NOVEL TUMOR MARKER

Abstract

cerns a gene product largely homologous to the epithelial growth factor receptor (EGFR). It further refers to mRNA coding for such epithelial growth factor receptor. The present invention provides such an epithelial growth factor receptor which is characterized in that either exons 12 to 14 or exons 12 to 15 are deleted. These novel variants of the epithelial growth factor receptor can be used for a diagnosis, stratification, therapy guidance of a tumor or therapy guidance of tumor surgery.

Description

BACKGROUND OF THE INVENTION

[0001]The present invention concerns a gene product largely homologous to the epithelial growth factor receptor (EGFR). It further refers to mRNA coding for such epithelial growth factor receptor. Further, the present invention concerns the use of such a gene product as tumor marker, particularly for epithelial tumors like colon cancer, lung cancer, prostate cancer, breast cancer or other solid tumors.

[0002]Histochemical studies of primary tumors often neglect the fact that in normal and tumor cells, different variants (also called derivatives or mutants) of therapy target genes, e.g. EGFR DNA rearrangement, EGFR transcription and/or protein variants exist. The growth factor receptor EGFR is of special importance in most epithelial tumors like colon cancer, lung cancer, prostate cancer or breast cancer, because expression in primary tumors correlates with a shorter survival time/rate. Therefore, an expression of EGFR is a prerequisite for EGFR based therapy (antibodies like Erbitux or receptor tyrosine kinase inhibitors like Iressa). But about 30% of patients do not respond to antibody therapies. The reason could be that a part of the N-terminal domain of die EGFR is deleted as has been described for the variant EGFRvIII. These variants e.g. naturally occurring through alternative splicing or in tumor cells through somatic mutation) exist in the whole EGFR sequence. Soluble proteins arise without cytosolic C-terminal ends as well as membrane anchored proteins without the N-terminal ligand-binding domain. The following tumor-specific variants of EGFR exist (Kuan et al., 2001 Endocr. Relat. Cancer 8, 83-96: "EGF mutant receptor vII as a molecular target in cancer therapy"):

EGFR wt, EGFR vI, EGFR vII, EGFR vIII, EGFR vIII/Δ12-13, EGFR TDM/2-7, EGFR vIV, EGFR vV, EGFR TDM/18-25 and EGFR TDM/18-26.

[0003]Variant vII is the most abundant in different tumors such as breast or ovarian cancer but not all tumors express this variant (e.g. small cell lung carcinoma (SCLC)).

[0004]Another set of variants are short deletions (<10 amino acids) or single nucleotide mutations in the receptor tyrosine kinase domain of the EGFR that are expressed in tumor cells in non-small cell lung carcinoma (NSCLC) or neuroblastoma.

[0005]Today, two new classes of therapies exist: (1) antibodies that compete with the ligands of EGFR for example EGF or TGF-α and block the N-terminal binding domain of the receptor (Erbitux) and (2) small molecular weight receptor tyrosine kinase inhibitors (Iressa, Tarceva) that bind to the intracellular receptor tyrosine kinase domain and block the ATP binding site reversibly or irreversibly. Both therapies eliminate signal transduction of the wild-type EGFR.

[0006]But the clinical experience shows that not all patients (about 20%) respond to such therapies even if the primary tumor was tested positive for EGFR expression by IHC and/or amplification with FISH. On the other hand, patients, which were tested negative with immunohistochemical methods, sometimes respond to Erbitux therapy. One reason can be that not all patients express the wild-type EGFR but some of the EGFR variants.

[0007]Variants without the N-terminal binding domain (e.g. variant III) do not bind Pantimumab. EGFR variants without somatic mutations in the receptor tyrosine kinase domain need higher doses of, for example Iressa, to block the whole signal transduction of the EGFR. Thus, Iressa or Tarceva have a pronounced effect in NSCLC patients with single nucleotide mutations. Some new antibodies against different EGFR variants (e.g. EGFRvIII) are in preclinical studies.

[0008]The present invention addresses the problem of different EGFR variants existing in tumors and normal cells. It is the object of the present invention to provide improved diagnosis, stratification and/or therapy guidance of a tumor.

[0009]This object is solved by the gene product according to claim 1, the gene according to claim 8, the polynucleotide according to claim 9, the methods according to claims 15 and 16 and their use. Further improvements and modifications of the gene product, the gene, the polynucleotide and the methods are given in the respective dependent claims.

[0010]The present invention for the first provides two new EGFR variants which are preferentially expressed in tumor cells. This concerns primarily epithelial tumors, like colon cancer, lung cancer, prostate cancer or breast cancer or any other solid tumor. The new EGFR variants allow the detection of tumor cells even if present detection methods fail. They, therefore, provide improved detection of tumors, improved stratification of tumors and provide the possibility to improve therapy guidance of a tumor therapy or a therapy after tumor surgery or use the gene or gene product as target for therapeutical intervention, e.g. by antibodies, antisense RNA etc.

[0011]The newly provided EGFR variants show deletions of exons 12 to 14 (EGFR c.EX12_--14del) or exons 12 to 15 (EGFR c.EX12_--15del). Such variants have not been described up to date and have not been correlated to tumor cells. As will be shown in the following examples, the new EGFR variants are closely related to tumor cells and therefore provide novel tumor markers. The same is true for the mRNA coding for the new EGFR variants.

[0012]To summarize, the present invention provides new EGFR genomic or splice variants that allow identifying cancer patients since these variants are not expressed in healthy donors. It is further possible to detect cancer patients with a high risk of not responding to certain antibody therapies directed towards the N-terminals of the EGFR protein.

[0013]These new EGFR variants therefore allow for the detection of the presence of a tumor by analyzing a sample for the presence of the novel EGFR variants. As sample, any kind of sample like tissue sample, body fluid sample, like blood samples and the like, are suitable. Of course, samples derived from the primary tumor or from micrometastases are also well suited to stratify the tumor and influence therapy guidance for this tumor before or after tumor surgery.

[0014]In the following, examples for isolation of the novel EGFR variants as well as their statistical detection in tumor samples are provided.

[0015]The new EGFR variants were isolated from mRNA of neuronal cell lines of the brain (U87MG and U251) by RT-PCR with a primer pair with the following sequences: forward primer: 3'AAACTGCACCTCCATCAGTG5' (SEQ ID NO:6) and reverse primer: 3'ATTCGTTGGACAGCCTTCAAG5' (SEQ ID NO:7) under the following conditions:

95° C., 15 min., (94° C., 30 s, 60° C., 30 s, 72° C., 1 min.) for 45 cycles, 72° C., 5 min; 0.5-1 μM of each primer and HotStarTaq Mix (Qiagen GmbH, Hilden, Germany) in a volume of 50 μl. By using this primer pair, unexpectedly, two fragments of about 414 bp and 256 bp were detected.

[0016]In the following, the nucleotide sequence (cDNA) and corresponding amino acid sequence of both fragments is provided. Numbers of positions of the nucleotides are from EGFR wild type sequence X00588.

TABLE-US-00001 Sequence of EGFR c.EX12_14del fragment (414 bp) (SEQ ID NO:2) nt 1266 AAACTGCACCTCCATCAGTGGCGATCTCCACATCCTGCCGGTGGCATTTA GGGGTGACTCCTTCACACATACTCCTCCTCTGGATCCACAGGAACTGGAT ATTCTGAAAACCGTAAAGGAAATCACAGGGTTTTTGCTGATTCAGGCTTG GCCTGAAAACAGGACGGACCTCCATGCCTTTGAGAACCTAGAAATCATAC GCGGCAGGACCAAGCAACAGGACCAGACAACTGTATCCAGTGTGCCCACT ACATTGACGGCCCCCACTGCGTCAAGACCAGCCCGGCAGGAGTCATGGGA GAAAACAACACCCTGGTCTGGAAGTACGCAGACGCCGGCCATGTGTGCCA CCTGTGCCATCCAAACTGCACCTACGGATGCACTGGGCCAGGTCTTGAAG GCTGTCCAACGAAT nt 2103 and the corresponding amino acid sequence (SEQ ID NO:5): NCTSISGDLHILPVAFRGDSFTHTPPLDPQELDILKTVKEITGFLLIQAW PENRTDLHAFENLEIIRGRTKQQDQTTVSSVPTTLTAPTASRPARQESWE KTTPWSGSTQTPAMCATCAIQTAPTDALGQVLKAVQR

[0017]Therein, the underlined amino acid Q is found in exchange for H. The dotted underlined amino acids are modified due to a shift in the reading frame.

[0018]EGFR c.EX12_--14del shows a frame shift in the open reading frame and a premature stop codon in exon 18 resulting in an EGFR protein with a truncated C-terminus. The frame shift results in exchange of H against Q (underlined amino acid) and completely new amino acids (dotted underlined). The following shows the amino acid sequence above which has been completed up to the stop codon in exon 18:

TABLE-US-00002 (SEQ ID NO:4) NCTSISGDLHILPVAFRGDSFTHTPPLDPQELDILKTVKEITGFLLIQAW PENRTDLHAFENLEIIRGRTKQQDQTTVSSVPTTLTAPTASRPARQESWE KTTPWSGSTQTPAMCATCAIQTAPTDALGQVLKAVQRMGLRSRPSPLGWW GPSSCCWWWPWGSASSCEGATSFGSARCGGCCRRGSLWSLLHPVEKLPT KLS* Sequence of EGFR c.EX12_15del fragment (256 bp) (SEQ ID NO:1) nt 1266 AAACTGCACCTCCATCAGTGGCGATCTCCACATCCTGCCGGTGGCATTTA GGGGTGACTCCTTCACACATACTCCTCCTCTGGATCCACAGGAACTGGAT ATTCTGAAAACCGTAAAGGAAATCACAGGGTTTTTGCTGATTCAGGCTTG GCCTGAAAACAGGACGGACCTCCATGCCTTTGAGAACCTAGAAATCATAC GCGGCAGGACCAAGCAACAATGCACTGGGCCAGGTCTTGAAGGCTGTCCA ACGAAT nt 2103 and the corresponding amino acid sequence (SEQ ID NO:3): NCTSISGDLHILPVAFRGDSFTHTPPLDPQELDILKTVKEITGFLLIQAW PENRTDLHAFENLEIIRGRTKQQCTGPGLEGCPTN

[0019]Again an exchange of H against Q is observed (underlined amino acid).

[0020]These fragments were isolated, cloned and sequenced with standard procedures. The sequence of the 414 bp fragment shows a deletion of exon 12 to 14 of the EGFR and the sequence of the 256 bp fragment shows a deletion of exon 12 to 15. The variants were named EGFR c.EX12_--14del and EGFR c.EX12_--15del, respectively. The predicted amino acid sequence of variant EGFR c.EX12_--14del shows a frame shift in the open reading frame and a premature stop codon in exon 18 resulting in an EGFR protein with a truncated C-terminus. In contrast, the predicted amino acid sequence of variant EGFR c.EX12_--15del shows no frame shift resulting in an EGFR with a deletion of 194 amino acids.

[0021]Further, cancer cell lines were analyzed for the variants. Both variants could be detected in breast, colon, prostate and lung carcinoma cell lines. To determine that these new variants are tumors associated, the PCR was done with cDNA of blood from 22 healthy donors and 33 colon cancer patients isolated with AdnaTest ColonCancerSelect® (AdnaGen AG, Langenhagen, Germany) and Sensiscript Reverse Transcriptase® (Qiagen GmbH, Hilden, Germany). None of the healthy volunteers show the variants. 2 patients show the variant EGFR c.EX12_--14del and one of these patients shows the variant EGFR c.EX12_--15del additionally.

[0022]In the case of breast cancer, PCR was done with cDNA of blood from 23 healthy donors and 33 breast cancer patients with metastases (M1) isolated with AdnaTest BreastCancerSelect® (AdnaGen AG, Langenhagen, Germany) and Sensiscript Reverse Transcriptase (Qiagen GmbH, Hilden, Germany). None of the healthy volunteers show the variants. 2 patients show the variant EGFR c.EX12_--14del and 3 patients show the variant EGFR c.EX12_--15del.

[0023]To summarize, the present invention provides two new molecular tumor targets. This will allow an improved detection, stratification and therapy guidance of cancer patients and have an effect on future therapy decisions. Further, these two new EGFR variants (genes and mRNA) may be used as new tumor targets for therapeutical invention, e.g. with anti-bodies directed against the new EGFR protein variants or antisense RNA corresponding to the respective polynucleotides. Such antibodies and antisense RNA are also comprised within this invention. With the newly discovered EGFR variants, it is possible to further analyze the patients for therapy guidance and monitoring.

Sequence CWU 1

71256DNAHomo sapiens 1aaactgcacc tccatcagtg gcgatctcca catcctgccg gtggcattta ggggtgactc 60cttcacacat actcctcctc tggatccaca ggaactggat attctgaaaa ccgtaaagga 120aatcacaggg tttttgctga ttcaggcttg gcctgaaaac aggacggacc tccatgcctt 180tgagaaccta gaaatcatac gcggcaggac caagcaacaa tgcactgggc caggtcttga 240aggctgtcca acgaat 2562414DNAHomo sapiens 2aaactgcacc tccatcagtg gcgatctcca catcctgccg gtggcattta ggggtgactc 60cttcacacat actcctcctc tggatccaca ggaactggat attctgaaaa ccgtaaagga 120aatcacaggg tttttgctga ttcaggcttg gcctgaaaac aggacggacc tccatgcctt 180tgagaaccta gaaatcatac gcggcaggac caagcaacag gaccagacaa ctgtatccag 240tgtgcccact acattgacgg cccccactgc gtcaagacca gcccggcagg agtcatggga 300gaaaacaaca ccctggtctg gaagtacgca gacgccggcc atgtgtgcca cctgtgccat 360ccaaactgca cctacggatg cactgggcca ggtcttgaag gctgtccaac gaat 414385PRTHomo sapiens 3Asn Cys Thr Ser Ile Ser Gly Asp Leu His Ile Leu Pro Val Ala Phe1 5 10 15Arg Gly Asp Ser Phe Thr His Thr Pro Pro Leu Asp Pro Gln Glu Leu 20 25 30Asp Ile Leu Lys Thr Val Lys Glu Ile Thr Gly Phe Leu Leu Ile Gln 35 40 45Ala Trp Pro Glu Asn Arg Thr Asp Leu His Ala Phe Glu Asn Leu Glu 50 55 60Ile Ile Arg Gly Arg Thr Lys Gln Gln Cys Thr Gly Pro Gly Leu Glu65 70 75 80Gly Cys Pro Thr Asn 854202PRTHomo sapiens 4Asn Cys Thr Ser Ile Ser Gly Asp Leu His Ile Leu Pro Val Ala Phe1 5 10 15Arg Gly Asp Ser Phe Thr His Thr Pro Pro Leu Asp Pro Gln Glu Leu 20 25 30Asp Ile Leu Lys Thr Val Lys Glu Ile Thr Gly Phe Leu Leu Ile Gln 35 40 45Ala Trp Pro Glu Asn Arg Thr Asp Leu His Ala Phe Glu Asn Leu Glu 50 55 60Ile Ile Arg Gly Arg Thr Lys Gln Gln Asp Gln Thr Thr Val Ser Ser65 70 75 80Val Pro Thr Thr Leu Thr Ala Pro Thr Ala Ser Arg Pro Ala Arg Gln 85 90 95Gly Ser Trp Glu Lys Thr Thr Pro Trp Ser Gly Ser Thr Gln Thr Pro 100 105 110Ala Met Cys Ala Thr Cys Ala Ile Gln Thr Ala Pro Thr Asp Ala Leu 115 120 125Gly Gln Val Leu Lys Ala Val Gln Arg Met Gly Leu Arg Ser Arg Pro 130 135 140Ser Pro Leu Gly Trp Trp Gly Pro Ser Ser Cys Cys Trp Trp Trp Pro145 150 155 160Trp Gly Ser Ala Ser Ser Cys Glu Gly Ala Thr Ser Phe Gly Ser Ala 165 170 175Arg Cys Gly Gly Cys Cys Arg Arg Gly Ser Leu Trp Ser Leu Leu His 180 185 190Pro Val Glu Lys Leu Pro Thr Lys Leu Ser 195 2005137PRTHomo sapiens 5Asn Cys Thr Ser Ile Ser Gly Asp Leu His Ile Leu Pro Val Ala Phe1 5 10 15Arg Gly Asp Ser Phe Thr His Thr Pro Pro Leu Asp Pro Gln Glu Leu 20 25 30Asp Ile Leu Lys Thr Val Lys Glu Ile Thr Gly Phe Leu Leu Ile Gln 35 40 45Ala Trp Pro Glu Asn Arg Thr Asp Leu His Ala Phe Glu Asn Leu Glu 50 55 60Ile Ile Arg Gly Arg Thr Lys Gln Gln Asp Gln Thr Thr Val Ser Ser65 70 75 80Val Pro Thr Thr Leu Thr Ala Pro Thr Ala Ser Arg Pro Ala Arg Gln 85 90 95Glu Ser Trp Glu Lys Thr Thr Pro Trp Ser Gly Ser Thr Gln Thr Pro 100 105 110Ala Met Cys Ala Thr Cys Ala Ile Gln Thr Ala Pro Thr Asp Ala Leu 115 120 125Gly Gln Val Leu Lys Ala Val Gln Arg 130 135620DNAArtificialArtificial primer 6aaactgcacc tccatcagtg 20721DNAArtificialArtificial primer 7attcgttgga cagccttcaa g 21

Claims

1. A gene product largely homologous to the epithelial growth factor receptor (EGFR), characterized in that the amino acids encoded by either exons 12 to 14 or exons 12 to 15 are deleted.

2. The gene product according to claim 1, characterized in that it comprises the following amino acid sequence: TABLE-US-00003 (SEQ ID NO:5) NCTSISGDLHILPVAFRGDSFTHTPPLDPQELDILKTVKEITGFLLIQAW PENRTDLHAFENLEIIRGRTKQQDQTTVSSVPTTLTAPTASRPARQESWE KTTPWSGSTQTPAMCATCAIQTAPTDALGQVLKAVQR.

3. The gene product according to claim 2, characterized in that it comprises the following amino acid sequence: TABLE-US-00004 (SEQ ID NO:4) NCTSISGDLHILPVAFRGDSFTHTPPLDPQELDILKTVKEITGFLLIQAW PENRTDLHAFENLEIIRGRTKQQDQTTVSSVPTTLTAPTASRPARQESWE KTTPWSGSTQTPAMCATCAIQTAPTDALGQVLKAVQRMGLRSRPSPLGWW GPSSCCWWWPWGSASSCEGATSFGSARCGGCCRRGSLWSLLHPVEKLPT KLS.

4. The gene product according to claim 1, characterized in that it comprises the following amino acid sequence: TABLE-US-00005 (SEQ ID NO:3) NCTSISGDLHILPVAFRGDSFTHTPPLDPQELDILKTVKEITGFLLIQAW PENRTDLHAFENLEIIRGRTKQQCTGPGLEGCPTN.

5. The gene product according to claim 1, characterized in that with the exception of said deleted exons, its amino acid sequence is largely homologous or identical to the epithelial growth factor receptor in its native state (wild-type) or any of its known variants or derivatives.

6. The gene product according to claim 1, characterized in that with the exception of said deleted exons, its amino acid sequence is largely homologous or identical to variants vI, vII, vIII, vIII/Δ12-13, TDM/2-7, vIV, vV, TDM/18-25, TDM/18-26.

7. The gene product according to claim 1, characterized in that it comprises further amino acid sequence modifications, like such as caused by single nucleotide mutations or nucleotide deletions in its corresponding gene or mRNA etc.

8. A gene coding for a gene product according to claim 1.

9. A polynucleotide largely homologous or identical to the messenger ribonucleic acid (mRNA) coding for the epithelial growth factor receptor (EGFR), wherein the nucleotides coding for exons 12 to 14 or exons 12 to 15 of the epithelial growth factor receptor (EGFR) are deleted.

10. The polynucleotide according to claim 9, characterized in that it comprises a nucleotide sequence corresponding to the following cDNA-Sequence (SEQ ID NO:1): TABLE-US-00006 AAACTGCACCTCCATCAGTGGCGATCTCCACATCCTGCCGGTGGCATTTA GGGGTGACTCCTTCACACATACTCCTCCTCTGGATCCACAGGAACTGGAT ATTCTGAAAACCGTAAAGGAAATCACAGGGTTTTTGCTGATTCAGGCTTG GCCTGAAAACAGGACGGACCTCCATGCCTTTGAGAACCTAGAAATCATAC GCGGCAGGACCAAGCAACAGGACCAGACAACTGTATCCAGTGTGCCCACT ACATTGACGGCCCCCACTGCGTCAAGACCAGCCCGGCAGGAGTCATGGGA GAAAACAACACCCTGGTCTGGAAGTACGCAGACGCCGGCCATGTGTGCCA CCTGTGCCATCCAAACTGCACCTACGGATGCACTGGGCCAGGTCTTGAAG GCTGTCCAACGAAT.

11. The polynucleotide according to claim 9, characterized in that it comprises a nucleotide sequence corresponding to the following cDNA-Sequence (SEQ ID NO:2): TABLE-US-00007 AAACTGCACCTCCATCAGTGGCGATCTCCACATCCTGCCGGTGGCATTTA GGGGTGACTCCTTCACACATACTCCTCCTCTGGATCCACAGGAACTGGAT ATTCTGAAAACCGTAAAGGAAATCACAGGGTTTTTGCTGATTCAGGCTTG GCCTGAAAACAGGACGGACCTCCATGCCTTTGAGAACCTAGAAATCATAC GCGGCAGGACCAAGCAACAATGCACTGGGCCAGGTCTTGAAGGCTGTCCA ACGAAT.

12. The polynucleotide according to claim 9, characterized in that with the exception of the nucleotides coding for said deleted exons its nucleotide sequence is largely homologous or identical to the mRNA coding for the epithelial growth factor receptor in its native state (wild-type) or any of its known variants or derivatives.

13. The polynucleotide according to the claim 12, characterized in that with the exception of the nucleotides coding for said deleted exons its nucleotide sequence is largely homologous or identical to variants vI, vII, vIII, vIII/Δ12-13, TDM/2-7, vIV, vV, TDM/18-25, TDM/18-26.

14. The polynucleotide according to claim 9, characterized in that it comprises further nucleotide sequence modifications, like single nucleotide modifications, nucleotide deletions etc.

15. A method for detecting the presence of a tumor, comprising (a) analyzing a sample for the presence of a gene product, a gene and/or a polynucleotide that comprises or codes for gene product largely homologous to epithelial growth factor receptor (EGFR) that lacks either exons 12 to 14 or exons 12 to 15 or the corresponding amino acid sequences, and (b) deciding on the presence of a tumor if presence of a gene product or polynucleotide is detected.

16. The method according to claim 15, further comprising diagnosing, stratifying or guiding therapy of a tumor or after tumor surgery.

17-18. (canceled)

19. The method according to claim 17, characterized in that the tumor is an epithelial tumor, e.g. colon cancer, lung cancer, prostate cancer, breast cancer or other solid tumors.

20. A method for targeting a tumor cell with a therapeutic agent comprising (a) providing a composition a gene product, a gene and/or a polynucleotide that comprises or codes for gene product largely homologous to epithelial growth factor receptor (EGFR) that lacks either exons 12 to 14 or exons 12 to 15 or the corresponding amino acid sequences, wherein said gene product, gene or polynucleotide is linked to or further codes for said therapeutic agent, and (b) delivering said composition to said tumor cell.