markers that are relevant in other diseases no individual feature or alteration reliably identified the sensitive tumors to EGFR inhibition. Mutation in the KRAS gene, an almost universal finding in pancreatic cancer, is unlikely to be a resistant mechanism in this disease as opposed to lung or colorectal cancer. Otherwise, the positive outcome of the pivotal trial is difficult to explain. In this work, two of the sensitive tumors in fact carried KRAS mutations.

[62] Despite initial MLPA analysis in the learning set suggested that the two sensitive tumors had gains in pathway-related genes such as EGFR, PIK3CA and Aktl, this did not identify a sensitive case prospectively, and after incorporating to the analysis the 8 additional cases from the validation set no solid pattern was found. We are uncertain as to the reasons of the lack of correlation between EGFR and HER2 FISH and MLPA results in our samples. The EGFR pathway was not present in the top scoring pathways of the cases with neither mutations nor increased copy number/dosage, suggesting that none of these individual genetic abnormalities was responsible for the observed pattern. Altogether this suggests that the mechanistic basis for higher pathway gene expression is not related to a single genetic alteration.

[63] For this work we took advantage of the PancXenoBank, a collection of individual pancreas cancer tumors obtained from patients with pancreatic cancer (15). Generally, before entering clinical trials, new agents are tested against high-passage cell lines and typically a few xenografts established from these lines. It is unclear how representative those models are of the biology of pancreatic cancer, in view of the historic disconnect between preclinical and clinical results in this disease. We have shown that directly xenografted tumors retain the key features of the originator tumor, represent the heterogeneity of the disease, are easily amenable to treatment with different drugs, and offer and endless source to tumors for complex biological studies (31). Indeed in this study we were able to conduct a large set of complex biological studies as well as compare the activity of different agents against each individual tumor. While obviously clinical specimens and clinical response data is more valuable, the detailed biological and therapeutic assessment conducted in this work is not possible in the clinical setting as patients are not treated with more than 2 or 3 drugs and available tissues are not adequate in quantity and quality for broad biological testing. We propose this platform is useful for screening purposes and best candidate selection that now can be tested in focused clinical studies.

[64] In summary, EGFR inhibition showed activity in a subset of cases from a direct xenograft pancreatic cancer platform. This subset was characterized by EGFR pathway upregulation as assessed by gene expression. The EGFR pathway activation only predicted response to EGFR inhibitors and not to other agents. No single genetic abnormality, including mutations and copy number variation in key components of the pathway was individually responsible for the global activation of it. The data suggest the presence of global pathway activation rather than specific oncogene addiction. These results can be readly applied to clinical trials with EGFR inhibitors in pancreatic cancer and provide a framework to explore biomarkers of drug activity in this disease.

References— each of which is expressly incorporated herein:

1. Moore MJ, Goldstein D, Hamm J, Figer A, Hecht JR, Gallinger S, et al. Erlotinib plus gemcitabine compared with gemcitabine alone in patients with advanced pancreatic cancer: a phase HI trial of the National Cancer Institute of Canada Clinical Trials Group. J Clin Oncol 2007;25(15): 1960-6.

2. Feinberg AP, Vogelstein B. Hypomethylation distinguishes genes of some human cancers from their normal counterparts. Nature 1983;301(5895):89-92.

3. Lynch TJ, Bell DW, Sordella R, Gurubhagavatula S, Okimoto RA, Brannigan BW, et al. Activating mutations in the epidermal growth factor receptor underlying responsiveness of non- small-cell lung cancer to gefitinib. N Engl J Med 2004;350(21):2129-39.

4. Paez JG, Janne PA, Lee JC, Tracy S, Greulich H, Gabriel S, et al. EGFR mutations in lung cancer: correlation with clinical response to gefitinib therapy. Science 2004;304(5676): 1497-500.

5. Cappuzzo F, Hirsch FR, Rossi E, Bartolini S, Ceresoli GL, Bemis L, et al. Epidermal growth factor receptor gene and protein and gefitinib sensitivity in non-small-cell lung cancer. J Natl Cancer Inst 2005;97(9):643-55.

6. Cappuzzo F, Varella-Garcia M, Shigematsu H, Domenichini I, Bartolini S, Ceresoli GL, et al. Increased HER2 gene copy number is associated with response to gefitinib therapy in epidermal growth factor receptor-positive non-small-cell lung cancer patients. J Clin Oncol 2005;23(22):5007-18.

7. Hirsch FR, Varella-Garcia M, Bunn PA, Jr., Franklin WA, Dziadziuszko R, Thatcher N, et al. Molecular predictors of outcome with gefitinib in a phase III placebo-controlled study in advanced non-small-cell lung cancer. J Clin Oncol 2006;24(31):5034-42.

8. Immervoll H, Hoem D, Kugarajh K, Steine SJ, Molven A. Molecular analysis of the EGFR-RAS-RAF pathway in pancreatic ductal adenocarcinomas: lack of mutations in the BRAF and EGFR genes. Virchows Arch 2006;448(6):788-96.

9. Lee J, Jang KT, Ki CS, Lim T, Park YS, Lim HY, et al. Impact of epidermal growth factor receptor (EGFR) kinase mutations, EGFR gene amplifications, and KRAS mutations on survival of pancreatic adenocarcinoma. Cancer 2007;109(8):1561-1569. 10. Saxby AJ, Nielsen A, Scarlett CJ, Clarkson A, Morey A, Gill A, et al. Assessment of HER-2 status in pancreatic adenocarcinoma: correlation of immunohistochemistry, quantitative real-time RT-PCR, and FISH with aneuploidy and survival. Am J Surg Pathol 2005;29(9):l 125- 34.

11. Tsiambas E, Karameris A, Dervenis C, Lazaris AC, Giannakou N, Gerontopoulos K, et al. HER2/neu expression and gene alterations in pancreatic ductal adenocarcinoma: a comparative immunohistochemistry and chromogenic in situ hybridization study based on tissue microarrays and computerized image analysis. Jop 2006;7(3):283-94.

12. Sjoblom T, Jones S, Wood LD, Parsons DW, Lin J, Barber TD, et al. The consensus coding sequences of human breast and colorectal cancers. Science 2006;314(5797):268-74.

13. van de Vijver MJ, He YD, van't Veer LJ, Dai H, Hart AA, Voskuil DW, et al. A gene- expression signature as a predictor of survival in breast cancer. N Engl J Med 2002;347(25): 1999-2009.

14. Glas AM, Floore A, Delahaye LJ, Witteveen AT, Pover RC, Bakx N, et al. Converting a breast cancer microarray signature into a high-throughput diagnostic test. BMC Genomics 2006;7:278.

15. Rubio-Viqueira B, Jimeno A, Cusatis G, Zhang X, Iacobuzio-Donahue C, Karikari C, et al. An In vivo Platform for Translational Drug Development in Pancreatic Cancer. Clin Cancer Res 2006; 12(15):4652-61.

16. Subramanian A, Tamayo P, Mootha VK, Mukherjee S, Ebert BL, Gillette MA, et al. Gene set enrichment analysis: a knowledge-based approach for interpreting genome- wide expression profiles. Proc Natl Acad Sci U S A 2005;102(43):15545-50.

17. Kanehisa M, Goto S, Hattori M, Aoki-Kinoshita KF, Itoh M, Kawashima S, et al. From genomics to chemical genomics: new developments in KEGG. Nucleic Acids Res 2006;34(Database issue):D354-7.

18. Frank E, Hall M, Trigg L, Holmes G, Witten IH. Data mining in bioinformatics using Weka. Bioinformatics 2004;20(15):2479-81.

19. Shi C, Eshleman SH, Jones D, Fukushima N, Hua L, Parker AR, et al. LigAmp for sensitive detection of single-nucleotide differences. Nat Methods 2004;l(2): 141-7. 20. Embuscado EE, Laheru D, Ricci F, Yun KJ, de Boom Witzel S, Seigel A, et al. Immortalizing the Complexity of Cancer Metastasis: Genetic Features of Lethal Metastatic Pancreatic Cancer Obtained from Rapid Autopsy. Cancer Biol Ther 2005 ;4(5).

21. Samuels Y, Wang Z, Bardelli A, Silliman N, Ptak J, Szabo S, et al. High frequency of mutations of the PIK3CA gene in human cancers. Science 2004;304(5670):554.

22. Schouten JP, McElgunn CJ, Waaijer R, Zwijnenburg D, Diepvens F, Pals G. Relative quantification of 40 nucleic acid sequences by multiplex ligation-dependent probe amplification. Nucleic Acids Res 2002;30(12):e57.

23. Hanahan D, Weinberg RA. The hallmarks of cancer. Cell 2000;100(l):57-70.

24. Teyssier JR, Bartram CR, Deville J, Potron G, Pigeon F. c-abl Oncogene and chromosome 22 "bcr" juxtaposition in chronic myelogenous leukemia. N Engl J Med 1985;312(21):1393-4.

25. Slamon DJ, Godolphin W, Jones LA, Holt JA, Wong SG, Keith DE, et al. Studies of the HER-2/neu proto-oncogene in human breast and ovarian cancer. Science 1989;244(4905):707- 12.

26. Slamon DJ, Leyland- Jones B, Shak S, Fuchs H, Paton V, Bajamonde A, et al. Use of chemotherapy plus a monoclonal antibody against HER2 for metastatic breast cancer that overexpresses HER2. N Engl J Med 2001 ;344(l l):783-92.

27. Shepherd FA, Rodrigues Pereira J, Ciuleanu T, Tan EH, Hirsh V, Thongprasert S, et al. Erlotinib in previously treated non-small-cell lung cancer. N Engl J Med 2005;353(2): 123-32.

28. O'Brien SG, Guilhot F, Larson RA, Gathmann I, Baccarani M, Cervantes F, et al. Imatinib compared with interferon and low-dose cytarabine for newly diagnosed chronic-phase chronic myeloid leukemia. N Engl J Med 2003;348(l l):994-1004.

29. Jimeno A, Rubio-Viqueira B, Amador ML, Oppenheimer D, Bouraoud N, Kulesza P, et al. Epidermal growth factor receptor dynamics influences response to epidermal growth factor receptor targeted agents. Cancer Res 2005;65(8):3003-10.

30. Vivanco I, Palaskas N, Tran C, Finn SP, Getz G, Kennedy NJ, et al. Identification of the JNK Signaling Pathway as a Functional Target of the Tumor Suppressor PTEN. Cancer Cell 2007;l l(6):555-69. 31. Jimeno A, Rubio-Viqueira B, Amador ML, Grunwald V, Maitra A, Iacobuzio-Donahue C, et al. Dual mito gen-activated protein kinase and epidermal growth factor receptor inhibition in biliary and pancreatic cancer. MoI Cancer Ther 2007;6(3): 1079-88.

Claims

WE CLAIM:

1. A method for predicting sensitivity or resistance of a tumor to an EGFR inhibitory drug or biological, comprising: comparing expression of at least 2 genes of the EGFR signaling pathway in a tumor sample of a subject to expression of the at least 2 genes in control samples of a plurality of human tumors resistant to an EGFR inhibitory drug, and determining an average of the differences of expression of the at least 2 genes relative to the control samples, wherein the genes are selected from the group consisting of ABL2, AKT2, AKT3, BTC, CAMK2A, EGF, EREG, FRAPl, HBEGF, HRAS, MAPKlO, MAPK8, MAPK9, NRAS, NRG2, NRG4, PAK6, PIK3CA, PIK3R3, PRKCBl, SHC3, SHC4, SOSl, SRC, TGFA, ABLl, AKTl, ARAF, AREG, BAD, BRAF, CAMK2B, CAMK2D, CAMK2G, CBL, CBLB, CBLC, CDKNlA, CDKNlB, CRK, CRKL, EGFR, EIF4EBP1, ELKl, ERBB2, ERBB3, ERBB4, GABl, GRB2, GSK3B, JUN, KRAS, MAP2K1, MAP2K2, MAP2K4, MAP2K7, MAPKl, MAPK3, MYC, NCKl, NCK2, NRGl, NRG3, PAKl, PAK2, PAK3, PAK4, PAK7, PIK3CB, PIK3CD, PIK3CG, PIK3R1, PIK3R2, PIK3R5, PLCGl, PLCG2, PRKCA, PRKCG, PTK2, RAFl, RPS6KB1, RPS6KB2, SHCl, SHC2, SOS2, STAT5A, and STAT5B; predicting that the tumor of the subject will be sensitive to EGFR inhibitory drugs if the average of the differences of expression of the at least 2 genes is higher in the tumor sample than in the control samples, or predicting that the tumor of the subject will be resistant to EGFR inhibitory drugs if the average of the differences of expression of the at least 2 genes is lower in the tumor sample than in the control samples.

2. The method of claim 1 further comprising: measuring expression of at least 2 genes of the EGFR signaling pathway in the tumor sample by measuring mRNA prior to the step of comparing.

3. The method of claim 1 further comprising: recommending an EGFR inhibitory drug for a subject whose tumor is predicted to be sensitive to EGFR inhibitory drugs, or recommending that an EGFR inhibitory drug not be prescribed for a subject whose tumor is not identified as sensitive to EGFR inhibitory drugs.

4. The method of claim 1 wherein the cancer is pancreatic cancer.

5. The method of claim 1 wherein the cancer is lung cancer.

6. The method of claim 1 wherein the cancer is breast cancer.

7. The method of claim 3 wherein the inhibitor is erlotinib.

8. The method of claim 3 wherein the inhibitor is cetuximab.

9. The method of claim 3 wherein the inhibitor is gefitinib.

10. The method of claim 1 wherein average expression of at least 25 genes is compared.

11. The method of claim 1 wherein average of the differences of expression of at least 25 genes is determined, said set of 25 genes comprising: AB L2, AKT2, AKT3, BTC, CAMK2A, EGF, EREG, FRAPl, HBEGF, HRAS, MAPKlO, MAPK8, MAPK9, NRAS, NRG2, NRG4, PAK6, PIK3CA, PIK3R3, PRKCBl, SHC3, SHC4, SOSl , SRC, and TGFA.

12. The method of claim 1 wherein average of the difference of expression of at least 40 genes is determined.

13. The method of claim 1 wherein average of the difference of expression of at least 55 genes is determined.

14. The method of claim 1 wherein average of the differences of expression of at least 87 genes is determined, said set of 87 genes comprising: AB L2, AKT2, AKT3, BTC, CAMK2A, EGF, EREG, FRAPl , HBEGF, HRAS, MAPKlO, MAPK8, MAPK9, NRAS, NRG2, NRG4, PAK6, PIK3CA, PIK3R3, PRKCBl, SHC3, SHC4, SOSl, SRC, TGFA, ABLl, AKTl , ARAF, AREG, BAD, BRAF, CAMK2B, CAMK2D, CAMK2G, CBL, CBLB, CBLC, CDKNlA, CDKNlB, CRK, CRKL, EGFR, EIF4EBP1, ELKl, ERBB2, ERBB3, ERBB4, GABl, GRB2, GSK3B, JUN, KRAS, MAP2K1 , MAP2K2, MAP2K4, MAP2K7, MAPKl, MAPK3, MYC, NCKl, NCK2, NRGl, NRG3, PAKl, PAK2, PAK3, PAK4, PAK7, PIK3CB, PIK3CD, PIK3CG, PIK3R1, PIK3R2, PIK3R5, PLCGl, PLCG2, PRKCA, PRKCG, PTK2, RAFl, RPS6KB1, RPS6KB2, SHCl, SHC2, SOS2, STAT5A, and STAT5B.

15. The method of claim 1 wherein said step of comparing is performed by computer.

16. The method of claim 1 further comprising: transmitting the prediction of sensitivity or resistance to the subject or subject's physician.

17. A method for predicting resistance of a tumor to an EGFR inhibitory drug, comprising: comparing expression of a gene in a tumor sample of a subject to expression of a plurality of control samples of tumors sensitive to an EGFR inhibitory drug, wherein the gene is selected from the group consisting of ATP2A3, DDC, HLXB9, KIAA0282, MAOA, MAPK7, PIP5K1B, PLCD3, PRSS3, TAC3, THRA, THRB and VIPRl ; predicting that the tumor of the subject is resistant to EGFR inhibitory drugs if the expression is higher in the tumor sample than in the control sample.

18. The method of claim 17 further comprising: recommending an alternate therapy to EGFR inhibitory drugs or biologicals to the subject.

19. The method of claim 17 further comprising: if the tumor sample is found to have higher expression of ATP2A3 than the control samples, then recommending a combination therapy of an EGFR inhibitory drug or biological and a sarcoplasmic reticulum ATPase calcium inhibitor.

20. The method of claim 17 further comprising: if the tumor sample is found to have higher expression of DDC than the control samples, then recommending a combination therapy of an EGFR inhibitory drug or biological and a DDC inhibitor.

21. The method of claim 17 further comprising: if the tumor sample is found to have higher expression of MAOA than the control samples, then recommending a combination therapy of an EGFR inhibitory drug or biological and a MAOA inhibitor.

22. The method of claim 17 further comprising: if the tumor sample is found to have higher expression of ATP2A3 than the control samples, then recommending a combination therapy of an EGFR inhibitory drug or biological and Artimesin.

23. The method of claim 17 further comprising: if the tumor sample is found to have higher expression of DDC than the control samples, then recommending a combination therapy of an EGFR inhibitory drug or biological and Carbidopa.

24. The method of claim 17 further comprising: if the tumor sample is found to have higher expression of MAOA than the control samples, then recommending a combination therapy of an EGFR inhibitory drug or biological and Phenelzine.

25. The method of any of claims 19-24 wherein the EGFR inhibitory drug or biological is selected from the group consisting of erlotinib, cetuximab, and gefϊtinib.

26. A kit for treating a tumor resistant to EGFR inhibitory drugs, comprising: an EGFR inhibitory drug or biological; and an inhibitor of a protein selected from the group consisting of: ATP2A3, DDC, HLXB9, KIAA0282, MAOA, MAPK7, PIP5K1B, PLCD3, PRSS3, TAC3, THRA, THRB and VIPRl ; wherein the EGFR inhibitory drug or biological and the inhibitor are in a single or separated containers.

27. The kit of claim 26 wherein the EGFR inhibitory drug or biological is selected from the group consisting of erlotinib, cetuximab, and gefitinib.

28. The kit of claim 26 wherein the inhibitor is selected from the group consisting of artimesin, carbodopa, and phenelzine.

29. A diagnostic reagent for assessing susceptibility or resistance to EGFR inhibitory drugs comprising: a solid support comprising oligonucleotide probes which are complementary to at least 2 genes are selected from the group consisting of ABL2, AKT2, AKT3, BTC, CAMK2A, EGF, EREG, FRAPl, HBEGF, HRAS, MAPKlO, MAPK8, MAPK9, NRAS, NRG2, NRG4, PAK6, PIK3CA, PIK3R3, PRKCBl, SHC3, SHC4, SOSl, SRC, TGFA, ABLl , AKTl, ARAF, AREG, BAD, BRAF, CAMK2B, CAMK2D, CAMK2G, CBL, CBLB, CBLC, CDKNlA, CDKNlB, CRK, CRKL, EGFR, EIF4EBP1, ELKl, ERBB2, ERBB3, ERBB4, GABl, GRB2, GSK3B, JUN, KRAS, MAP2K1, MAP2K2, MAP2K4, MAP2K7, MAPKl , MAPK3, MYC, NCKl, NCK2, NRGl, NRG3, PAKl, PAK2, PAK3, PAK4, PAK7, PIK3CB, PIK3CD, PIK3CG, PHGRl, PIK3R2, PIK3R5, PLCGl, PLCG2, PRKCA, PRKCG, PTK2, RAFl, RPS6KB1, RPS6KB2, SHCl, SHC2, S0S2, STAT5A, STAT5B; ATP2A3, DDC, HLXB9, KIAA0282, MAOA, MAPK7, PIP5K1B, PLCD3, PRSS3, TAC3, THRA, THRB and VIPRl ; wherein the at least 2 genes selected from the group comprise at least 50 % of the genes for which the solid support contains oligonucleotide probes.

30. A method of providing a prediction of sensitivity or resistance of a tumor to an EGFR inhibitory drug or biological, comprising: obtaining expression data for at least 2 genes of the EGFR signaling pathway in a tumor sample of a subject, wherein the genes are selected from the group consisting of ABL2, AKT2, AKT3, BTC, CAMK2A, EGF, EREG, FRAPl, HBEGF, HRAS, MAPKlO, MAPK8, MAPK9, NRAS, NRG2, NRG4, PAK6, PIK3CA, PIK3R3, PRKCBl, SHC3, SHC4, SOSl, SRC, TGFA, ABLl, AKTl , ARAF, AREG, BAD, BRAF, CAMK2B, CAMK2D, CAMK2G, CBL, CBLB, CBLC, CDKNlA, CDKNlB, CRK, CRKL, EGFR, EIF4EBP1, ELKl, ERBB2, ERBB3, ERBB4, GABl, GRB2, GSK3B, JUN, KRAS, MAP2K1, MAP2K2, MAP2K4, MAP2K7, MAPKl, MAPK3, MYC, NCKl, NCK2, NRGl, NRG3, PAKl, PAK2, PAK3, PAK4, PAK7, PIK3CB, PIK3CD, PIK3CG, PIK3R1, PIK3R2, PIK3R5, PLCGl, PLCG2, PRKCA, PRKCG, PTK2, RAFl, RPS6KB1, RPS6KB2, SHCl, SHC2, S0S2, STAT5A, and STAT5B; analyzing the data by comparing expression of the at least 2 genes to expression of the at least 2 genes in control samples of a plurality of human tumors resistant to an EGFR inhibitory drug, and determining an average of the differences of expression of the at least 2 genes; transmitting to the subject or the subject's physician or a diagnostic laboratory which generated the expression data a prediction that the tumor of the subject will be sensitive to EGFR inhibitory drugs if the average of the differences of expression is higher in the tumor sample than in the control samples, or a prediction that the tumor of the subject will be resistant to EGFR inhibitory drugs if the average of the differences of expression is lower in the tumor sample than in the control samples.

31. The method of claim 30 wherein the steps of obtaining and transmitting are performed on the internet.

32. The method of claim 1, 17, or 30 wherein the control samples comprise human xenografts.

33. The method of claim 1, 17, or 30 wherein the control samples comprise human biopsied tumors.

34. The method of claim 1, 17, or 30 wherein the control samples comprise human xenografts and human biopsied rumors.

35. A computer that accesses a database of expression data of EGFR pathway genes in EGFR inhibitor-resistant tumor samples and compares the database values to test sample values and provides a prediction of EGFR inhibitor-sensitivity or EGFR inhibitor-resistance of the test sample based on an average of the differences of the expression of the genes compared to the database values, wherein the expression data of EGFR pathway genes which are compared and the differences of the expression averaged comprise at least 25 of ABL2, AKT2, AKT3, BTC, CAMK2A, EGF, EREG, FRAPl, HBEGF, HRAS, MAPKlO, MAPK8, MAPK9, NRAS, NRG2, NRG4, PAK6, PIK3CA, PIK3R3, PRKCBl, SHC3, SHC4, SOSl, SRC, TGFA, ABLl, AKTl, ARAF, AREG, BAD, BRAF, CAMK2B, CAMK2D, CAMK2G, CBL, CBLB, CBLC, CDKNlA, CDKNlB, CRK, CRKL, EGFR, EIF4EBP1, ELKl, ERBB2, ERBB3, ERBB4, GABl, GRB2, GSK3B, JUN, KRAS, MAP2K1, MAP2K2, MAP2K4, MAP2K7, MAPKl, MAPK3, MYC, NCKl, NCK2, NRGl, NRG3, PAKl, PAK2, PAK3, PAK4, PAK7, PIK3CB, PIK3CD, PIK3CG, PIK3R1, PIK3R2, PIK3R5, PLCGl, PLCG2, PRKCA, PRKCG, PTK2, RAFl, RPS6KB1, RPS6KB2, SHCl, SHC2, S0S2, STAT5A, and STAT5B.

36. The computer of claim 35 wherein the database expression data comprises human xenograft data.

37. The computer of claim 35 wherein the database expression data comprises human biopsied tumor data.

38. The computer of claim 35 wherein the database expression data comprises human biopsied tumor data and xenograft data.

39. The computer of claim 35 which can further access a database of expression data from EGFR inhibitor-sensitive tumors for one or more genes selected from the group consisting of ATP2A3, DDC, HLXB9, KIAA0282, MAOA, MAPK7, PIP5K1B, PLCD3, PRSS3, TAC3, THRA, THRB and VIPRl ; wherein the computer compares the database values to test sample values and provides a prediction of EGFR inhibitor-resistance of the test sample based on expression values of the genes compared to the database values.

40. The computer of claim 35 which provides a recommendation of a combination treatment based on expression levels of ATP2A3, DDC, and/or MAOA.

41. A method for treating a patient with a tumor predicted to be resistant to EGFR inhibitory drugs, comprising: administering to the patient an EGFR inhibitory drug or biological; and administering to the patient an inhibitor of a protein selected from the group consisting of ATP2A3, DDC, HLXB9, KIAA0282, MAOA, MAPK7, PIP5K1B, PLCD3,

PRSS3, TAC3, THRA, THRB and VIPRl .

42. The method of claim 41 wherein the EGFR inhibitory drug or biological is selected from the group consisting of erlotinib, cetuximab, and gefitinib.

43. The method of claim 41 wherein the inhibitor is selected from the group consisting of artimesin, carbidopa, and phenelzine.

44. The method of claim 41 wherein if the tumor over expresses MAOA relative to EGFR inhibitor sensitive tumors, an inhibitor of MAOA is administered.

45. The method of claim 41 wherein if the tumor over expresses DDC relative to EGFR inhibitor sensitive tumors, an inhibitor of DDC is administered.

46. The method of claim 41 wherein if the tumor over expresses ATP2A3 relative to EGFR inhibitor sensitive tumors, an inhibitor of ATP2A3 is administered.