FIELD OF THE INVENTION The present invention is directed to a mutation of the human epidermal growth factor receptor gene, as a marker for determining response to monoclonal antibody treatment.

BACKGROUND ART

Epidermal growth factor receptor gene (EGFR) is a transmembrane tyrosine- kinase receptor that belongs to the epidermal growth factor family of receptors (ErbB family), which includes four closely related receptor tyrosine kinases: EGFR (ErbB-1 ), HER2/c-neu (ErbB-2), Her 3 (ErbB-3) and Her 4 (ErbB-4). Upon ligand binding, EGFR activates intracellular signaling pathways, mainly the RAS-RAF-MEK-ERK cascade and the PI3KAkt pathway, that regulate key oncogenic events such as apoptosis, cell growth,

angiogenensis and metastasis. Aberrant activation or overexpression of EGFR has been reported in several types of cancer (i.e. Mendelsohn J, Baselga J et al., "Epidermal growth factor receptor targeting in cancer".

Semin Oncol - 2006, Vol. 33, pp.: 369-38). Mutations in EGFR gene have been described in lung cancer. Examples of such mutation are disclosed for instance in the document of Lynch TJ 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, Vol. 350, pp:2129-2139; or in

Paez JG et al., "EGFR mutations in lung cancer: correlation with clinical response to gefitinib therapy", Science-2004, Vol. 304, pp. :1497-500; or in Pao W et al., "EGF receptor gene mutations are common in lung cancers from "never smokers" and are associated with sensitivity of tumors to gefitinib and erlotinib". Proc Natl Acad Sci U S A-2004, Vol. 101 , pp.:13306-1331 1 .).

Metastasic colorectal cancer (mCRC) is the second leading cause of death from cancer in the Western Countries world. A therapy based on monoclonal antibodies (moAbs), e.g. cetuximab and panitumumab, which are directed against EGFR, provides significant survival benefit to patients with mCRC and are now standard components of therapy regimens for these patients, i.e. either alone or in combination with other antineoplasic drug(s). One of these moAbs, cetuximab (Erbitux) is also indicated for the treatment of patients with squamous cell carcinoma of the head and neck, also named head and neck cancer, in combination with platinum-based chemotherapy.

The moAbs bind to foreign antigens expressed on cancer cells. Once bound, the cancer cells are marked for destruction by the patient's immune system. In addition to targeting cancer cells, moAbs can be designed to act on other cell types and molecules necessary for tumor growth. For example, antibodies can neutralize growth factors and thereby inhibit tumor expansion. It is possible to create a moAb specific to almost any extracellular/cell surface target (such as cancer cells). In summary, moAbs can be used to destroy malignant tumor cells and prevent tumor growth by blocking specific cell receptors. Therapeutic moAbs cetuximab and panitumumab bind to EGFR and prevent the activation of intracellular signaling pathways driven by EGFR (i.e., the RAS-RAF-MEK-ERK cascade and PI3K-akt pathway).

Unfortunately, not all patients with mCRC respond to a therapy regimen comprising moAbs. The lack of response of a patient with mCRC to such a treatment could be primary, i.e. since the beginning of anti-EGFR moAb treatment; known as primary resistance. Moreover, all mCRC patients that initially respond to anti-EGFR moAbs invariably develop secondary

resistance, i.e. acquired resistance to anti-EGFR moAb. In both cases, the result is treatment failure. The mechanisms that contribute to the acquisition of such treatment resistance in mCRC patients is still not fully known. The same resistance to anti-EGFR moAb therapy (primary or secondary) is observed in patients with head and neck cancer. KRAS (also known as V-Ki-ras2 Kirsten rat sarcoma viral oncogene homolog) is an EGFR downstream effector, and a marker of primary resistance to anti- EGFR moAbs. KRAS has a significant impact on the optimization of treatment of mCRC patients. Forty per cent of colorectal tumors harbour a mutation in the KRAS gene and these patients do not benefit from anti-EGFR moAbs. In current clinical practice all mCRC patients who are being considered for anti- EGFR moAb therapy should undergo KRAS testing, and patients should be excluded from cetuximab or panitumumab therapy if a KRAS mutation is detected.

Nevertheless, a fraction of mCRC patients with wild-type KRAS tumors still do not benefit from anti-EGFR moAbs. The response rate to anti-EGFR moAbs in wild-type KRAS patients is approximately 60% when combined with

chemotherapy and less than 20% when administered alone in chemotherapy- refractory patients, as derived from Amado et al., "Wild-type KRAS is required for panitumumab efficacy in patients with metastasic colorectal cancer", J. Clin Oncol - 2008, Vol. 28, pp.: 1626-1634.

Activating mutations of other EGFR downstream genes such as BRAF

(serine/threonine-protein kinase B-Raf ) and PI3K (phosphatidylinositol 3- kinase), as well as loss of expression of PTEN (phosphatase and tensin homolog), and alterations in other EGFR regulatory proteins are being evaluated as potential candidates for response to anti-EGFR therapy with inconclusive results so far. Information regarding the association between the mutations in theses genes and the response to anti-EGFR therapy can be derived from the documents of De Roock et al., "Effects of KRAS, BRAF, NRAS, and PIK3CA mutations on the efficacy of cetuximab plus

chemotherapy in chemotherapy-refractory metastasic colorectal cancer: a retrospective consortium analysis", Lancet Oncol - 2010, Vol. 1 1 , pp.:753- 762; or in the document of Loupakis et al., "PTEN expression and KRAS mutations on primary tumors and metastases in the prediction of benefit of cetuximab plus irinotecan for patients with metastasic colorectal cancer", J Clin Oncol - 2009. Vol. 27, pp.:2622-2629.

The studies carried out so far to elucidate a potential role of the EGFR as a marker of response to anti-EGFR moAb are inconclusive. EGFR protein expression, as detected by immunohistochemistry, is not a reliable predictive marker of response to anti-EGFR moAbs. However, there is increasing evidence supporting EGFR gene copy number as a potential biomarker of response to anti-EGFR moAbs . Regarding the association of nucleotide changes in the EGFR gene with response to anti-EGFR moAbs-based therapy, the state of the art, in particular Gongalves et al. In "A polymorphism of EGFR extracellular domain is associated with progression free-survival in metastasic colorectal cancer patients receiving cetuximab-based treatment", BMC Cancer - 2008, Vol. 8, pp.: 169, describes a polymorphism in the extracellular portion of the EGFR gene, resulting in the amino acid

substitution R521 K associated with cetuximab benefit in mCRC patients. The polymorphism or single nucleotide polymorphism is the one identified as the variation CM942312 from January 201 1 , retrievable from the database

Ensembl (www.ensembl.org). It corresponds to the codon change at position 521 AGG-AAG in the mRNA sequence idetified as NM_005228 version 3, available on 26.06.201 1 from GenBank.

Additionally, also document WO2008/88860 discloses that patients with metastasic or non-metastasic gastrointestinal neoplasm or malignant tumour having the polymorphism R497K in the EGFR gene are likely to show responsiveness to single agent anti-EGFR moAb-based therapy (e.g.

cetuximab or panitumumab). This mutation is the same disclosed by

Gongalves et al. (supra), but being identified with the ancient designation. Finally, document WO20050854732 discloses the association of 12

polymorphisms in the regulatory region of the EGFR gene, which induce over- expression of the EGFR protein, with decreased efficacy of an EGFR- targeting therapeutic agent for the treatment of cancer in a patient. In summary, the results showed in the documents comprised in the state of the art are not only inconclusive, but also do not fully clarify the fraction of mCRC patients with wild-type KRAS tumors who still do not benefit from anti- EGFR mAb-based therapy. In view of the above, it is therefore necessary to identify additional predictive biomarkers of resistance to anti-EGFR moAb therapy in patients with mCRC.

SUMMARY OF THE INVENTION The inventors have identified a mutation in the extracellular domain of EGFR

(domain III) that correlates with resistance to the treatment with some moAbs used in the cancer therapy. In particular, the invention is based on the surprising identification of the serine by an arginine amino acid substitution in position 492 of the EGFR protein. Mutated protein has the amino acid sequence identified herein as SEQ ID NO: 10. Wild type protein has the amino acid sequence SEQ ID NO: 8. The mutation is known herein as S492R. Individuals with the mutation showed resistance to the treatment with cetuximab, thus rendering usefulness such a therapeutic approach, and making enforceable other drug strategies. Thus, a first aspect of the present invention relates to a peptide sequence comprising SEQ ID NO: 1 (TKIIRNRGE). This amino acid sequence is a fragment derived from the entire amino acid sequence codifying for EGFR, corresponding to the amino acid sequence SEQ ID NO: 8, wherein the serine at position 492 has been substituted by an arginine.

Advantageously, the mutated peptide comprising SEQ ID NO: 1 is still sensitive to anti-EGFR moAbs other than cetuximab being relevant for use in cancer therapy. In particular it is sensitive to moAbs useful in the therapy of metastasic colorectal cancer (mCRC) and chemotherapy to head and neck cancer.

In a second aspect the invention aims an oligonucleotide comprising a sequence coding for SEQ ID NO: 1 . A further aspect of the invention is a set of primers consisting of SEQ ID Nos: 3 (gggacctccggtcagaaaa) and 4 (cggtgacttactgcagctgttt).

This set of primers allows amplifying the genomic region comprising the portion of the EGFR coding region wherein the nucleotide changes resulting in the mutation of the present invention are located. They are thus related with the novel amino acidic mutation identified by the inventors.

In particular, the invention is based on the surprising identification of the serine by an arginine amino acid substitution in position 492 of the EGFR protein (mutation known herein as S492R). This amino acid change is the result of the nucleotide change C→A at position 1722 (also known herein as C1722A) of the mRNA variant 1 of the EGFR gene. The amino acid change of the present invention may also be the result of the nucleotide change A→G at nucleotide position 1720 (also known herein as A1720G) of the mRNA variant 1 of the EGFR gene. Finally, the amino acid mutation of the present invention can also be the result of each of the following nucleotide(s) changes in the mRNA variant 1 of the EGFR gene: C→G at position 1722 (also known herein as C1722G), A→C at position 1720 and C→T at position 1722 (also known herein as A1720C/C1722T), A→C at position 1720 (also known herein as A1720C); and A→C at position 1720 and C→G at position 1722 (also known herein as A1720C/C1722G).

Another aspect of the invention is an oligonucleotide consisting of SEQ ID NO: 5 (cacctctgtttcttataatt).

This oligonucleotide of SEQ ID NO: 5 is complementary to the mutated region of the EGFR coding region wherein the nucleotide changes resulting in the mutation of the present invention are located. Thus, it hybridizes with a fragment of the nucleotide sequence carrying the mutation. It allows detecting the nucleotide change C→A at position 1722 and/or 1720 disclosed above. As already indicated above, each of the above nucleotide changes refers to the mRNA, transcript variant 1 sequence of the EGFR gene (also known as ERBB1 , PIG61 , proto-oncogene c-ErbB-1 , avian erythroblastic leukemia viral (v-erb-b) oncogene homolog receptor tyrosine-protein kinase erbB-1 , or HER1 ). The sequence of the mRNA, transcript variant 1 , of the EGFR gene is that corresponding to SEQ ID NO: 7 (or GenBank accession number

NM_005228.3, available on 26.06.201 1 ) as well as any variant thereof, wherein said variant codes for the EGFR protein. The EGFR protein corresponds to SEQ ID NO: 8 (GenBank accession number NP_005219.2 version of 17.07.201 1 ) or any variant thereof that maintains the basic structure of the EGFR protein.

Also another aspect of the invention is an oligonucleotide consisting of SEQ ID NO: 6 (cacctctgttgcttataa). This oligonucleotide of SEQ ID NO: 6 is complementary to wild-type region of the EGFR coding region wherein the identified mutation is located (in case of a mutation exits).

Both oligonucleotides are suitable probes allowing detecting the presence or not of the mutation leating to the arginine by serine change.

Another aspect of the invention is a kit which comprises the oligonucleotide as defined before and corresponding to SEQ ID NO: 5.

Advantageously, this kit is a usable tool to detect the presence of the S492R mutation in an easy and fast way since it includes the probe complementary to the mutated EGRF coding region.

Also another aspect of the invention is the kit as defined above, for use in the prediction of the response of a subject to a therapy regimen comprising cetuximab and/or panitumumab.

Further, the invention also relates to an in vitro method of identifying the presence or absence of an arginine at position 492 of the amino acid sequence corresponding to SEQ ID NO: 8 in a sample taken from a subject, comprising determining the amino acid at position 492 of SEQ ID NO: 8 by means selected from the group consisting of genotype methods, and/or protein sequencing methods. The in vitro method of the invention allows identifying a peptide sequence comprising SEQ ID NO: 1 (TKIIRNRGE).

Finally, another aspect of the invention is a in vitro method of predicting the response of a subject therapy regimen comprising cetuximab and/or panitumumab, wherein the method comprises: i) determining the presence or absence of an arginine at position 492 of the amino acid sequence

corresponding to SEQ ID NO: 8 in a sample taken from the subject, as defined in the method disclosed above; ii) correlating the presence of the arginine identified in step i) with resistance of the subject to the therapy regimen comprising cetuximab, or correlating the absence of the arginine identified in step i) with response of the subject to therapy regimen comprising panitumumab. The put into practice of the in vitro method of predicting the response of a subject to a therapy regimen comprising cetuximab and/or panitumumab, implies the advantage of accommodating the more suitable therapy for the subject, and avoids wrong or not useful enough therapeutically approaches incurring waste time, which is an essential aspect for the subject and the success of the treatment, especially if the subject is affected with cancer.

BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 , related to Example 3, is a direct binding assay, showing interaction of cetuximab and panitumumab to interact with wild-type EGFR (wt EGFR) and S492R EGFR. WT-ECD-Fc means extracellular domain Fc (fragment crystallisable); S492R-ECD-Fc means extracellular domain Fc (fragment crystallisable); [Ab] (ng/ml) is the tested antibody concentration in nanograms per millilitre; and OD 450/620 nm is the optical density.

Fig. 2A is a structural modelling of the interaction between EGFR domain III and cetuximab, confirming the position of the mutation of the present invention (arginine at position 492 of the EGFR protein) at the interface of both molecules.

Fig. 2B, related to Example 3, is a Western blot analysis of total and phosphorylated EGFR (at Tyr1068; named herewith pEGFR) NIH3T3 cell lysates overexpressing wild-type EGFR (wt EGFR) and S492R EGFR mutant cultured in the presence of cetuximab or panitumumab. Tub means tubulin.

Fig. 2C, related to Example 3, is a Western blot analysis of total EGFR of lysates of NIH3T3 cells expressing wild-type EGFR (wt EGFR) and S492R EGFR, immunoted with cetuximab and panitumumab. E means empty; SN supernatant and IP immunoprecipitated; and I means input.

Fig. 2D, related to Example 3, is a Flow cytometry binding analysis of trypsinized NIH3T3 overexpressing wild-type EGFR (wt EGFR) and S492R

EGFR mutant incubated with cetuximab or panitumumab as primary antibodies and using a secondary antibody conjugated with phicoeritrin directed against human IgG. C means counts; FL2H denotes the maximal signal intensity in the second channel of fluorescence detection with a band pass of 585± 21 that is used to detect the phycoerythrin (PE) fluorescence; E means empty.

DETAILED DESCRIPTION OF THE INVENTION In general, the following words or phrases have the indicated definition when used in the description, examples and claims. The term "therapy regimen" as used in the state of the art and also herein refers to any therapy intended to prevent, slow, arrest or reverse the growth of a precancerous lesion, cancer or a cancer metastasis. It includes chemotherapy, radiation therapy, immunotherapy, monoclonal antibody therapy or other methods.

By "response" is to be understood any kind of improvement either clinical or non-clinical selected from, but not limited to, measurable reduction in tumour size or evidence of disease or disease progression, stable disease, increase or elongation of progression of free survival or reduction in toxicity.

"Progression free survival" indicates the length of time during and after treatment that the cancer does not grow. Progression free survival includes the amount of time patients have experienced a complete response or partial response, as well as the amount of time patients have experienced stable disease.

"A complete response" to a therapy defines patients with valuable but non- measurable disease, whose tumour and all evidence of disease disappeared.

"A partial response" to a therapy defines patients with anything less than complete response.

The expression "genotype methods" includes all those methodologies and processes suitable for determining the genotype or, which is the same for identifying the nucleotide in a given position. Examples of said methodologies encompass Sanger sequencing, pyrosequencing, allele-specific PCR, denaturing high pressure liquid chromatography (DHPLC), Allele Specific Primer Extension (ASPE), DNA biochips/microarrays and dynamic allele- specific hybridization (DASH).

For "protein sequencing methods" is to be understood any technique allowing to determine the amino acid sequence of a protein, as well as which conformation the protein adopts and the extent to which it is complexed with any non-peptide molecules. The determination of amino acid composition may be performed by hydrolysis or separation of the amino acids. Known technologies include the Sanger sequencing, Edman degradation and mass spectrometry.

As already explained above, the teachings according to the state of the art suggest, on the one hand, that mutations in the regulatory region of the EGFR gene that result in the over-expression of the corresponding protein are associated with decreased efficacy of an EGFR-targeting therapeutic agent for the treatment of cancer in a patient (cf. WO2005/854732); but also that a nucleotide change in the coding region of the EGFR gene is associated with responsiveness to single agent anti-EGFR mAb based therapy in patients with metastasic or non-metastasic gastrointestinal neoplasm or malignant tumour (cf. WO2008/88860). These are contradictory results that moreover have not been further confirmed, e.g. by analyzing, at least in vitro, the effect of the nucleotide change identified in the coding region of the EGFR gene, hence, affecting the EGFR protein. Consequently, it is not clear whether the nucleotide change identified in WO2008/88860 is a mutation causing the responsiveness, or on the contrary, another mutation in linkage

desequilibrium with it, and located in another gene, is causing the

responsiveness. In contrast with the findings disclosed in the state of the art, the present invention is based on a novel mutation in the coding region of the EGFR gene. The novel mutation of the present invention is useful to predict the response to moAb-based therapy of a patient with mCRC and/or with head and neck cancer (squamous cell carcinomas).

As already indicated above, each of the disclosed nucleotide changes lead to the substitution of a serine to an arginine at position 492 of the protein sequence corresponding to SEQ ID NO: 8. Serine 492 is located within the extracellular domain of EGFR (also known as domain III). Arginine is an amino acid with a bulky side chain, whereas serine is a polar amino acid. The present invention, hence, is based on the finding that the substitution of the amino acid located at position 492 of the EGFR protein (i.e. serine) by a bulky amino acid (e.g. arginine) interferes with the binding of the mAb cetuximab to EFGR. In other words, that the amino acidic change of the present invention located in the epitope of EGFR that binds to cetuximab specifically disrupts the cetuximab-EGFR interaction. Furthermore, and even more surprisingly, the amino acidic change of the present invention, which results in cetuximab resistance, does not affect the EGFR binding of another moAb, namely panitumumab, which is a moAb widely used for cancer therapy of metastasic colorectal cancer (mCRC).

Accordingly, this invention provides methods to select the appropriate therapy for patients suffering from mCRC, wherein the appropriate therapy comprises administration of the effective amount of cetuximab and/or panitumubab.

The isolated peptide comprising SEQ ID NO: 1 is the key product leading to the detection of a mutated form of EGFR protein of great interest in the field of cancer therapy. This mutated form of the protein is also detectable in the form of an oligonucleotide comprising a sequence coding for SEQ ID NO: 1 .

In a preferred embodiment the oligonucleotide coding for SEQ ID NO: 1 comprises a nucleic acid sequence selected from the group consisting of SEQ ID NO: 2, SEQ ID NO: 13, SEQ ID NO: 14, SEQ ID NO: 15, SEQ ID NO: 16 and SEQ ID NO: 17. All these sequences can be schematically

represented by the generic sequence defined in formula (I): ccaaaattataXYZaacagaggtga (I) wherein nucleotides XYZ are nucleotide combinations selected from the group consisting of: AGA, AGG, CGT, CGC, CGA and CGG.

The detection of the nucleotide changes in the EGFR gene disclosed in the present invention is preferably carried out with the following method: After amplification of the target DNA sequence corresponding to SEQ ID NO: 9, the amplicon is analyzed using a mutated probe (SEQ ID NO: 5) and a wild type probe (SEQ ID NO: 6). SEQ ID NO: 9 corresponds to GenBank accession number NG_007726, version 1 (NG_007726.1 ), available on 19.06.201 1 .

In a preferred embodiment the mutated probe is that corresponding to SEQ ID NO: 5, whereas the wild type probe is that corresponding to SEQ ID NO: 6, these probes are labelled in the 5' end with 6-carboxyfluorescein (FAM ) and the label VIC®, and with a Minor Groove Binder (MGB) that includes a non- fluorescent extintor of fluorescency at the 3' end. Schematically, the probes are as follows: wild type probe (labelled SEQ ID NO: 6):

VIC-5'-CACCTCTGTTGCTTATAA-3'-MGB mutation-specific probe (labelled SEQ ID NO: 5): FAM-5'-CACCTCTGTTTCTTATAATT-3'-MGB

The probes may also include other suitable labels allowing the specific detection of the same once hybridized with the target fragment of the EGFR gene. Examples of other labels include chemiluminescent labels,

radioisotopic labels, colorimetric labels, etc.

To the best of the applicant's knowledge, the specific portion of the EGFR genomic region wherein the nucleotide changes resulting in the mutation of the present invention are located has been of no interest in the state of the art, since no mutation with industrial applicability has been identified therein. Consequently, the state of the art does not disclose primers which hybridize and amplify said genomic region.

Therefore, the novel primers disclosed in the present invention specifically designed to amplify the genomic region comprising the portion of the EGFR coding region wherein the nucleotide changes resulting in the mutation of the present invention are located, are clearly linked with the novel amino acidic mutation identified by the inventors. The preferred primers used for the amplification are those corresponding to

SEQ ID NO: 3 and SEQ ID NO: 4. These primers are located, respectively, on nucleotides positions 146318-146297, and 146253-146271 of the genomic sequence of the EGFR gene, SEQ ID NO: 9 (corresponding to GenBank accession number NG_007726, version 1 (NG_007726.1 ), available on 19.06.201 1 ).

These methods are not limited by the technique that is used to identify the nucleotide changes of the present invention. Any technique known in the art suitable for the detection of the nucleotide changes of interest can be used in the methods of the present invention. Suitable techniques include but are not limited to amplification reactions. In a preferred embodiment, the amplification technique used is the polymerase chain reaction. In a more preferred embodiment, the PCR is a real time PCR. The amplification conditions can be adjusted as necessary, and easily, by one of skill in the art; an example of amplification conditions suitable for the methods of the present invention are provided in the example section.

Alternatively, the mutation of interest is detected by a sequencing reaction, such as the Sanger sequencing, which gives the information of the codon or amino acid present in the region of interest. Examples of suitable methods to detect the mutation encompass Sanger sequencing, pyrosequencing, allele- specific PCR, denaturing high pressure liquid chromatography (DHPLC).

Thus, the in vitro method of the invention for identifying the presence or absence of an arginine at position 492 of the amino acid sequence

corresponding to SEQ ID NO: 8 in a sample taken from a subject, is preferably carried out using genotyping methodologies. In another preferred embodiment the presence or absence of an arginine at position 492 of the amino acid sequence corresponding to SEQ ID NO: 8 is carried out at the protein level by means of protein sequencing methods. In a preferred embodiment of the kit according to the invention, the kit comprises the oligonucleotide consisting of SEQ ID NO: 5

(cacctctgtttcttataatt) and further the oligonudetide consisting of SEQ ID NO: 6 (cacctctgttgcttataa). In a more preferred embodiment the kit further comprises reagents for detecting mutations in the KRAS and/or PIK3CA and/or BRAF gene.

As above exposed, these genes are related with the resistance to the treatment of cancer (namely mCRC) with moAb (KRAS), or they are codifying for EGFR regulatory proteins (PIK3CA and BRAF) evaluated as potential candidates to drug target therapy. So then, in a more more preferred embodiment, the kit also includes tools and means (reagents) to detect the mutations in KRAS selected from the group consisting of G12A; G12C; G12D; G12R; G12S; G12V; G13A; G13C, G13D; G13V as defined by Karapetis et al., "K-ras Mutations and Benefit from Cetuximab in Advanced Colorectal Cancer", The New England Journal of Medicine - 2008, Vol. 359, pp.: 1757-1765. All these mutations are placed on codons 12 and 13 of the protein sequence of K-ras identified with the

GenBank accession number NP_004976.2 from 24.07.201 1 (named GTPase KRas isoform b precursor) and NP_203524.1 from 24.07.201 1 (named GTPase KRas isoform a precursor

In another preferred embodiment the kit also includes tools and means (reagents) to detect mutations in exons 9 and 20 of the PIK3CA gene that codifies for the PIK3CA protein with the GenBank accession number

NP_006209.2 from 17.07.201 1 ; and/or the V600E mutation placed on codon 600 of the protein sequence of BRAF identified with the GenBank accession number NP_004324.2 from 24.07.201 1 .

The kit of the invention, for use in the prediction of the response of a subject to a therapy regimen comprising cetuximab and/or panitumumab, is in a preferred embodiment for predicting the response of a subject to a cancer selected from the group consisting of metastasic colorectal cancer and head and neck cancer. Preferably the kit is for the prediction of response in case of metastasic colorectal cancer.

In a preferred embodiment of the in vitro method of predicting the response of a subject to a therapy regimen comprising cetuximab and/or panitumumab, the subject has already been treated with a therapy regimen comprising cetuximab.

In another preferred embodiment, the in vitro method of prediction is applicable to a subject affected with cancer. Preferably the cancer is selected from the group consisting of metastasic colorectal cancer and head and neck cancer.

The in vitro method of predicting the response of a subject to a therapy regimen comprising cetuximab and/or panitumumab is especially suitable for subjects affected with cancer. In particular preferred cancers are selected from the group consisting of metastasic colorectal cancer and head and neck cancer. The invention further provides methods of treating subjects affected with cancer, preferably mCRC or head and neck cancer, comprising: i)

determining the presence of the S492R EGFR mutant of the present invention; and ii) administering to said subject an effective amount of cetuximab, or a composition thereof, if the mutation is absent, or

panitumumab, or a composition thereof, if the mutation is present.

The anti-EGFR moAbs, cetuximab and panitumumab can be administered alone, as a composition, or in a therapy regimen including other compounds, such as chemotherapeutic drugs.

The anti-EGFR moAbs, cetuximab and panitumumab, or compositions thereof, are administered or delivered in an amount effective to treat the cancer (mCRC or head and neck) and with any suitable formulation, e.g. including a pharmaceutically acceptable carrier. The formulation can further comprise one or more preservatives and/or stabilizers.

The in vitro method for predicting the response of a subject to a therapy regimen comprising cetuximab and/or panitumumab, is carried out in a sample comprising the tumour, in which the nucleotide changes in the EGFR gene of the present invention can be detected. In cases of mCRC, the sample can be used directly as obtained from the source or following a pre-treatment of the sample. The sample may additionally comprise normal tissue adjacent to said tumour. Accordingly, in case of mCRC the sample is selected from a primary colorectal cancer biopsy or a biopsy of a metastasis thereof. In other words, the sample may be a biopsy from colorectal cancer samples, including primary tumors and metastases. In a preferred embodiment, the metastasis is in the liver tissue.

The subject includes any mammal, including, but not limited to, a human or non-human mammal. Preferably the subject is a human.

Patients having the S492R mutation of the present invention are likely to show response to a therapy regimen not comprising cetuximab as measured by any suitable clinical or sub-clinical increase or elongation in progression free survival. In a preferred embodiment the therapy regimen is cetuximab alone or in combination with a chemotherapy regimen based on irinotecan, oxaliplatin and/or 5-fluorouracil (5-FU or 5FU). In a preferred embodiment the therapy regimen is panitumumab alone or in combination with a chemotherapy regimen based on irinotecan, oxaliplatin and/or 5-fluorouracil.

Throughout the description and claims the word "comprise" and variations of the word, are not intended to exclude other technical features, additives, components, or steps. Furthermore the word "comprise" and its variations encompasses the expression "consisting of. Additional objects, advantages and features of the invention will become apparent to those skilled in the art upon examination of the description or may be learned by practice of the invention. The following examples and drawings are provided by way of illustration, and they are not intended to be limiting of the present invention. Furthermore, the present invention covers all possible combinations of particular and preferred embodiments described herein.

EXAMPLES

Example 1. Tumor samples and patients

Tumor specimens were obtained during diagnosis or from surgical

procedures on mCRC patients. Biopsy was obtained from the most accessible malignant lesion (either primary tumor or metastasis). When necessary, biopsy of tumoral lesions from patients that demonstrated tumor regrowth (disease progression) after initial response to cetuximab- based therapy was collected. Specimens from matched normal tissue were obtained as control. DNA extraction and mutational analysis of KRAS (codons 12 and 13), BRAF (V600E) were performed as previously described in Mutational analysis in the doucment by Montagut et al., "Mitogen-activated protein kinase phosphatase- 1 (MKP-1 ) impairs the response to anti-epidermal growth factor receptor (EGFR) antibody cetuximab in metastasic colorectal cancer patients", Br. J Cancer - 2010, Vol. 102, pp.: 1 137-1 144. Mutational analysis of PIK3CA was performed as with the DxS PI3K Mutation Test Kit (DxS, Manchester, UK), as disclosed in the Procedures of De Roock et al., "Effects of KRAS, BRAF, NRAS, and PIK3CA mutations on the efficacy of cetuximab plus chemotherapy in chemotherapy-refractory metastasic colorectal cancer: a retrospective consortium analysis", Lancet Oncol - 2010, Vol. 1 1 , pp.:753-762. by the

Amplification of EGFR was assessed by fluorescent in situ hybridization (FISH) using the LSI EGFR/CEP7 probe (Abbott Molecular Inc., Des Plaines, IL), as previously described in Pao et al., "Acquired resistance of lung adenocarcinomas to gefitinib or erlotinib is associated with a second mutation in the EGFR kinase domain", PloS Med 2005; 2:e73.

Analysis of EGFR S492R was performed by direct sequencing. Briefly, the region of EGFR exon 12 containing the mutated region were amplified with primers 5'-TTGCAGTCGTCAGCCTGAAC-3' (direct primer or SEQ ID NO: 1 1 ) and 5'-TTAAATGGGAATAGCCCTTCAATATT-3' (reverse primer or SEQ ID NO: 12) in an Applied Biosytems Veriti Thermalcycler with the following conditions: 95°C for 10 minutes ; 40 cycles of 95°C, 1 minute, 60°C, V 30" and 72°C 1 minute; and a final extension of 10 minutes at 72°C. Sequencing was performed with BigDye v3.1 (Applied Biosystems, Foster City, CA) following the manufacturer ^' s instructions and analysed on a 3500Dx Genetic Analyzer (Applied Biosystems). The sequence data files were analyzed using SeqScape software (Applied Biosystems) and all mutations were confirmed with an independent PCR. Real time monitoring of PCR amplification of DNAs was done with Taqman Universal master mix (Applied Biosystems) using 37.5 nM nM of each probe and 37.5 nM of each primer, in a ABI Prism 7500 FAST (Applied Biosystems). All determinations were performed in duplicate to minimise intra-assay variations. Example 2.The S492R EGFR mutation and primary resistance to cetuximab Cetuximab (Erbitux) and Panitumumab (Vectibix) were obtained from Hospital del Mar's Pharmacy, both monoclonal antibodies were ready to use. Gefitinib was obtained from Selleck Chemicals (Houston, TX, USA) and was dissolved in DMSO and aliquoted and stored at -20°C. Purified EGF recombinant protein was purchased from Calbiochem (San Diego, CA, USA) and was dissolved in PBS 0,1 % BSA , aliquoted and stored at -20°C.

The EGFR extracellular domain was sequenced and KRAS and BRAF mutational status in primary tumor specimens was analyzed from 83 metastasic colorectal cancer patients prior to administration of cetuximab- based therapy. As a group, these patients had been heavily pre-treated with other therapy regimen before receiving cetuximab. In eighty-two per cent of the patients, cetuximab was given in combination with irinotecan.

Significantly, a C→A nucleotide change at position corresponding to nucleotide 1722 of SEQ ID NO: 7, resulting in the S492R amino acid change in the corresponding EGFR protein (SEQ ID NO: 8) was detected in the specimens of two patients. Both tumors were KRAS, BRAF and PIK3CA wild- type and did not display EGFR gene amplification. The S492R EGFR mutation was not detected in matched normal tissue available for said patients.

Example 3. Presence of S492R EGFR mutation and resistance to cetuximab To establish whether the S492R EGFR mutation of the invention was responsible for the observed resistance to cetuximab, full-length wild-type EGFR and the S492R EGFR mutation was ectopically expressed in cultured NIH3T3 mouse embryonic fibroblast cell line that lack detectable endogenous EGFR expression.

EGFR was stimulated with its natural ligand EGF in the presence of cetuximab or panitumumab in transfected cells. In wild-type EGFR cells, both cetuximab and panitumumab inhibited EGFR activation, whereas in cells carrying the S492R mutation, panitumumab, but not cetuximab, effectively blocked EGF-induced EGFR activation (FIG. 2B).

These conclusions were derived from the assay in FIG. 2B, which is a Western blot analysis of total and phosphorylated EGFR (at Tyr1068;

herewith named pEGFR) NIH3T3 cell lysates overexpressing wild-type EGFR (wt EGFR) and S492R EGFR mutant cultured in the presence of cetuximab or panitumumab. NIH3T3 cells overexpressing wild-type EGFR (wt EGFR) and S492R EGFR mutant were cultured in the presence of cetuximab or panitumumab (10pg/nnl), after 2h cells were stimulated with EGF 10 g/mL for 15 minutes. Cell lysates were subjected to Western blot analysis of total and phosphorylated EGFR (Tyr1068 from SEQ ID NO: 8) to determine the activation of the receptor. Cetuximab was not able to revert ligand-induced activation in S492R EGFR mutant cells, as observed in the band

corresponding to the pEGFR when the assay was performed with this moAb.

To collect lysates, cells where washed with PBS and scraped in lysis buffer Nonidet P-40 buffer (Tris-HCL (pH = 7.4) 50mM, NaCI 150mM, 1 % NP40, EDTA 5 mM, NaF 5 mM, Na3VO4 2 mM, PMSF 1 mM, Leupeptin 5 pg/mL and Aprotinin 5 g/mL). After shaking for 30 min at 4°C, the samples were centrifuged at 13200 rpm for 30 min and the supernatant was aliquoted and stored at -20°C until use. Samples (30 g/lane) were subjected to SDS-page and transferred to nylon membranes. Western blotting was carried out according to standard procedures using horseradish peroxidase-conjugated secondary antibodies for signal detection. Target proteins were visualized after enhanced chemiluminescence treatment of membranes and subsequent exposure to X-ray film. The following antibodies were purchased from the manufacturers listed bellow: phospho EGFR (Y1068 or Tyr1068), EGFR, were obtained from Cell Signalling Technology (Beverly, MA, USA).

Moreover, flow cytometry as well as biochemical binding assays and immunoprecipitation showed that in cells expressing wild-type EGFR, both cetuximab and panitumumab could bind EGFR; however, in cells expressing S492R EGFR, panitumumab was able to bind to EGFR whereas cetuximab-

EGFR binding was not detected (see FIG. 1 , FIGs. 2C and 2D).

FIG. 1 shows the ability of cetuximab and panitumumab to interact with wild- type EGFR and S492R EGFR mutant in vitro by direct binding assay.

As above indicated, this assay was performed to further verify that the S492R EGFR directly impacted binding to cetuximab. The in vitro biochemical binding studies were performed using purified recombinant forms of the extracellular domain (EC) of wild type EGFR and the S492R mutant (FIG. 1 ).

The competitive binding assay was performed as follows: Anti-EGFR Ab binding to wild-type (WT) and mutant extracellular domain (ECD) of EFGR was compared in a competitive sandwich ELISA. Recombinant EGFR ECD human Fc fusion protein (WT, WT-ECD-Fc; or mutant, S492R-ECD-FC in FIG. 1 ) was immobilized onto a plastic surface overnight. The plate was then blocked with PBS containing BSA. Anti-EGFR Abs and a negative control hulgG Ab were serially diluted and mixed with an equal volume of biotin- labeled panitumumab or cetuximab at fixed concentration, the mixture was added onto the plate. The sample was incubated for 2 hours. The plate was washed and streptavidin-horseradish peroxidase (SA-HRP) conjugate was added as detection. The substrate tetramethyl benzidine (TMB) was added to the plate, and the reaction was stopped with acid. The plate was read at two wave- lenght values (OD 450/620 nm) and the Ab competitive binding results from WT and mutant ECD were compared.

Consistent with the cell-based assays, the biochemical binding studies confirmed that the S492R EGFR mutant is selectively defective for binding to cetuximab, but not to panitumumab. No detection of biotin-labeled cetuximab is observed at any Anti-EGFR Abs concentration in the experimental with S492R mutant, which means that no binding exists. This was also concluded from the results of a immunoprecipitation assay

(FIG. 2C) of the cell lysates from NIH3T3 expressing wild-type EGFR (wt EGFR) and S492R EGFR after being immunoted with 10 g/ml cetuximab and panitumumab, wherein non-specific IgG was used as negative control. As shown in FIG. 2C, the Western blot analysis of total EGFR confirmed that cetuximab was not able to bind to and precipitate S492R EGFR mutant. The input (I) and supernatant (SN) fractions of the precipitates were used as controls to confirm the presence of EGFR in the cell lysates.

Finally, the above results were also confirmed by flow cytometry. FIG. 2D shows that while cetuximab and panitumumab were able to interact with 60 % of cells expressing wild-type EGFR (wt EGFR), only panitumumab was able to bind to cells expressing the S492R EGFR mutation. Trypsinized NIH3T3 cells overexpressing wild-type EGFR and S492R EGFR mutant were incubated with 1 g/ml of cetuximab or panitumumab as primary antibodies. The binding was analyzed by flow cytometry using a secondary antibody conjugated with phicoeritrin directed against human IgG. NIH3T3 cells expressing the empty vector (E) were used as a negative control. The histograms show the percentage of cells detected by both antibodies.

Flow citometry was performed as follows: For cell cycle distribution analysis, cells were grown and treated with cetuximab for 24 h, 48 h and 72 h. After treatment, cells were harvested by trypsinization, washed twice with cold PBS and fixed with 70 % ethanol overnight. Ethanol was removed by washing the cells twice with cold PBS. Cells were stained for DNA with PBS containing 50 g/ml propidium iodide and 100 g/ml of RNAse at least during 48 h at 4 °C protected from light. Cell cycle distribution was measured using FACScalibur flow cytometer (Beckton Dickinson). To measure cetuximab and panitumumab binding to EGFR, cells were harvested by trypsinization and washed twice with PBS. Cells were incubated with Fc blocking reagent (MACS®) for 15 minutes on ice to block unspecific Fc binding of immonuglobulins. Cells were washed and incubated with the monoclonal antibodies to detect EGFR binding for 30 minutes on ice. A goat anti-human IgGy Phicoeritrin conjugated (Invitrogen) was used as a secondary antibody. EGFR binding was analyzed using the FACScan flow Cytometer (Beckton Dickinson).

Example 4. Detection of the S492R EGFR mutation in a patient with colorectal cancer demonstrating acquired (or secondary treatment) resistance to cetuximab To assess the clinical relevance of this mutation as a mechanism of acquired resistance to cetuximab, it was examined whether the S492R EGFR mutation could be found in patients with metastasic colorectal cancer who experienced disease progression following an initial response to cetuximab. Paired tumor samples from 10 patients before receiving cetuximab therapy and after failure to treatment (post-treatment specimen) were analyzed. All pre-treatment samples were from the primary colon tumor except in one case where the specimen was from a liver lesion, which was a metastasized tissue from a primary mCRC. Post-treatment samples were from liver metastasis obtained by percutaneous biopsy with ultrasound guidance. Most patients had previously received at least one line of chemotherapy for metastasic disease and cetuximab was administered together with irinotecan or oxaliplatin in all cases.

The mutational status of the extracellular domain region of the EGFR protein as well as that of KRAS, BRAF and PIK3CA were assessed by DNA sequence analysis (as detailed above). EGFR gene copy number was also studied by FISH (as detailed in Example 1 )

All pre-treatment biopsies were wild-type for EGFR, KRAS, BRAF and PIK3CA. The post-cetuximab treated tumor samples did not harbour any known KRAS, BRAF or PIK3CA mutations; however, the S492R mutation was identified in two patients. Notably, the observed mutation in one patient was associated with nucleotide substitution A→C change at nucleotide 1720 of SEQ ID NO: 7, which also results in a serine to arginine substitution at amino acid 492 of the EGFR protein (SEQ ID NO: 8). Sequencing of normal cells from the patient showed only the wild-type sequence, indicating that the

S492R mutation was a somatic mutation. The observed mutation in the other patient was the same as in the in vitro studies.

One of the two patients carrying the S492R mutation was treated with cetuximab (400 mg/m ² initial dose followed by 250 mg/m ² /week thereafter) plus oxaliplatin 85 mg/m ² on day 1 , plus leucovorin 200 mg/m ² and

fluorouracil as a 400 mg/m ² bolus followed by a 600 mg/m ² infusion during 22 hours on days 1 and 2. Three months after onset of treatment, a computed tomographic (CT) scan showed a partial response according to the response evaluation criteria in solid tumors (RECIST) (Eisenhauer et ai, "New response evaluation criteria in solid tumours: revised RECIST guideline (version 1 .1 )", Eur J Cancer 2009, Vol. 45(2):228-247). After 10 months of treatment, however, hepatic lesions exhibited frank progression and new liver lesions appeared. Cetuximab treatment was discontinued and a biopsy from pre-existing liver lesion was then obtained for molecular analysis, revealing the S492R EGFR mutation. The patient was then treated with irinotecan- based chemotherapy but did not respond. Therapy with single agent panitumumab 6mg/Kg every 2 weeks was then initiated, and after two months of treatment, a CT scan showed a reduction in all liver lesions greater than 50%. REFERENCES CITED IN THE APPLICATION

Mendelsohn J, Baselga J et al., "Epidermal growth factor receptor targeting in cancer". Semin Oncol - 2006, Vol. 33, pp.: 369-38 Gongalves et al., "A polymorphism of the EGFR extracellular domain is associated with progression free-survival in metastasic colorectal cancer pateints receiving cetuximab-based treatment", BMC Cancer 2008, Vol 8:169.

Eisenhauer et al., "New response evaluation criteria in solid tumours: revised RECIST guideline (version 1 .1 )", Eur J Cancer 2009. Vol. 45(2):228-247.

De Roock et al., "Effects of KRAS, BRAF, NRAS, and PIK3CA mutations on the efficacy of cetuximab plus chemotherapy in chemotherapy-refractory metastasic colorectal cancer: a retrospective consortium analysis", Lancet Oncol - 2010, Vol. 1 1 , pp. :753-762

Loupakis et al., "PTEN expression and KRAS mutations on primary tumors and metastases in the prediction of benefit of cetuximab plus irinotecan for patients with metastasic colorectal cancer", J Clin Oncol - 2009, Vol. 27, pp.:2622-2629.

Karapetis et al., "K-ras Mutations and Benefit from Cetuximab in Advanced Colorectal Cancer", The New England Journal of Medicine - 2008, Vol. 359, pp.: 1757-1765.

Montagut et al., "Mitogen-activated protein kinase phosphatase-1 (MKP-1 ) impairs the response to anti-epidermal growth factor receptor (EGFR) antibody cetuximab in metastasic colorectal cancer patients", Br. J Cancer - 2010, Vol. 102, pp.: 1 137-1 144.

Amado et al., "Wild-type KRAS is required for panitumumab efficacy in patients with metastasic colorectal cancer", J. Clin Oncol - 2008, Vol. 28, pp.: 1626-1634

Pao et al., "Acquired resistance of lung adenocarcinomas to gefitinib or erlotinib is associated with a second mutation in the EGFR kinase domain", PloS Med 2005; 2:e73

Lynch TJ 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, Vol. 350, pp:2129-2139.

Paez JG et al., "EGFR mutations in lung cancer: correlation with clinical response to gefitinib therapy", Science-2004, Vol. 304, pp. :1497-500.

Pao W et al., "EGF receptor gene mutations are common in lung cancers from "never smokers" and are associated with sensitivity of tumors to gefitinib and erlotinib". Proc Natl Acad Sci U S A-2004, Vol. 101 , pp.:13306-1331 1