JP2005224238A - Prognosis diagnosing marker for predicting treatment response and/or survival time of patient suffering from mammary cell growth disorder - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a method for conducting diagnosis of prognosis and/or prediction of result of estrogenic treatment of a patient suffering from mammary cell growth disorder, especially, mastocarcinoma. <P>SOLUTION: This method determines an expression level of a gene PITX2 of a mammary cell, genetic variation of a genomic DNA related to the PITX2, or epigenetic variation thereof, wherein a specific base sequence derived from breast cancer, an oligonucleotide, and an antibody are used. <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

  The present invention relates to a cell proliferation disorder of breast tissue characterized by determining the expression level of gene PITX2, or genetic or epigenetic changes in genomic DNA associated with gene PITX2 and / or its regulatory or promoter regions And / or methods for predicting the survival and / or therapeutic response of patients diagnosed with. The invention also relates to nucleic acid sequences, oligonucleotides, antibodies that can be used in the described methods.

(breast cancer)
In European and American women, breast cancer is the most frequently diagnosed cancer and the second leading cause of cancer death. In women aged 40-55 years, breast cancer is the leading cause of death (Greenlee et al., 2000). In 2002, there were 204,000 new cases of breast cancer in the United States and comparable numbers in Europe.

  Breast cancer is defined as the uncontrolled growth of cells within breast tissue. The breast consists of 15-20 leaflets connected by a duct. Cancer most commonly occurs in the ducts, but is also found in the leaflets that have a rare type of cancer called inflammatory breast cancer. It is recognized by those skilled in the art that there is a continuing need to improve methods for early detection, classification, and treatment of breast cancer. Breast cancer is inherently difficult to classify and detect against the detection of the cervix and some other common cancers such as the skin.

(Treatment of breast cancer)
The first stage of all treatment is an assessment of the patient's condition relative to a defined category of disease. However, the value criteria for such systems are inherently dependent on the nature of the classification. Breast cancer is graded by its size, location, and the occurrence of metastases. Treatment methods include the use of surgery, radiation therapy, chemotherapy, and endocrine therapy, which is also used as an adjunct therapy to surgery. In general, more aggressive diseases are treated with more aggressive therapies.

  Most of the early cancers are operable, that is, the tumor can be completely removed by surgery, but about 1/3 of patients with lymph node negative disease and those with lymph node positive disease About 50-60% develop metastases during the follow-up period.

  Based on this finding, systemic adjuvant therapy was introduced in both node-positive and node-negative breast cancers. Systemic adjuvant therapy, given after surgical removal of the tumor, has been shown to significantly reduce the risk of recurrence. Endocrine therapy, which is also called hormone therapy (for hormone receptor positive tumors), different chemotherapy regimens, and antibody treatments based on new drugs such as Herceptin (an antibody against epidermal growth factor receptor): Types of adjuvant therapy are available.

  The growth of the majority of breast cancers (approximately 70-80%) depends on the presence of estrogens. Thus, one important target for adjuvant therapy is estrogen removal (eg, by ovariectomy), blocking of receptors by competitors (eg, tamoxifen), or inhibition of androgen conversion to estrogen (eg, aromatase inhibitors) ) Block estrogen synthesis or block its action on tumor cells. This type of treatment is called "endocrine treatment". Endocrine therapy is believed to be effective only in tumors that express hormone receptors (estrogen receptor (ER) and / or progesterone receptor (PR)). Recently, the vast majority of women with hormone receptor positive breast cancer have accepted several forms of endocrine therapy, regardless of lymph node status. The drug most frequently used in this scenario is tamoxifen.

  However, even in hormone receptor positive patients, not all patients will benefit from endocrine therapy. Adjuvant endocrine therapy reduces mortality by 22%, while in the case of metastasis (progressive), the response rate to endocrine therapy is 50-60%.

  Because tamoxifen has relatively few side effects, treatment can be justified even for low-benefit patients. However, these patients may require additional, more aggressive adjuvant therapy. Even in the earliest and least aggressive tumors, such as lymph node negative, hormone receptor positive tumors, approximately 21% of patients 10 years after the first diagnosis if patients receive only tamoxifen monotherapy as an adjunct therapy (Lancet. 1998 May 16; 351 (9114): 1451-67. Tamoxifen for early breast cancer: an overview of the randomized trials. Early Breast Cancer Trialists' Collaborative Group.). Similarly, a significant number of patients with hormone receptor negative disease are adequately treated with surgery and, in some cases, radiation therapy alone, while other patients require additional chemotherapy.

  Several cytotoxic regimens have been shown to be effective in reducing the risk of breast cancer recurrence (Mansour et al., 1998). According to recent treatment guidelines, the majority of node-positive patients have received adjuvant chemotherapy in both the United States and Europe because of the risk of recurrence. Nonetheless, not all patients do not relapse, and none even recur even without chemotherapy, but some patients are still receiving chemotherapy due to recently used criteria. In hormone receptor positive patients, chemotherapy is usually given prior to endocrine therapy, whereas hormone receptor negative patients receive only chemotherapy.

  The situation for node-negative patients is particularly complex. In the US, cytotoxic chemotherapy is recommended for node-negative patients if the tumor is larger than 1 cm. In Europe, if one or more risk factors are present, such as tumors larger than 2 cm, negative hormone receptor status, or three tumor staging, or age under 35 years, chemotherapy is Considered for node-negative cases. In general, there is a tendency to select premenopausal women for additional chemotherapy, whereas for postmenopausal women, chemotherapy is often omitted. Compared to endocrine therapy, especially tamoxifen or aromatase inhibitors, chemotherapy is more toxic, such as short-term side effects such as nausea, vomiting, bone marrow loss, and increased cardiotoxicity and secondary cancer risk Has a long-term effect.

(Necessity that has long been felt in the prior art)
Currently, it is not clear which breast cancer patients should be selected for more aggressive therapy and which breast cancer patients will recover without additional aggressive treatment, and the selection of appropriate patients The great need for is that clinicians agree. Choosing the right patient for adjunct treatment and the difficulty of choosing the right adjunct treatment, and lack of appropriate criteria, are much less frequent than chemotherapy recommended based on data from the New Mexico Tumor registry It has also been shown by a recent study that revealed its use in (Du et al., 2003). This study provided substantial evidence that there is a need to successfully select patients for chemotherapy or other more aggressive forms of breast cancer therapy.

(Prior art of gene PITX2)
PITX2 (also known as PTX2, RS, RGS, ARP1, Brx1, IDG2, IGDS, IHG2, RIEG, IGDS2, IRID2, Otlx2, RIEG1, MGC20144) is a transcription factor within the paired-like class of homeodomain factors It is known to belong to the PTX subfamily of PTX1, PTX2, and PTX3 defining new families. The gene PITX2 (according to NM_153426) encodes a paired-like homeodomain transcription factor 2, which is known to be expressed during the development of anterior structures such as the eyes, teeth, and anterior pituitary gland.

Toyota et al. (2001) (Blood 97 : p 2823-9.) (Non-patent document 1) discovered hypermethylation of the PITX2 gene in most acute myeloid leukemias. Furthermore, in this study, hypermethylation of PITX2 is positively correlated with ER gene methylation and decreased expression levels. Means for analyzing the methylation pattern of the PITX2 gene have also been described in numerous patent applications (respectively, WO 02/077272 relates to the use of methylation markers to distinguish between AML and ALL). WO 01/19845 (Patent Document 2) relates to several differently methylated sequences useful for the diagnosis of several cell proliferation disorders, WO 02/00927 and WO 01/092565 (Patent Documents 3 and 4). ) Each relating to the use of methylation markers in the diagnosis of diseases associated with the developmental gene or with DNA transcription).

  Loss of heterozygosity on chromosome 4 (sometimes referred to as “LOH” below) is a known characteristic of many types of tumors. Shivapurkar et al [Cancer Research 59, 3576-3580, August 1, 1999] (Non-Patent Document 2) observed loss of heterozygosity in multiple regions of chromosome 4 in breast cancer samples and cell lines. The deletion at 4q25-26 was present in 67% of the samples analyzed. However, the region to be analyzed (between markers D4S1586 and D4S175) was not mapped to the PITX2 gene, and no effect on PITX2 expression was obtained. Furthermore, the studies performed did not show suitability of genes or region loci for prognostic applications.

  Although methylation of PITX2 has been linked to diseases such as development, transcription, and cancer, a role in the prediction or response of breast cancer patients' outcome to endocrine therapy has not been recognized to date.

(Prior art in expression analysis)
The protein encoded by the gene rather than the expression of the gene can be studied at four different levels: first, the protein expression level can be determined directly, and second, it is determined at the transcription level of the mRNA. And third, epigenetic changes such as DNA methylation characteristics of genes or histone characteristics of genes can be analyzed, because methylation is often associated with inhibition of protein expression and 4. The gene itself may be analyzed for genetic changes such as mutations, deletions, polymorphisms, etc. that affect the expression of the gene product.

  The level of observation that has been studied by recent methodological developments in molecular biology is the genes themselves, transcription of these genes into RNA, and translation into the resulting protein. However, how the activation and inhibition of a specific gene is controlled in a specific cell and tissue at a specific time in the development of an individual depends on the degree, nature, or nature of the gene methylation. May be related. In this regard, the pathogenic situation is itself evident in changes in the methylation pattern of individual genes or genomes.

  Applied to the field of broad analysis of the entire genome of all these biological processes, the four terms are called: Proteomics, Transcriptomics, Epigenomics ( Or Methylomics) and Genomics. Methods and techniques that can be used for studying expression or studying changes involved in expression for all of these levels are well described in the literature and are therefore known to those skilled in the art. They are described in molecular biology textbooks and numerous scientific papers.

  Methods for analyzing protein expression of a single gene are prior art. It usually requires an antibody specific for the gene product of interest. Appropriate techniques are ELISA and immunohistochemistry.

  Analysis of mRNA levels has also been well described. Today, the value criterion is reverse transcription PCR.

  In order to avoid duplication, a more detailed description of the prior art relating to existing known techniques is given as part of the present invention within the description of the present invention.

  US Patent Application 2003/0198970 by Gareth Roberts is a “genetic creation” of humans, that is, a technique on how to determine genetic changes such as deletions, polymorphisms, mutations, etc. that may vary between individuals. And some of the methods, describing the variability of individuals in disease, response to therapy, and the potential role of this genetic sequence information in prognosis. However, no epigenetic differences are mentioned. The gene PITX2 is listed within this application as the name of one of a long comprehensive list of about 2500 other gene names, whose expression plays a role in several types of therapeutic responses It is suggested that can be fulfilled. However, this is an assumption based solely on speculation because no experiments have been disclosed that show any kind of association between genetic changes in PITX2 and individual variability in treatment response.

  A region that is hardly established in this situation is the field of epigenomics or epigenetics, ie the field related to the analysis of DNA methylation patterns.

(Prior art in methylation analysis)
5-Methylcytosine is the most frequent covalent base change in eukaryotic DNA. DNA methylation plays an important role in the control of gene expression in mammalian cells. It plays a role in, for example, transcriptional regulation, genetic imprinting, and tumorigenesis. DNA methyltransferase is required for DNA methylation and catalyzes the transfer of the methyl group from S-adenosylmethionine to a cytosine residue, which is a modified base found mostly at the CpG site of the genome, 5-methyl Form cytosine. The presence of methylated CpG islands in the promoter region of genes can suppress their expression. This process may be due to the presence of 5-methylcytosine, which is clearly prevented by the binding of transcription factors or other DNA-binding proteins that block transcription. In different types of tumors, aberrant or accidental methylation of CpG islands in the promoter region has been discovered for many cancer-related genes, resulting in silencing of their expression. Such genes include genes that suppress metastasis, angiogenesis, and tumor suppressor genes that repair DNA (Momparler and Bovenzi (2000) J. Cell Physiol. 183: 145-54). Therefore, the identification of 5-methylcytosine as an element of genetic information is a consideration. However, since 5-methylcytosine has the same base-pairing behavior as cytosine, the position of 5-methylcytosine can be identified by sequencing. Furthermore, epigenetic information carried by 5-methylcytosine is completely lost during PCR amplification.

(Methylation analysis technology)
Furthermore, DNA methylation has also been described to play a role in the field of pharmacogenetics. A similar approach related to the application of information related to genomic genetic changes to the analysis of an individual's response to treatment is described, for example, in US application 2003/0198970 by Gareth Roberts, in application WO 02/037398. It is tailored to the application of information about epigenetic changes in the genome based on DNA methylation analysis, and is investigating treatment selection guidelines and individual therapeutic responses.

  An example of the application of this concept is given by Esteller et al. (Esteller et al. (2000) N Engl J Med. 2000 Nov 9; 343 (19): 1350-4.) Where methylation of the MGMT promoter in gliomas is an alkylating agent. It was shown to be a useful predictor of tumor response to. More recently, Fruehwald has compiled a series of studies showing that DNA methylation is associated with the aggressiveness of various cancers (Fruhwald MC. DNA methylation patterns in cancer: novel prognostic indicators? Am J Pharmacogenomics. 2003 ; 3 (4): 245-60).

  An example of the potential for analysis of epigenetic changes, such as DNA methylation analysis in relation to predicting treatment response—with regard to breast cancer—is San Marton et al., San Antonio Breast Cancer Symposium, San Antonio, TX, December, 3 -6, 2003 (Non-Patent Document 3). Breast cancer patients initially treated by surgical removal of the tumor were treated for metastases with tamoxifen. The first tumor sample was analyzed for aberrant methylation patterns. Patients were then divided into two subclasses according to their objective tumor response: patients with progressive disease (increasing the size of metastases), and complete or partial remission of recurrent tumors Patient (reduces the size of metastases). The two subclasses are distinguished based on their methylation pattern. This provides a clear indication that the methylation pattern described in the study can serve as a predictive therapeutic response tool for endocrine treatments such as tamoxifen. The result of this study is the matter of patent application WO 04/035803 published on April 29, 2004 (Patent Document 5): Methods and nucleic acids for improved treatment of breast cancer cell proliferative disorders. PITX2 is also listed as a predictive marker in the application, however, the use of the marker is only described as a therapeutic response marker, not as a prognostic marker.

  Recently, several predictive markers are under evaluation. Recently, the only commonly used treatment that targets the endocrine pathway is tamoxifen, however, many of the biomarkers associated with tamoxifen response have the same mechanism of action, or all target the same pathway. It is predictable that it is related to drugs. For example, expression of endocrine receptors (hereinafter also referred to as “ER”) and progesterone receptors (hereinafter also referred to as “PR”) is used to select patients for any treatment that targets the endocrine pathway. . A marker associated with tamoxifen response is bcl-2. High expression levels of bcl-2 have shown promising correlations with tamoxifen therapy response and survival in patients with metastatic disease, adding valuable information to the ER negative patient subgroup (J Clin Oncology, 1997, 15 5: 1916 -1922; Endocrine, 2000, 13 (1): 1-10 (Non-Patent Documents 4 and 5)). There is conflicting evidence related to independent predictive value of c-erbB2 (Her2 / neu) overexpression in patients with advanced breast cancer that needs further evaluation and demonstration (British J of Cancer, 1999, 79 ( 7/8): 1220-1226 (Non-patent document 6); J Natl Cancer Inst, 1998, 90 (21): 1601-1608 (Non-patent document 7)).

  Other predictive markers include SRC-1 (steroid receptor coactivator-1), CGA mRNA overexpression, cell kinetics, and S phase fraction assay (Breast Cancer Res and Treat, 1998, 48: 87-92; Oncogene, 2001, 20: 6955-6959 (Non-Patent Document 8)). In recent years, both uPA (urokinase-type plasminogen activator) and PAI-1 (plasminogen activator inhibitor type 1) can be useful in defining subgroups of patients that benefit from poorer prognosis and adjuvant systemic therapy. (J Clinical Oncology, 2002, 20 n04 (Non-patent Document 9)). However, all of these markers require further evaluation in future trials because none of them has a demonstrated response.

  Furthermore, the results of the study were presented by Paik et al., The San Antonio Breast Cancer Symposium, San Antonio, TX, December, 3-6, 2003 (Non-Patent Document 10). It was provided by analyzing the RNA expression pattern by RT-PCR.

  However, it is unlikely that the marker is suitable for use in commercial testing with a large number of genes. It is preferred that a more limited number of genes be analyzed for commercially available tests.

  Also recently published is a study on the prognostic power of methylation analysis in breast cancer patients. Muller et al. (Muller HM, Widschwendter A, Fiegl H, Ivarsson L, Goebel G, Perkmann E, Marth C, Widschwendter M. (2003) DNA methylation in serum of breast cancer patients: an independent prognostic marker. Cancer Res. 2003 Nov 15 63 (22): 7641-5. (Non-Patent Document 11)) described a set of genes that can be used as biomarkers for prognosis in breast cancer patients by analysis of serum before treatment. The specific aberrant methylation patterns of the two genes found in the DNA from the pre-treatment serum of cancer patients indicated whether their prognosis was good or bad. The analyzed DNA was not tissue-derived DNA but serum DNA. The presence of the most promising tumor-specific pattern indicates the presence of tumor-derived DNA; however, the absence of a specific methylation pattern causes a tumor that does not show this methylation pattern or enough DNA to flow into the bloodstream. There may be no tumor. A good or poor prognosis is defined as a long-term or short-term “overall survival” after surgery without adjuvant therapy. This result is therefore relevant for patients who do not receive post-surgical treatment. The marker is therefore considered purely prognostic (unless otherwise proven). Markers do not provide any information regarding therapeutic response, but may only provide very basic guidance on tumor aggressiveness. On this basis, clinicians can only speculate on the suitability of treatment options. However, regardless of tumor aggressiveness, it is standard to provide the majority of patients with tamoxifen (or other endocrine therapy) as an adjunct therapy, so these markers are not applicable to most patients.

  Thus, the need for improved treatment techniques for breast cancer patients that are not met by current technology has long been felt and still exists: more particularly, among these markers, the above None of the specific problems outlined in can be answered, i.e., patients treated by primary therapy (often surgical), endocrine therapy, such as but not limited to tamoxifen or aromatase Whether the patient is a good candidate for treatment with an inhibitor alone, or if the patient has been treated with further adjuvant treatment (eg, but not limited to chemotherapy) instead of or in addition to the endocrine treatment Whether or not it shows a better prognosis.

  Regardless of treatment, a pure prognostic marker for cancer patients is not a preferred solution to the need in the prior art as described above. While the markers provide some indication of tumor aggressiveness and thus guide the choice of treatment that may be necessary, they do not take into account the generational change of cancer with respect to therapeutic response. Thus, patients with a poor prognosis (determined using the pure prognostic marker) may respond well to endocrine therapy as well as adjunct therapy, regardless of disease aggressiveness. If the treatment has an inadequate response, alternative and / or additional treatments may be suitable for treatment even if the patient has a good prognosis.

  In one aspect, the present invention provides a prognostic marker, PITX2, which is recognized as a gene encoding the protein PITX2; its mRNA transcript is NM_153426, however it is “pure prognostic” is not. This marker can be used to provide guidance information on whether adjunct chemotoxic therapy is approved in addition to endocrine treatments such as tamoxifen, or whether this is an unnecessary burden on the patient. It solves the need in the prior art as outlined in.

  It is disclosed herein that aberrant expression of the gene PITX2 correlates with prognosis and / or outcome prediction of treatment with estrogen treatment of breast cell proliferative disorders patients, particularly breast carcinoma.

  This marker thus provides a new means of characterizing breast carcinoma. Abnormal expression of the gene PITX2 is an indicator of recurrence and / or survival of breast cancer patients. The invention described herein is therefore particularly useful for making improved therapeutic decisions.

  Relapse of the patient when targeted by the estrogen receptor, synthesis or conversion pathway, or otherwise treated with one or more therapies necessary for the metabolism, production or secretion of estrogen And / or an indicator of survival.

  The invention described herein targets the estrogen receptor pathway or is required for estrogen metabolism, production, or secretion from an individual optionally treated with other treatments in addition to or instead of the treatment It is particularly useful for distinguishing between individuals who have been properly treated with one or more treatments. Preferably “other treatment” includes, but is not limited to, chemotherapy or radiation therapy.

  Thus, if the patient is treated with an adjunct therapy that targets the endocrine pathway, the marker may be used in the treatment of breast cancer patients by allowing patient classification according to their likely outcome. Particularly preferred. It is further preferred to provide further adjuvant treatments, in particular but not limited to chemotherapy, instead of or in addition to said endocrine therapy, to patients with poor treatment results. A marker suitable for said purpose is also referred to below as “auxiliary marker”.

The present invention also relates to the use of PITX2 as an “adjunct marker”, which is useful as a “prognostic marker”, particularly in hormone receptor negative women who have not received any endocrine treatment.
WO 02/077272 WO 01/19845 WO 02/00927 WO 01/092565 WO 04/035803 Toyota et al. (2001), Blood 97: p 2823-9. Shivapurkar et al., Cancer Research 59, 3576-3580, August 1, 1999 Martens et al., The San Antonio Breast Cancer Symposium, SanAntonio, TX, December, 3-6, 2003 J Clin Oncology, 1997, 15 5: 1916-1922 Endocrine, 2000, 13 (1): 1-10 British J of Cancer, 1999, 79 (7/8): 1220-1226 J Natl Cancer Inst, 1998, 90 (21): 1601-1608 Breast Cancer Res and Treat, 1998, 48: 87-92; Oncogene, 2001, 20: 6955-6959 J Clinical Oncology, 2002, 20 n04 Paik et al., The San Antonio Breast Cancer Symposium, San Antonio, TX, December, 3-6, 2003 Muller HM, Widschwendter A, Fiegl H, Ivarsson L, Goebel G, Perkmann E, Marth C, Widschwendter M. (2003) DNA methylation in serum of breast cancer patients: an independent prognostic marker.Cancer Res. 2003 Nov 15; 63 ( 22): 7641-5.

(Description)
Breast cancer characterization with respect to prognosis and / or treatment outcome allows physicians to make informed decisions about treatment plans with an appropriate balance of risk and benefit for the patient.

  In the present invention, the terms “estrogen receptor positive” and / or “progesterone receptor positive” when used to describe a breast cell proliferative disorder means that proliferating cells express the hormone receptor.

  In the present invention, the term “aggressive” refers to a high likelihood of recurrence after surgery; patient survival below the mean or below the median; disease-free below the mean or below the median Means one or more of: survival; relapse-free survival below mean or median; occurrence of tumor-related complications below mean; rapid progression of tumor or metastasis. Depending on the aggressiveness of the disease, the appropriate treatment or treatments may be chemotherapy, radiotherapy, surgery, biological therapy, immunotherapy, antibody therapy, therapy using molecular targeted drugs, estrogen receptor modulator therapy, estrogen It may be selected from the group consisting of receptor down-regulator treatment, aromatase inhibitor treatment, ovariectomy, treatment giving LHRH analogs, or other centrally acting drugs that affect estrogen production. Where the cancer is characterized as “aggressive”, it is particularly preferred that a treatment such as but not limited to chemotherapy be provided in addition to or in place of the endocrine targeted treatment. Indicators of tumor aggressiveness criteria in the prior art include, but are not limited to, tumor stage, tumor grade, lymph node status and survival.

  Unless otherwise stated, as used herein, the term “survival” is considered to include all of the following: survival to death, also known as overall survival (the death is irrelevant to the cause, or “Relapse-free survival” (the term recurrence includes both local and remote recurrence); survival without metastases; disease-free (Disease) Survival (The term disease includes breast cancer and related diseases). The length of the survival period may be calculated by referring to a fixed starting point (eg, at the time of diagnosis or treatment) and a final point (eg, death, recurrence, or metastasis).

  As used herein, the term “prognostic marker” means an indicator of the prognostic likelihood of a disease, in particular the aggressiveness and metastatic potential of a breast tumor.

  As used herein, the term “predictive marker” means an indicator of response to therapy, preferably said response is defined by the patient's lifetime. Preferably it is used to define patients with high, low, medium length survival or relapse after treatment, which is a result of the inherent heterogeneity of the disease pathway.

  As defined herein, the term predictive marker may fall within the scope of the “prognostic marker” described herein in some situations, for example, the survival time when the prognostic marker varies depending on the treatment. In identifying patients with outcome, the marker is also a predictive marker for the treatment. Thus, unless otherwise noted, the two terms are not incompatible with each other.

  As used herein, the term “expression” refers to the transcription and translation of a gene, as well as genetic or epigenetic changes in the genomic DNA associated with the marker gene and / or its regulatory or promoter regions. Genetic changes include SNPs, point mutations, deletions, insertions, repeat lengths, transpositions, and other polymorphisms. Analysis of the expression level of either protein or mRNA, or analysis of changes in genetic or epigenetic marker genes in a patient individual, is summarized here as an analysis of gene “expression”.

  The level of gene expression is related to the level of gene transcription and translation, including but not limited to methylation analysis, loss of heterozygosity (also referred to herein as LOH), RNA expression level and protein expression level Or determined by analysis of the indicated factors.

  In addition, the activity of the transcribed gene is not limited, but includes genetic changes (including but not limited to SNPs, point mutations, deletions, insertions, repeat lengths, transpositions, and other polymorphisms). Can be affected by mutations.

  The term “endocrine therapy” or “endocrine therapy” is meant to include therapy, single therapy, or multiple therapies that target the estrogen receptor pathway or estrogen synthesis pathway or estrogen conversion pathway, and these are estrogen metabolism. Necessary for production or secretion. Such treatments include, but are not limited to, estrogen receptor modulators, estrogen receptor downregulators, aromatase inhibitors, ovariectomy, LHRH analogs, and centrally acting drugs that affect estrogen production.

  The term “monotherapy” means the use of a single drug or other therapy.

  In the context of the present invention, the term “chemotherapy” means the use of a medicament or chemical to treat cancer. This definition excludes radiotherapy (treatment with high energy light or particles), hormone therapy (treatment with hormones or hormone analogs) and surgical treatment.

  In the context of the present invention, the term “adjuvant treatment” means the treatment of a cancer patient immediately after the initial non-chemotherapeutic therapy, eg surgical treatment. In general, the purpose of adjuvant therapy is to reduce the risk of recurrence.

  In the present invention, the term “determining a suitable treatment plan for a patient” is initiated based on, or essentially based on, or at least in part based on the analysis results of the present invention, It means the determination of a treatment plan for a patient that is changed and / or terminated (ie, a single therapy or a combination of different therapies used to prevent and / or treat cancer in the patient). One example is to start adjuvant endocrine therapy after surgery, and another is to change the dose of a particular chemotherapy. The determination can be based on the personal characteristics of the subject to be treated, in addition to the analysis results according to the present invention. In most cases, the actual determination of a treatment plan suitable for the subject is performed by the attending physician or doctor.

  In the present invention, the terms “obtain a biological sample” or “obtain a sample from a subject” do not include active recovery of a sample from an individual, eg, performing a biopsy. The term refers to obtaining a sample previously isolated from an individual. The sample can be isolated by standard means in the prior art, including but not limited to body fluids isolated by biopsy, surgical removal, aspiration. Further, the sample may be provided by three parties including a clinician, a courier, a commercial sample provider, and a sample collector.

  In the present invention, the term “CpG island” has (1) a frequency of CpG dinucleotides corresponding to “observation / prediction ratio”> 0.6, and (2) a genomic DNA that satisfies a standard having “GC content”> 0.5. Indicates a continuous region. CpG islands are typically, but not always, between about 0.2 and about 1 kb in length.

  In the present invention, the term “regulatory region” of a gene means a nucleotide sequence that affects the expression of the gene. The regulatory region may be located within, proximal or distal of the gene. Such regulatory regions include, but are not limited to, constitutive promoters, tissue specific promoters, development specific promoters, inducible promoters, and the like. Promoter regulatory elements may also include specific enhancer sequence elements that control the transcription or translation efficiency of the gene.

  In the present invention, the term “methylation” indicates the presence or absence of 5-methylcytosine (“5-mCyt”) at one or more CpG dinucleotides in a DNA sequence.

  In the present invention, the term “methylation state” means the degree of methylation present in a nucleic acid of interest, which can be expressed in absolute or relative terms, ie percentage or other numerical values, or other tissues As compared to, and hypermethylated, hypomethylated and expressed herein as having significantly similar or identical methylation status.

In the present invention, the term “hemi-methylation” or “hemimethylation” refers to the methylation state of a CpG methylation site, and is simply in one of two CpG dinucleotide sequences of a double-stranded CpG methylation site. Only one cytosine is methylated (eg, 5′-NNC ^M GNN-3 ′ (upper strand): 3′-NNGCNN-5 ′ (lower strand)).

  In the present invention, the term “hypermethylation” means that 5-mCyt at one or more CpG dinucleotides in the DNA sequence of the test DNA sample is found in the corresponding CpG dinucleotide in the normal control DNA sample. Shows the average methylation state corresponding to the presence of increasing amounts of mCyt.

  In the present invention, the term “hypomethylation” means that 5-mCyt at one or more CpG dinucleotides in the DNA sequence of the test DNA sample is found in the corresponding CpG dinucleotide in the normal control DNA sample. The average methylation state corresponding to being present in a reduced amount relative to the amount of mCyt is shown.

  In the present invention, the term “microarray” broadly refers to both “DNA microarray” and “DNA chip”, and as recognized in the prior art, encompasses all prior art recognized solid supports, including nucleic acids It includes all methods of attaching molecules or synthesizing nucleic acids thereon.

  “Genetic parameters” are gene mutations and polymorphisms, as well as sequences required for their regulation. What are manifested as genetic changes or mutations are in particular insertions, deletions, point mutations, inversions and polymorphisms, particularly preferably SNPs (single nucleotide polymorphisms).

  An “epigenetic change” or “epigenetic parameter” is a change in the DNA bases of genomic DNA and the sequences that are further required for their regulation, in particular the cytosine methylation. In addition, epigenetic parameters include, for example, histone acetylation associated with DNA methylation, which cannot be directly analyzed using the described methods.

  In the present invention, the term “bisulfite reagent” refers to a reagent containing bisulfite, disulfate, bisulfate, or a combination thereof, and distinguishes between methylated CpG dinucleotide sequences and unmethylated CpG dinucleotide sequences. Is useful as disclosed herein.

  As used herein, the term “methylation analysis” refers to any analysis that determines the methylation status of one or more CpG dinucleotide sequences within a sequence of DNA.

  In the present invention, the term “MS.AP-PCR” (Methylation-Sensitive Arbitrarily-Primed Polymerase Chain Reaction) is a genome that uses CG-rich primers to focus on the region most likely to contain CpG dinucleotides. Shows a prior art recognition technique that allows for a global scan of, which is described by Gonzalgo et al., Cancer Research 57: 594-599, 1997.

In the present invention, the term “MethyLight ^™ ” refers to fluorescence-based real-time PCR recognized by the prior art described by Eads et al., Cancer Res. 59: 2302-2306, 1999.

In the present invention, the term “HeavyMethyl ^™ ” analysis refers to a methylation analysis that, in its example performed here, includes a methylation-specific blocking probe covering the CpG position between amplification primers.

  The term “Ms-SNuPE” (Methylation-sensitive Single Nucleotide Primer Extension) indicates a prior art recognition analysis described by Gonzalgo & Jones, Nucleic Acids Res. 25: 2529-2531, 1997.

  In the present invention, the term “MSP” (Methylation-specific PCR) is defined by Herman et al., Proc. Natl. Acad. Sci. USA 93: 9821-9826, 1996, and US Patent No. 5,786,146. Figure 2 shows the prior art perceived analysis described.

  In the present invention, the term “COBRA” (Combined Bisulfite Restriction Analysis) refers to the methylation analysis recognized by the prior art described by Xiong & Laird, Nucleic Acids Res. 25: 2532-2534, 1997.

  In the present invention, the term “hybridization” is understood as the binding of an oligonucleotide to a complementary sequence along the Watson-Crick base pair line in the sample DNA forming a duplex structure. It is.

  “Stringent hybridization conditions” are as described herein: hybridized in 5 × SSC / 5 × Denhardt solution / 1.0% SDS at 68 ° C. and 0.2 × SSC / 0.1% SDS at room temperature. Washing, or equivalents recognized by the prior art (for example, performing hybridization in 2.5 × SSC buffer at 60 ° C. and washing several times with low concentration buffer at 37 ° C. Conditions that continue and remain stable). Moderate stringent conditions include washing at 42 ° C. in 3 × SSC, as defined herein, or equivalents recognized by the prior art. The salt concentration and temperature parameters can be varied to achieve an optimal level of identity between the probe and the target nucleic acid. Guidance regarding such conditions is available in the prior art, for example, Sambrook et al., 1989, Molecular Cloning, A Laboratory Manual, Cold Spring Harbor Press, NY; and Ausubel et al. (Eds.), 1995, Current Protocols in Molecular Biology. , (John Wiley & Sons, NY), Unit 2.10.

  “Background DNA” as used herein refers to all nucleic acids originating from other sources than breast cells.

  Using the methods and nucleic acids described herein, a statistically significant model of patient recurrence, disease-free survival, metastasis-free survival, overall survival, and / or disease progression can be developed in a treatment plan. It can be used to help patients and clinicians in determining suitable treatment options to be included.

  In one aspect, the method provides a prognostic marker for cell proliferation disorders in breast tissue. Preferably, the prognosis is provided for a result selected from the group consisting of relapse; patient overall survival; metastasis-free survival; disease-free survival or possible disease progression.

  In a further aspect of the invention, the marker targets the estrogen receptor pathway or results from a treatment required for estrogen metabolism, generation, secretion as a therapy for patients suffering from cell proliferation disorders of breast tissue. Used as a predictive marker. This aspect of the method allows the physician to determine which treatment to use in addition to or instead of the endocrine treatment. The additional treatment is preferably a more aggressive therapy, such as but not limited to chemotherapy. Thus, it can be seen that the present invention reduces the problems associated with existing breast cell proliferative disorder prognosis, prediction and treatment response prediction methods.

  Using the methods and nucleic acids described herein, patient survival is determined prior to treatment for cell proliferation disorders in breast cancer tissue or to provide baseline information to patients and physicians regarding likely progression of the disease. Can be evaluated in the middle. Thus, the methods and nucleic acids exemplified herein improve the patient's quality of life and the likelihood of successful treatment by allowing both the patient and physician to make a more accurate determination of the patient's treatment options. Can help.

  The methods described herein can be used for improved treatment of all breast cancer cell proliferation disorder patients, both before and after menopause, regardless of lymph node or estrogen receptor status. However, it is particularly preferred that the patient is lymph node negative and estrogen receptor positive.

  The present invention enables breast cell proliferative disorders by predicting the likelihood of recurrence after surgery, by allowing improved prediction of patient survival, particularly with or without adjuvant endocrine therapy. Make available methods for improved treatment. In addition, the present invention provides a means for improved prediction of treatment outcome with endocrine therapy, wherein the therapy includes one or more treatments required for the estrogen status pathway, or estrogen metabolism, production, secretion.

  The method according to the present invention includes, but is not limited to, noninvasive ductal carcinoma, invasive ductal carcinoma, noninvasive lobular carcinoma, comedous cancer, inflammatory cancer, mucinous cancer, hard cancer, colloid cancer, tubular cancer, marrow It can be used to analyze cell proliferation disorders in a wide variety of breast tissues, including like cancer, metaplastic cancer, papillary cancer, and non-invasive papillary cancer, undifferentiated or anaplastic cancer, and Paget's disease of the breast.

  The method according to the present invention can be used to provide a prognosis of patients with breast cell proliferative disorders, and further said method is used to provide a prediction of patient survival and / or recurrence after treatment with endocrine therapy. Can be done.

  When the markers, methods, and nucleic acids disclosed herein are used as prognostic markers, it is particularly preferred that the prognosis is defined in terms of patient survival and / or recurrence. In this embodiment, patient survival time and / or recurrence is predicted by their gene expression or their genetic or epigenetic changes. In this aspect of the invention, it is particularly preferred that the patient is tested before receiving all adjuvant endocrine treatments.

  When the markers, methods, and nucleic acids disclosed herein are used as predictive markers, endocrine therapy is received as a secondary therapy for the first non-chemotherapeutic therapy, such as surgery, as shown in FIG. It is applied to the prediction of patient results (hereinafter also referred to as “auxiliary settings”). Such treatment is often prescribed for patients with stage 1-3 breast carcinoma. It is also preferred that said “result” is defined in terms of patient survival and / or recurrence.

  In this embodiment, patient survival time and / or recurrence is predicted by their gene expression or their genetic or epigenetic changes. By detecting patients with a metastatic or disease-free survival time below the mean or below the median and / or patients with a high likelihood of recurrence, the physician may replace the endocrine targeted therapy, or In addition to this, the patient may choose to recommend further treatment, such as but not limited to chemotherapy.

  The invention described herein provides prognostic and predictive biomarkers for novel breast cancer cell proliferation disorders.

  It is described herein that aberrant expression of the gene PITX2 and / or their regulatory or promoter regions is associated with prognosis and / or outcome prediction of estrogen treatment of patients with breast cell proliferation disorders, particularly breast carcinoma.

  This marker thereby provides a new means of characterizing breast cell proliferative disorders. As described herein, determination of the expression of the gene PITX2 and / or their regulatory or promoter regions allows for prognostic prediction of patients with breast tissue proliferative disorders. In an alternative embodiment, the expression of the gene PITX2 and / or their regulatory or promoter region targets the estrogen receptor, synthesis or conversion pathway or otherwise is required for estrogen metabolism, production, secretion 1 or Allows prediction of treatment response of patients treated with further treatment.

  The invention described herein adds to the above treatment of an individual thereby targeting the estrogen receptor pathway or that can be appropriately treated with one or more therapies necessary for estrogen metabolism, production, or secretion. It is also useful for distinguishing from individuals who are additionally treated with other therapies. Preferably "other treatments" include but are not limited to chemotherapy or radiotherapy. It is particularly preferred that the prognostic and / or therapeutic response is indicated in terms of recurrence, survival or outcome.

In a further embodiment of the invention, the abnormal expression of a plurality of genes comprising the gene PITX2 and / or their regulatory or promoter regions is analyzed. The plurality of genes are also referred to below as “gene panels”. Analysis of multiple genes increases the accuracy of the provided prognosis and / or outcome prediction of estrogen treatment. The gene panel preferably consists of no more than 7 genes and / or their promoter regions associated with prognosis and / or prediction of treatment response in breast carcinoma patients. More preferably, the panel comprises genes PITX2 and one or more genes selected from ABCA8, CDK6, ERBB2, ONECUT2, PLAU, TBC1D3 and TFF1, and / or their regulatory regions. It is particularly preferred that the gene panel is selected from the group of gene panels consisting of:
・ PITX2, PLAU and TFF1
・ PITX2 and PLAU
・ PITX2 and TFF1

  It is particularly preferred that a gene panel comprising PITX2 and TFF1 is used to predict the outcome of patient treatment with endocrine therapy. It is particularly preferred that a gene panel comprising PITX2 and PLAU is used to provide prognosis for patients. Preferably, the patient is analyzed before receiving all endocrine treatments.

  In a further embodiment, the present invention relates to new methods and sequences for the prognosis of patients diagnosed with breast cell proliferative disease. In a further aspect, the present invention relates to a new method and sequence, as a tool for selecting a suitable treatment for patients diagnosed with breast cell proliferative disease based on the prediction of recurrence, survival or possible outcome. Can be used.

  More particularly, the present invention provides new methods and arrangements for patients diagnosed with breast cell proliferative disease, which allows an improved selection of suitable adjuvant therapies. Furthermore, it is preferred that patients with poor prognosis after endocrine monotherapy receive chemotherapy in addition to or instead of endocrine therapy.

  One aspect of the present invention is the provision of a method that provides prognosis and / or outcome prediction of endocrine therapy in patients with breast tissue cell proliferation disorders. Preferably, said prognosis and / or prediction is given with respect to the likelihood of relapse or survival of the early patient. More preferably, the survival period is a disease-free survival period or a metastasis-free survival period. It is also preferred that the disease is breast cancer. These methods include analysis of the expression level of the gene PITX2 and / or their regulatory regions.

In a further embodiment, the method comprises the gene PITX2 and one or more genes selected from the group consisting of ABCA8, CDK6, ERBB2, ONECUT2, PLAU, TBC1D3 and TFF1, and / or their regulatory regions Analysis of the expression of the "panel". It is particularly preferred that said gene panel is selected from the group of gene panels consisting of:
・ PITX2, PLAU and TFF1
・ PITX2 and PLAU
・ PITX2 and TFF1

  It is particularly preferred that the expression of the gene panel comprising PITX2 and TFF1 is determined in order to predict the outcome of treatment of the patient with endocrine therapy. It is also particularly preferred that the expression of the gene panel comprising PITX2 and PLAU is determined to provide a prognosis for the patient. The patient is preferably analyzed prior to any endocrine treatment.

  Expression determination can be accomplished by any means standard in the art, however, it is accomplished by analysis of LOH, methylation, protein expression, mRNA expression, genetic sequence or other epigenetic changes Most preferred.

  Particularly preferred is the analysis of the DNA methylation profile of the genomic sequence of the gene PITX2 shown in SEQ ID NO: 149 and / or their regulatory or promoter regions. More preferred are within the following sites of SEQ ID NO: 149: nucleotide 2,700-nucleotide 3,000; nucleotide 3,900-nucleotide 4,200; nucleotide 5,500-nucleotide 8,000; nucleotide 13,500-nucleotide 14,500; nucleotide 16,500-nucleotide 18,000; nucleotide 18,500-nucleotide 19,000; Analysis of the methylation status of the CpG position from nucleotide 21,000 to nucleotide 22,500. Particularly preferred is an analysis of the methylation status of eight specific CpG dinucleotides covered by the four subsequences of SEQ ID NO: 149 shown in SEQ ID NOs: 1, 13, 18 and 19. Where the method comprises analysis of PITX2 and a gene panel comprising one or more genes selected from ABCA8, CDK6, ERBB2, ONECUT2, PLAU, TBC1D3 and TFF1, and / or their regulatory or promoter regions, The gene sequence is preferably selected from the group consisting of SEQ ID NO: 69 to SEQ ID NO: 75 and SEQ ID NO: 150 listed in Table 1.

  This methodology provides a further advance in the state of the art in that the method can be applied to any subject regardless of estrogen and / or progesterone receptor status. Thus, in a preferred embodiment, the subject need not be tested for estrogen or progesterone receptor status.

  In a further aspect of the invention, the disclosed matter provides a novel nucleic acid sequence useful for the analysis of methylation within said gene, and the other aspect is a prognosis of endocrine therapy in patients diagnosed with breast cell proliferative disease. And / or novel uses of the genes and gene products and methods, analyses, and kits for the purpose of predicting results.

  In one embodiment, the present invention provides prognosis and / or predicts outcome of endocrine therapy in patients suffering from breast cell proliferative disease by analysis of expression of gene PITX2 and / or its regulatory regions A method is disclosed. Preferably, said endocrine treatment is adjuvant endocrine monotherapy. The method may be possible by any gene expression analysis including, but not limited to, mRNA expression analysis, or protein expression analysis, or by analysis of genetic changes that lead to alternative expression (including LOH). However, in the most preferred embodiment of the invention, said expression is determined by analysis of the methylated region of the gene PITX2 and its CpG site within its promoter or regulatory element.

  In one embodiment of the method, abnormal expression of gene PITX2 and / or its panel may be detected by analysis of loss of heterozygous form of the gene. In the first step, DNA is isolated from a biological sample of the patient's tumor. Isolated DNA is analyzed for LOH by any method that is a standard in the prior art including, but not limited to, amplification of a locus or associated microsatellite marker. The amplification may be performed by any method that is a standard in the prior art, including polymerase chain reaction (PCR), strand displacement amplification (SDA) and isothermal amplification.

  The level of amplification is then detected by methods known in the prior art, including but not limited to gel electrophoresis and probe detection (including real-time PCR). Further, the amplification may be labeled to aid the detection. Suitable detectable labels include, but are not limited to, fluorescent labels, radioactive labels, and mass labels, the preferred use of which is described herein.

  With respect to the heterozygous control sample, detection of a decrease in the amount of amplification corresponding to one of the amplified alleles in the test sample indicates LOH.

  In a detection system for breast cancer recurrence, a sample is obtained from a patient to detect the level of mRNA encoding PITX2 and / or a panel comprising said gene. The acquisition of the sample should be aimed at the availability of isolated biological material representing the specific tissue relevant to the intended use, rather than the collection of the sample as in performing a biopsy. Is preferably meant. The sample can be a tumor tissue sample from a surgically removed tumor, a biopsy sample as obtained by a surgeon and provided to an analyst, or a sample such as blood, plasma, serum, and the like. The sample may be processed to extract the nucleic acid contained therein. The nucleic acid generated from the sample is subjected to gel electrophoresis or other separation techniques. Detection involves contacting the DNA sequence serving as a probe with the sample nucleic acid, and in particular mRNA, to form a hybrid duplex. Hybridization stringency is determined by a number of factors, including temperature, ionic strength, length of time, and formamide concentration during hybridization and washing procedures. These factors are outlined, for example, in Sambrook et al. (Molecular Cloning: A Laboratory Manual, 2nd ed., 1989). Detection of the resulting duplex is usually accomplished by the use of a labeled probe. Alternatively, the probe may not be labeled, but may be detectable directly or indirectly by specific binding with a labeled ligand. Suitable labels and methods for labeling probes and ligands are known in the prior art, eg radioactive labels, biotin, fluorescent groups, chemiluminescent groups that can be incorporated by known methods (eg nick translation, or kinasing) (Eg, dioxetane, especially triggered dioxetane), enzymes, antibodies and the like.

  In order to increase the sensitivity of detection in samples of mRNA encoding PITX2 and / or the panel containing the gene, reverse transcription / polymerase chain reaction techniques are transcribed from mRNA encoding PITX2 and / or the panel containing the gene. Can be used to amplify the cDNA. Reverse transcription / PCR methods are known in the prior art (see, eg, Watson and Fleming, supra).

  The reverse transcription / PCR method can be performed as follows. Cellular total RNA is isolated, for example, by the standard guanidine isocyanate method, and total RNA is reverse transcribed. Reverse transcription involves the synthesis of DNA in an RNA template using reverse transcriptase and a 3 ′ primer. Typically, a primer includes an oligo (dT) sequence. The cDNA thus generated is then amplified using the PCR method and amplified using PITX2 and / or specific primers of the panel containing said gene (Belyavsky et al., Nucl Acid Res 17: 2919-2932 , 1989; Krug and Berger, Methods in Enzymology, Academic Press, NY, Vol. 152, pp. 316-325, 1987, which is incorporated herein by reference).

  The present invention is also used to predict the likelihood of breast cancer patient recurrence and / or survival before or after surgical tumor removal with or without adjuvant endocrine monotherapy by testing biological samples. As described in certain embodiments. A representative kit is for PITX2 mRNA and / or one or more nucleic acids as described above that selectively hybridize to mRNA from a gene of the panel comprising PITX2, and for each of the one or more nucleic acid segments. A container may be included. In some embodiments, the nucleic acid segments may be mixed in a single tube. In a further embodiment, the nucleic acid segment may also include one primer pair for amplifying the target mRNA. Such kits may also include buffers, solutions, solvents, enzymes, nucleotides, or other components for hybridization, amplification, or detection reactions. Preferred kit components include reagents for reverse transcription PCR, in situ hybridization, Northern analysis and / or RPA.

  The present invention further provides a method for detecting the presence of PITX2 and / or a panel of polypeptides comprising said protein in a sample obtained from a patient. Preferably, the sequence is essentially the same as the sequence as shown in FIG. Any method known in the prior art for detecting proteins can be used. Such methods include, but are not limited to, immunodiffusion, immunoelectrophoresis, immunochemical methods, binder-ligand assay, immunohistological techniques, aggregation, and complement assay (eg, , Basic and Clinical Immunology, edited by Sites and Terr, Appleton & Lange, Norwalk, Conn. Pp 217-262, 1991, which is hereby incorporated by reference). Preferably, a binder-ligand immunoassay method comprising reacting an antibody with one or more epitopes of PITX2 and / or its panel and competitively replacing labeled PITX2 protein and / or its panel or derivatives thereof ( binder-ligand immunoassay method).

  One embodiment of the invention involves the use of an antibody that is specific for the gene PITX2 and a polypeptide encoded by a panel comprising said gene. Such antibodies can be used to detect recurrence of breast cancer patients, preferably under adjuvant endocrine monotherapy, by comparing the patient's level of PITX2 marker expression and / or expression of a panel comprising PITX2 to the expression of the same marker in normal individuals. And / or may be useful to provide a prognosis of potential survival. In certain embodiments, production of monoclonal or polyclonal antibodies can be induced by the use of PITX2 and / or other polypeptides of the panel as antigens. Such antibodies may in turn be used to detect proteins expressed as markers for the prognosis of recurrence in breast cancer patients under adjuvant endocrine monotherapy. The level of such proteins present in the patient's peripheral blood may be quantified by conventional methods. Antibody-protein binding can be detected and quantified by various methods known in the art, such as by labeling with fluorescent or radioactive ligands. The invention further comprises a kit for performing the above procedure, wherein such kit comprises antibodies specific for PITX2 and / or their gene panel polypeptides.

  Many competitive and non-competitive protein binding immunoassays are well known in the prior art. The antibody used for such analysis may be unlabeled, for example, as used in an agglutination test, or may be labeled for use in a wide variety of analytical methods. For use in radioimmunoassay (RIA), enzyme immunoassay, eg, enzyme-linked immunosorbent assay (ELISA), fluorescence immunoassay, etc., the label is a radionuclide, enzyme, fluorescent material, chemiluminescent material, enzyme substrate, or cofactor. , Enzyme inhibitors, particles, pigments and the like. Polyclonal or monoclonal antibodies against PITX2 and / or their panels or their epitopes can be made for use in immunoassay by any of a number of methods known in the art. One way to prepare antibodies against a protein is to select and prepare the amino acid sequence of all or part of the protein, to chemically synthesize that sequence, and to inject it into an appropriate animal, usually a rabbit or mouse ( Milstein and Kohler Nature 256: 495-497, 1975; Gulfre and Milstein, Methods in Enzymology: Immunochemical Techniques 73: 1-46, edited by Langone and Banatis, Academic Press, 1981, which are hereby incorporated by reference). Methods of preparation of PITX2 and / or panels thereof, or epitopes thereof include, but are not limited to, chemical synthesis, recombinant DNA techniques, or isolation from biological samples.

  In one aspect, the present invention is prior art in that it is the first single marker that can be used to predict the likelihood of recurrence or survival of breast cancer patients under adjuvant endocrine monotherapy. Provides significant progress in the current status of

  In the most preferred embodiment of the invention, expression analysis is performed by methylation analysis. More preferably, the methylation status of the CpG dinucleotide in the genomic sequence according to SEQ ID NO: 149 and the sequence complementary thereto is analyzed. SEQ ID NO: 149 discloses the gene PITX2 and its promoters and regulatory elements, wherein the fragment comprises a CpG dinucleotide that indicates an endocrine therapy-specific methylation pattern, indicates prognosis, and / or predicts outcome. More preferred are within the following sites of SEQ ID NO: 149: nucleotide 2,700-nucleotide 3,000; nucleotide 3,900-nucleotide 4,200; nucleotide 5,500-nucleotide 8,000; nucleotide 13,500-nucleotide 14,500; nucleotide 16,500-nucleotide 18,000; nucleotide 18,500-nucleotide 19,000; Analysis of the methylation status of the CpG position from nucleotide 21,000 to nucleotide 22,500. Analysis of the subsequence of gene PITX2 shown in SEQ ID NO: 1 is also preferred.

Expression of a “gene panel” wherein the method comprises genes and / or their regulatory or promoter regions and one or more genes selected from the group consisting of ABCA, CDK6, ERBB2, ONECUT2, PLAU, TBC1D3 and TFF1 Most preferably, the expression analysis is performed by methylation analysis. It is particularly preferred that CpG methylation of a gene panel selected from the group of gene panels comprising:
・ PITX2, PLAU and TFF1
・ PITX2 and PLAU
-PITX2 and TFF1.

  It is particularly preferred that the methylation of the gene panel comprising PITX2 and TFF1 is determined in order to predict the outcome of patient treatment with endocrine therapy. It is also particularly preferred that the methylation of the gene panel comprising PITX2 and PLAU is determined to provide a prognosis for the patient. The patient is preferably analyzed prior to any endocrine treatment.

  Hypermethylation of PITX2 and other selected genes and / or their sequences as described herein is associated with poor prognosis and / or outcome of endocrine treatment of breast cell proliferative disorders, most preferably breast carcinoma doing.

  The methylation pattern of the gene PITX2 and its promoter and regulatory elements has never been analyzed for prognosis or prediction of outcome of endocrine therapy in patients diagnosed with breast cell proliferative disorders. Due to the degeneracy of the genetic code, the sequence as specified in SEQ ID NO: 149 is substantially all upstream of the promoter region of the gene encoding the polypeptide having the biological activity of the polypeptide encoded by PITX2. To be interpreted to include sequences that are similar and equivalent.

  Most preferably, the following method is used to detect methylation in the gene PITX2 and / or its regulatory or promoter region, wherein said methylated nucleic acid is present in excess background DNA and background There is 100-1000 times the concentration of DNA at which DNA is detected.

  The method for analyzing methylation comprises contacting a nucleic acid obtained from a subject with at least one reagent or a series of reagents, wherein the reagent or series of reagents comprises methylated and unmethylated CpG in the target nucleic acid. Distinguish dinucleotides.

  Preferably, the method comprises the following steps: In a first step, a tissue sample to be analyzed is obtained. The raw material may be any suitable raw material, preferably the sample raw material consists of histological slide, biopsy, paraffin embedded tissue, body fluid, serum, feces, urine, blood, nipple aspirate and combinations thereof Selected from the group. Preferably, the source is tumor tissue, biopsy, serum, urine, blood, or nipple aspirate. The most preferred source is a tumor sample that has been surgically removed from the patient or from a biopsy sample of the patient.

  The DNA is then isolated from the sample. Genomic DNA may be isolated by any method that is standard in the art, including the use of commercially available kits. In brief, if the DNA of interest is encased by / in the cell membrane, the biological sample must be ground and lysed by enzymatic, chemical or physical means. The DNA solution may then remove proteins and other contaminants, for example by degradation with protease K. Genomic DNA is then recovered from the solution. This can be performed by various means including salting out, organic extraction, or binding of DNA to a solid support. The choice of method is influenced by several factors including time, loss, and required amount of DNA.

  The genomic DNA sample is then processed in a manner that converts cytosine bases that are not methylated at the 5 ′ position to uracil, thymine, or other bases that differ from cytosine with respect to hybridization behavior. This is understood here as “processing” or “preprocessing”.

  The above pretreatment of genomic DNA is preferably performed using bisulfite (bisulfate, disulfate) followed by uracil or base pairing behavior of unmethylated cytosine nucleobases different from cytosine Alkaline decomposition occurs which results in conversion to other bases. Encapsulation of the DNA to be analyzed in an agarose matrix, thereby preventing DNA diffusion and reconstitution (bisulfite only reacts with single-stranded DNA), removal of all precipitates, and purification steps by rapid dialysis (Olek A et al., A modified and improved method for bisulfite based cytosine methylation analysis, Nucleic Acids Res. 24: 5064-6, 1996) is one preferred example of a method for performing the pretreatment. More preferably, the bisulfite treatment is performed in the presence of a radical scavenger or a DNA denaturant.

  The treated DNA is then analyzed to determine the methylation status of gene PITX2 (and / or its regulatory region) (prior to treatment) in relation to prognosis and / or outcome of endocrine therapy. In a further embodiment of the method, the methylation status of gene PITX2 and / or its regulatory region, and one or more genes selected from the group comprising ABCA8, CDK6, ERBB2, ONECUT2, PLAU, TBC1D3 and TFF1, And / or the methylation status of their regulatory or promoter regions. It is particularly preferred that the methyl status of the gene panel selected from the group of gene panels comprising PITX2, PLAU and TFF1; PITX2 and PLAU; PITX2 and TFF1 is determined. More preferably, the sequence of the gene described in the attached sequence listing (see Table 3) is analyzed.

  In the third step of the method, the pretreated DNA fragment is amplified. When the DNA source is free DNA from serum or DNA extracted from paraffin, the size of the amplified fragment is particularly preferably between 100 and 200 base pairs long, and the DNA source is a cell source ( For example, when extracted from a tissue, biopsy, cell line), the amplification product is preferably between 100 and 350 base pairs long. It is particularly preferred that the amplification product comprises a sequence of at least 20 base pairs comprising at least 3 CpG dinucleotides. Said amplification is carried out using a set of primer oligonucleotides according to the invention and preferably a thermostable polymerase. Amplification of several DNA segments can be performed simultaneously in one and the same reaction vessel, and in one embodiment of the method, 6 or more fragments are amplified simultaneously. Typically, amplification is performed using the polymerase chain reaction (PCR). The set of primer oligonucleotides is reverse to each of at least 18 base pair long segments of SEQ ID NO: 2-5, SEQ ID NO: 76 to SEQ ID NO: 103, SEQ ID NO: 151 to SEQ ID NO: 158 and complementary sequences thereof It comprises at least two oligonucleotides having sequences that hybridize under complementary, identical, or stringent or highly stringent conditions.

  In a preferred embodiment of the method, the primer may be selected from the group consisting of SEQ ID NO: 6 to SEQ ID NO: 10.

  In an alternative embodiment of the method, the methylation status of a preselected CpG position within a nucleic acid sequence comprising SEQ ID NO: 1, SEQ ID NO: 60 to SEQ ID NO: 75, SEQ ID NO: 149 and SEQ ID NO: 150 is methylation specific It may be detected by a primer oligonucleotide. This technique (MSP) is described in Herman US Pat. No. 6,265,171. The use of methylation state specific primers for bisulfite-treated DNA amplification allows the distinction between methylated and unmethylated nucleic acids. The MSP primer pair includes at least one primer that hybridizes to a bisulfite treated CpG dinucleotide. Thus, the primer sequence comprises at least one CpG, TpG or CpA dinucleotide. MSP primers specific for unmethylated DNA contain a “T” at the 3 ′ position of the C position in CpG. Preferably, therefore, the base sequence of said primer is at least 18 which hybridizes to SEQ ID NO: 2 to SEQ ID NO: 5, and SEQ ID NOs: 151, 152, 155 and 156 and nucleic acid sequences pretreated with sequences complementary thereto. It is necessary to include a sequence having a nucleotide length, wherein the base sequence of the oligomer comprises at least one CpG, tpG or Cpa dinucleotide. In one embodiment of the method according to the invention it is particularly preferred that the MSP primer comprises between 2 and 4 CpG, tpG or Cpa dinucleotides. It is further preferred that the dinucleotide is located within the 3 ′ half of the primer where the particular dinucleotide is present within the first 9 bases from the 3 ′ end of the molecule, for example when the primer is 18 bases long. . In addition to CpG, tpG or Cpa dinucleotide, the primer further comprises a number of bisulfite converting bases (ie cytosine converted to thymine, or guanine converted to adenine in the hybridizing strand). preferable. In a further embodiment, the primer is designed to contain at most 2 cytosine or guanine bases.

  Fragments obtained by amplification can have a detectable label, either directly or indirectly. Preference is given to labels in the form of fluorescent labels, radionuclides or separable molecular fragments having a typical mass that can be detected with a mass spectrometer. When the label is a mass label, the labeled amplification product preferably has one positive or negative net charge for better detection performance with a mass spectrometer. Detection is performed and visualized, for example, with a matrix-assisted laser desorption / ionization mass spectrometer (MALDI) or using an electrospray mass spectrometer (ESI).

  Matrix-assisted laser desorption / ionization mass spectrometry (MALDI-TOF) is a very effective development for the analysis of biomolecules (Karas and Hillenkamp, Anal Chem., 60: 2299-301, 1988). The analyte is embedded in a light absorbing matrix. The matrix is deposited by short laser pulses and carries analyte molecules into the vapor phase in an unfragmented manner. The analyte is ionized by collision with matrix molecules. The load charge accelerates ions to the field-free flight tube. Due to their different masses, the ions are accelerated at different velocities. Smaller ions reach the detector faster than larger ones. MALDI-TOF analyzers are well suited for the analysis of peptides and proteins. Analysis of nucleic acids is somewhat difficult (Gut and Beck, Current Innovations and Future Trends, 1: 147-57, 1995). Sensitivity for nucleic acid analysis is almost 100 times worse than peptides and decreases non-uniformly as fragment size increases. Furthermore, for nucleic acids with multiple negatively charged backbones, the ionization process through the matrix significantly reduces efficiency. In the MALDI-TOF analyzer, the selection of the matrix plays an extremely important role. For peptide desorption, several very effective matrices have been discovered that produce very fine crystals. However, there are currently several DNA-reactive matrices, but the difference in sensitivity between peptides and nucleic acids has not been reduced. However, this difference in sensitivity can be reduced by chemically changing the DNA in a manner more similar to the peptide. For example, phosphorothioate nucleic acids, those in which the normal phosphate of the backbone is replaced by thiophosphate, can be converted to neutral DNA using simple alkylation chemistry (Gut and Beck, Nucleic Acids Res. 23: 1367 -73, 1995). The binding of a charge tag to this modified DNA results in an increase in MALDI-TOF sensitivity to the same level as seen with peptides. A further advantage of charge tagging is that it increases the stability of the analyte to impurities, which makes detection of unmodified substrates significantly difficult.

  In a particularly preferred embodiment of the method, the amplification of step 3 is performed in the presence of at least one blocker oligonucleotide. The use of such blocker oligonucleotides is described in Yu et al., BioTechniques 23: 714-720, 1997. The use of blocking oligonucleotides can improve the specificity of amplification of a subpopulation of nucleic acids. The blocking probe hybridized to the nucleic acid suppresses or prevents polymerase-mediated amplification of the nucleic acid. In one embodiment of the method, the blocking oligonucleotide is designed to hybridize to background DNA. In a further embodiment of the method, the oligonucleotide is designed to prevent or inhibit amplification of unmethylated nucleic acid relative to methylated nucleic acid, or vice versa.

  Blocking probe oligonucleotides are simultaneously hybridized to the bisulfite treated nucleic acid along with PCR primers. Nucleic acid PCR amplification products are terminated at the 5 ′ position of the blocking probe, and thus nucleic acid amplification is suppressed where a sequence complementary to the blocking probe is present. The probe may be designed to hybridize to the bisulfite treated nucleic acid in a methylation state specific manner. For example, for detection of methylated nucleic acids in a population of unmethylated nucleic acids, suppression of amplification of nucleic acids that are not methylated at the location of interest includes “TpG” at the location of interest versus “CpG” Performed by use of a blocking probe. In one embodiment of the method, the sequence of the blocking oligonucleotide is from the group consisting of SEQ ID NOs: 2 to 5, 151, 152, 155 and 156, preferably comprising one or more CpG, TpG, CpA dinucleotides. It is identical or complementary to a selected molecule that is complementary or identical to a sequence of at least 18 base pairs in length. In one embodiment of the method, the sequence of the oligonucleotide is selected from the group consisting of SEQ ID NO: 15 and SEQ ID NO: 16 and sequences complementary thereto.

  For PCR methods using blocking oligonucleotides, efficient interference with polymerase-mediated amplification requires that the blocker oligonucleotide not be extended by the polymerase. Preferably, this is accomplished by the use of a blocker that is a 3 ′ deoxy oligonucleotide or an oligonucleotide derivatized at the 3 ′ position with other than a “free” hydroxy group. For example, 3′-O-acetyl oligonucleotides represent a preferred class of blocker molecules.

  Furthermore, polymerase-mediated degradation of the blocker oligonucleotide is previously prevented. Preferably, such prevention involves either the use of a polymerase that has lost 5′-3 ′ exonuclease activity, or the use of a modified blocker oligonucleotide, eg, having a thioate bridge at the 5 ′ end that exhibits blocker molecule nuclease resistance. Including. Certain applications may not require such blocker 5 'modifications. For example, if the blocker and primer binding sites overlap, thereby preventing primer binding (eg, by excess blocker), degradation of the blocker oligonucleotide is substantially prevented. This is because the polymerase extends the primer toward and through the blocker (in the 5'-3 'direction)-there is usually no step that results in degradation of the hybridized blocker oligonucleotide.

  Particularly preferred blocker / PCR embodiments include the use of peptide nucleic acid (PNA) oligomers as blocking oligonucleotides, as described herein, for the purposes of the present invention. Such PNA blocker oligomers are well suited because they are not degraded or extended by the polymerase.

  In one embodiment of the method, the binding site of the blocking oligonucleotide is the same as or overlaps with that of the primer, thereby preventing hybridization to the binding site of the primer. In a further preferred embodiment of the method, two or more such blocking oligonucleotides are used. In a particularly preferred embodiment, hybridization of one blocking oligonucleotide prevents hybridization of the forward primer, and hybridization of the reverse primer binds to the amplification product of the forward primer by hybridization of the other probe (blocker) oligonucleotide. Prevents hybridization.

  In an alternative embodiment of the method, the blocking oligonucleotide hybridizes to a location between the reverse of the treated background DNA and the position of the forward primer, thereby preventing extension of the primer oligonucleotide.

  It is particularly preferred that the blocking oligonucleotide is present at at least 5 times the primer concentration.

  In the fourth step of the method, the amplification obtained during the third step of the method is analyzed to confirm the methylation status of the CpG dinucleotide prior to treatment.

  In one embodiment where the amplificates are obtained by amplification and / or blocking oligonucleotides obtained by MSP amplification, the presence or absence of the amplificates is essentially covered by primers and / or blocking oligonucleotides according to their base sequence. It is an indicator of the methylation state of the CpG position. All possible known molecular biology methods are possible for this detection, including but not limited to gel electrophoresis, sequencing, liquid chromatography, hybridization, real-time PCR analysis, or combinations thereof . This step of the method further serves as a quantitative control for the previous step.

  In the fourth step of the method, the amplification obtained by both standard and methylation-specific PCR is used to determine the methylation status of CpG of the genomic DNA isolated in the first step of the method. Further analysis. This may be performed by hybridization-based methods such as, but not limited to, array techniques and probe-based techniques, as well as techniques such as sequencing and extension templates.

  In one embodiment of the method, the amplification synthesized in step 3 is subsequently hybridized to a set or array of oligonucleotides and / or PNA probes. In this connection, the hybridization is carried out in the following manner: the set of probes used during the hybridization is preferably composed of at least two oligonucleotides or PNA oligomers; Acting as a probe that hybridizes to the oligonucleotides pre-bound to; non-hybridized fragments are subsequently removed; said oligonucleotides in SEQ ID NO: 2 to SEQ ID NO: 5 and SEQ ID NOs: 151, 152, 155 and 156 Includes at least one base sequence having a length of at least 9 nucleotides, reverse-complementary or identical to a segment of the identified base sequence, and the segment comprises at least one CpG, TpG, or CpA dinucleotide. In a further embodiment, the oligonucleotide comprises at least 9 nucleotides that are reverse-complementary or identical to a segment of the base sequence identified in SEQ ID NOs: 2-5, SEQ ID NOs: 151 to 158, and SEQ ID NOs: 76 to 103. The segment includes at least one base sequence having a length, and the segment includes at least one CpG, TpG, or CpA dinucleotide.

  In a preferred embodiment, the dinucleotide is present in the middle of 1/3 of the oligomer. For example, if the oligomer comprises one CpG dinucleotide, the dinucleotide is the 13th-mer 5 ′ end, preferably the 5th to 9th nucleotide. In a further embodiment, one oligonucleotide is for the analysis of each CpG dinucleotide within the sequence according to SEQ ID NOs 1 and 149 and equivalent positions within SEQ ID NOs 2 to 5 and SEQ ID NOs 151, 152, 155 and 156. Exists. One oligonucleotide is present in each CpG dinucleotide in the sequence according to SEQ ID NO: 1, SEQ ID NO: 149, 150, and SEQ ID NO: 60-75 and in SEQ ID NO: 2-5 and SEQ ID NO: 151-158, SEQ ID NO: 76-103 Exists for analysis of equivalent positions. The oligonucleotide may be present in the form of a peptide nucleic acid. Unhybridized amplification is then removed. Hybridized amplification is detected. In this context, it is preferred that the label attached to the amplification product can correspond to each position on the solid phase where the oligonucleotide sequence is present.

  In yet a further embodiment of the method, the genomic methylation status at the CpG position is confirmed by oligonucleotides that are simultaneously hybridized to the bisulfite treated DNA by PCR amplification primers (the primers are methylation specific). Or standard).

A particularly preferred embodiment of the method is the use of fluorescence-based real-time quantitative PCR (Heid et al., Genome Res. 6: 986-994, 1996; see also US Pat. No. 6,331,393). There are two preferred embodiments that utilize this method. One embodiment known as TaqMan ^™ analysis uses a dual-labeled fluorescent oligonucleotide probe. TaqMan ^™ PCR reactions employ the use of non-extendable responsive oligonucleotides called TaqMan ^™ probes, which are designed to hybridize to CpG rich sequences present between forward and reverse amplification primers. The TaqMan ^™ probe further comprises a fluorescent “reporter moiety” and “quencher moiety” covalently linked to a linker moiety (eg, phosphoramidite) attached to the nucleotide of the TaqMan ^™ oligonucleotide. The hybridized probe is displaced and degraded by the polymerase in the amplification reaction, thereby inducing an increase in fluorescence. For analysis of methylation in nucleic acids following bisulfite treatment, the probe is methylated as described in US Pat. No. 6,331,393, also incorporated herein by reference, also known as MethyLight analysis. It must be specific. The second preferred embodiment of this MethyLight technology is the use of dual probe technology (Lightcycler®), where each probe has a donor or recipient fluorescent moiety and the hybridization of two probes in close proximity to each other is Indicated by increasing or fluorescent amplification primers. Both these techniques may be adapted in a manner suitable for use with bisulfite-treated DNA, and even for methylation analysis within CpG dinucleotides.

  Also, any combination of these probes or a combination of these probes with other known probes may be used.

  In a further preferred embodiment of the method, the fourth step of the method consists of the template for oligonucleotide extension purposes such as MS-SNuPE as described in Gonzalgo and Jones, Nucleic Acids Res. 25: 2529-2531, 1997. Including use. In said embodiment, the methylation specific single nucleotide extension primer (MS-SNuPE primer) is from the group of SEQ ID NO: 2 to SEQ ID NO: 5 and SEQ ID NO: 151, 152, 155 and 156. One or more of the selected sequences, preferably at least 9, but preferably matches or is complementary to a sequence of only 25 nucleotides in length. However, it is also preferred to use fluorescently labeled nucleotides instead of radiolabeled nucleotides.

  In a still further embodiment of the method, the fourth step of the method involves sequencing the amplification generated in the third step of the method followed by sequence analysis (Sanger F. et al., Proc Natl Acad Sci USA 74: 5463-5467, 1977).

In the most preferred embodiment of the methylation analysis method, the genomic nucleic acid is the first three steps outlined above, namely:
a) obtaining a biological sample having the genomic DNA of interest from the subject;
b) extracting or otherwise isolating genomic DNA;
c) the genomic DNA of b) or its by one or more reagents that convert the unmethylated cytosine base at its 5-position to another different base or uracil so that it can be detected by cytosine with respect to hybridization properties Process the fragment; where:
d) Amplification after the treatment of c) is carried out by the use of a methylation specific mode, ie methylation specific primers or blocking oligonucleotides,
e) Detection of amplification is performed with a real-time detection probe as described above:
Isolated and processed.

  Preferably, the subsequent amplification of d) is carried out with a methylation specific primer as described above, said methylation specific primer comprising SEQ ID NO: 2 to SEQ ID NO: 5 and SEQ ID NO: 151, 152, 155 and 156 And a sequence having a length of at least 9 nucleotides that hybridizes to a sequence that has been processed by one and a sequence complementary thereto, wherein the base sequence of the oligomer comprises at least one CpG dinucleotide. In addition, further methylation specific primers are also used for the analysis of the gene panel as described above, said primers according to one of SEQ ID NO: 76 to SEQ ID NO: 103 and SEQ ID NO: 153, 154, 157 and 158 Including a processed nucleic acid sequence and a sequence having a length of at least 9 nucleotides that hybridizes to a sequence complementary thereto, wherein the base sequence of the oligomer comprises at least one CpG dinucleotide.

  In an alternative most preferred embodiment of the method, the subsequent amplification of d) is performed in the presence of a blocking oligonucleotide as described above. The blocking oligonucleotide comprises a sequence having a length of at least 9 nucleotides that hybridizes to the processed nucleic acid sequence according to SEQ ID NO: 2 to SEQ ID NO: 5, SEQ ID NO: 151, 152, 155 and 156 and the sequence complementary thereto. Preferably, the base sequence of the oligomer comprises at least one CpG, TpG, or CpA dinucleotide.

  In addition, additional blocking oligonucleotides may also be used in the analysis of the gene panel as described above, wherein the blocking oligonucleotide is one of SEQ ID NO: 76 to SEQ ID NO: 103 and SEQ ID NO: 153, 154, 157 and 158. And the nucleotide sequence of said oligomer comprises at least one CpG, TpG, or CpA dinucleotide that has a length of at least 9 nucleotides that hybridizes to a sequence complementary thereto.

  Step e) of the method, ie SEQ ID NO: 2-5, SEQ ID NO: 151-158, and SEQ ID NO: 76 to SEQ ID NO: 103, most preferably SEQ ID NO: 2 to SEQ ID NO: 5, and SEQ ID NO: 151, 152, 155 and 156 The detection of the indicator of the specific amplification of the methylation status of one or more CpG positions by is carried out by the real-time detection method described above.

  A further embodiment of the invention provides a method for the analysis of the methylation status of the gene PITX2 and / or their regulatory regions without the need for pretreatment. Further, the method may be used for methylation analysis of gene PITX2 and / or their regulatory regions, and one or more selected from the group comprising ABCA8, CDK6, ERBB2, ONECUT2, PLAU, TBC1D3, TFF1 And / or the methylation status of their regulatory or promoter regions. It is particularly preferred that the methylation status of a gene panel selected from the group of gene panels comprising PITX2, PLAU and TFF1; PITX2 and PLAU; PITX2 and TFF1 is determined.

  In the first step of such additional embodiments, the genomic DNA sample is isolated from a tissue or cell source. Preferably, such source comprises a cell line, histological slide, biopsy tissue, body fluid, or paraffin-embedded breast tumor tissue. Extraction may be by standard means to those skilled in the art including, but not limited to, the use of detergent lysates, sonification and vortexing with glass beads. Once nucleic acid is extracted, genomic duplex DNA is used for analysis.

  In a preferred embodiment, the DNA is degraded prior to treatment, which may be by standard means in the state of the art, but preferably by a methylation sensitive restriction endonuclease.

  In the second step, the DNA is then degraded by one or more methylation sensitive restriction enzymes. Degradation is performed such that the hydrolysis of DNA at the restriction site provides information on the methylation status of specific CpG dinucleotides.

  In the third step, which is an optional but preferred embodiment, the restriction fragments are amplified. This is preferably carried out using the polymerase chain reaction, the amplification having a suitable detectable label, as discussed above, namely a fluorophore label, a radionuclide and a mass label.

  In the fourth step, an amplification product is detected. Detection may be by means that are standard in the prior art, such as, but not limited to, gel electrophoresis analysis, hybridization analysis, incorporation of detectable tags into PCR products, DNA array analysis, MALDI or ESI analysis. Also good.

  In the final step of the method, prognosis and / or prediction of outcome of endocrine therapy is determined. Preferably, the correlation of gene expression levels with endocrine therapy prognosis and / or outcome prediction is performed substantially without human intervention. Poor prognosis and / or prediction of outcome of endocrine therapy is determined by abnormal levels and hypermethylation of mRNA and / or protein. It is particularly preferred that the hypermethylation is above the mean value of the disease or above the median value in the specific setting.

  It is particularly preferred that the sample classification is performed by algorithm means.

  In one embodiment, a machine that learns predictors is trained on the methylation pattern at the studied CpG site of a sample having a known state. A selection of CpG positions that is distinguishable for machines that learn predictors is used for the panel. In a particularly preferred embodiment of the method, both methods are combined; that is, the machine learning the classification factor is at a selected CpG position that is methylated differently between classes by statistical analysis. Only trained.

  The development of an algorithmic method for sample classification based on the methylation status of CpG positions in the panel is shown in the examples.

  The disclosed invention provides a processed nucleic acid from SEQ ID NO: 1, SEQ ID NO: 149, SEQ ID NO: 150, and SEQ ID NO: 60 to SEQ ID NO: 75, wherein the processing comprises at least one non-sequence of the genomic DNA sequence. Suitable for converting methylated cytosine bases to uracil or other bases that are detectably different from cytosine for hybridization. The genomic sequence in question may contain one or more consecutive or random methylated CpG positions. Said treatment preferably comprises the use of a reagent selected from the group consisting of bisulfite, disulfate, hydrogen sulfate and combinations thereof. In a preferred embodiment of the present invention, the object is an adjacent nucleotide base having a sequence length of at least 16 selected from the group consisting of SEQ ID NO: 1, SEQ ID NO: 149, SEQ ID NO: 150 and SEQ ID NO: 60 to SEQ ID NO: 75. Analysis of a non-naturally occurring altered nucleic acid comprising the sequence: wherein said sequence comprises at least one CpG, TpA, or CpA dinucleotide and a sequence complementary thereto. The sequences of SEQ ID NO: 2-5, SEQ ID NO: 151 to SEQ ID NO: 158 and SEQ ID NO: 76 to SEQ ID NO: 103 are non-naturally occurring nucleic acids according to SEQ ID NO: 1, SEQ ID NO: 149, SEQ ID NO: 150 and SEQ ID NO: 60 to SEQ ID NO: 75. The resulting variation is provided, where each genomic sequence change results in the synthesis of a nucleic acid having a sequence that is unique and distinct from said genomic sequence as follows. For each sense strand genomic DNA, eg, SEQ ID NO: 1, four converted forms are disclosed. “C” to “T”, but “CpG” remains “CpG” in the first type (ie, with respect to the genomic sequence, all “C” residues of the “CpG” dinucleotide sequence are methylated. The second type discloses the complement (ie, the antisense strand) of the disclosed genomic DNA sequence, “C” to “T”, but “CpG” is “CpG”. (I.e. corresponding to the case where the sequence is methylated and therefore not converted for all "C" residues of the CpG dinucleotide). The “upmethylated” converted sequences of SEQ ID NO: 1, SEQ ID NO: 149, SEQ ID NO: 150 and SEQ ID NO: 60 to SEQ ID NO: 75 are from SEQ ID NO: 2-5, SEQ ID NO: 151 to SEQ ID NO: 158 and SEQ ID NO: 76 This corresponds to the sequence of SEQ ID NO: 103. A third chemically converted form of each genomic sequence is provided, with “C” to “T” including that of the CpG dinucleotide sequence for all “C” residues (ie, for genomic sequences, CpG Corresponding to the case where all “C” residues of the dinucleotide sequence are not methylated); the last chemically converted form of each sequence is the complement of the disclosed genomic sequence (ie, the antisense strand) ) And “C” to “T” for all “C” residues includes that of the “CpG” dinucleotide sequence (ie, the CpG nucleotide sequence for the complement (antisense strand) of each genomic sequence) Corresponding to the case in which all “C” residues of are not methylated). SEQ ID NO: 1, SEQ ID NO: 149, SEQ ID NO: 150 and SEQ ID NO: 60 to SEQ ID NO: 75 are “downmethylated” converted sequences are SEQ ID NO: 2-5, SEQ ID NO: 151 to SEQ ID NO: 158 and SEQ ID NO: 76 to SEQ ID NO: 103 Corresponding to

  The present invention discloses oligonucleotides or oligomers for detecting cytosine methylation status in the genome or pretreated DNA according to SEQ ID NO: 1, SEQ ID NO: 149 to SEQ ID NO: 158 and SEQ ID NO: 60 to SEQ ID NO: 103. The oligonucleotide or oligomer is a sequence of SEQ ID NO: 2-5, SEQ ID NO: 151 to SEQ ID NO: 158 and SEQ ID NO: 76 under moderately stringent or stringent conditions (as defined herein above). The processed nucleic acid sequence according to No. 103 and / or its complementary sequence, or the genomic sequence according to SEQ ID No. 1, SEQ ID No. 149, SEQ ID No. 150 and SEQ ID No. 60 to SEQ ID No. 75 and / or complementary thereto It comprises a nucleic acid sequence having a length of at least nine (9) oligonucleotides that hybridize to the sequence.

  Accordingly, the present invention provides moderate stringency and / or to all or part of the sequences of SEQ ID NO: 2-5, SEQ ID NO: 151 to SEQ ID NO: 158, and SEQ ID NO: 76 to SEQ ID NO: 103, or their complements. Or nucleic acid molecules that hybridize under stringent hybridization conditions (eg, oligonucleotides and peptide nucleic acid (PNA) molecules (PNA oligomers)). The hybridizing portion of the hybridizing nucleic acid is typically at least 9, 15, 20, 25, 30 or 35 nucleotides in length. However, longer molecules have inventive utility and are therefore within the scope of the present invention.

  Preferably, the hybridizing portion of the hybridizing nucleic acid of the present invention is the sequence of SEQ ID NO: 2-5, SEQ ID NO: 151 to SEQ ID NO: 158 and SEQ ID NO: 76 to SEQ ID NO: 103, a part thereof, or a part thereof It is at least 95%, or at least 98%, or 100% identical to the complement.

  Hybridizing nucleic acids of the type described herein can be used, for example, as primers (eg, PCR primers), or as diagnostic and / or prognostic probes or primers. Preferably, hybridization of the oligonucleotide probe to the nucleic acid sample is performed under stringent conditions, and the probe is 100% identical to the target sequence. Nucleic acid duplex or hybridization stability is expressed as the melting temperature or Tm, which is the temperature at which the probe dissociates from the target DNA. This melting temperature is used to define the required stringency conditions.

  Target nucleic acids (eg, allelic variants and SNPs) that are related to and substantially identical to the corresponding sequences of SEQ ID NO: 1, SEQ ID NO: 149, SEQ ID NO: 150, and SEQ ID NO: 60 to SEQ ID NO: 75, rather than identical In regard to, it is useful to first construct the lowest temperature at which only homozygous hybridization occurs at a particular salt (eg, SSC or SSPE) concentration. Then, assuming that a 1% mismatch results in a 1 ° C. decrease in Tm, the temperature of the last wash in the hybridization reaction is reduced accordingly (eg, if a sequence with> 95% identity with the probe is searched) If done, the temperature of the last wash will decrease by 5 ° C). In fact, the change in Tm can be between 0.5 ° C and 1.5 ° C per 1% mismatch.

An example of an oligonucleotide of the invention of length X (in nucleotides) is a set of consecutively overlapping oligonucleotides of length X, as indicated for example by the position of the polynucleotide as referenced in SEQ ID NO: 1 (Corresponding to a set of sense and antisense), where the oligonucleotides in each successively overlapping set (corresponding to a given X value) are as a limited set of Z oligonucleotides from the nucleotide position Defined:
n to (n + (X-1));
Where n = 1, 2, 3,… (Y- (X-1));
Y is equal to the length of SEQ ID NO: 1 (9001) (nucleotide or base pair);
X is equal to the common length (in nucleotides) of each oligonucleotide in the set (eg, X = 20 for 20-mer sets that are consecutively overlapping); and for length Y of a given SEQ ID NO The number of continuously overlapping oligomers of length X (Z) is equal to Y- (X-1). For example, when X = 20, Z = 9001-19 = 8,982 for either the sense or antisense set of SEQ ID NO: 1.

  Preferably, the set is limited to those oligomers comprising at least one CpG, TpG, or CpA dinucleotide.

  Examples of 20-mer oligonucleotides of the invention include the following set of oligomers (and the set of antisense complementary thereto), polynucleotide positions referenced in SEQ ID NO: 1: 1-20, 2- Indicated by 21, 3-22, 4-23, 5-24, ... and 8,982-9,001.

  Preferably, the set is limited to those oligomers comprising at least one CpG, TpG, or CpA dinucleotide.

  Similarly, examples of 25-mer oligonucleotides of the invention include the following set of oligomers (and a set of antisense complementary thereto), polynucleotide positions referenced in SEQ ID NO: 1: 1-25, It is indicated by 2-26, 3-27, 4-28, 5-29, ......... and 8,977-9,001.

  Preferably, the set is limited to those oligomers comprising at least one CpG, TpG, or CpA dinucleotide.

  The present invention relates to each of SEQ ID NO: 1-5, SEQ ID NO: 149 to SEQ ID NO: 158, and SEQ ID NO: 60 to SEQ ID NO: 103 (sense and antisense), length X, eg, X = 9, 10, 17, 20 , 22, 23, 25, 27, 30 or 35 nucleotides, including multiple consecutively overlapping sets of oligonucleotides or altered oligonucleotides.

  The oligonucleotides or oligomers according to the present invention are effective for confirming genetic and epigenetic parameters of genomic sequences corresponding to SEQ ID NO: 1, SEQ ID NO: 149, SEQ ID NO: 150 and SEQ ID NO: 60 to SEQ ID NO: 75. The right tools. Preferred sets of such oligonucleotides of length X or altered oligonucleotides are those of oligomers corresponding to SEQ ID NO: 1-5, SEQ ID NO: 49 to SEQ ID NO: 158 and SEQ ID NO: 60 to SEQ ID NO: 103 (and their complements). These are consecutively overlapping sets. Preferably, the oligomer comprises at least one CpG, TpG, or CpA dinucleotide.

  Particularly preferred oligonucleotides or oligomers according to the invention are those in which the cytosine of the CpG dinucleotide (or corresponding converted TpG or CpA dinucleotide) sequence is in the middle of one third of the oligonucleotide; The nucleotide is, for example, 13 bases in length, and the CpG, TpG or CpA dinucleotide is located within the 5th to 9th nucleotides from the 5 ′ end.

  The oligonucleotides of the invention can also be altered by chemically linking the oligonucleotide to one or more moieties or conjugates that enhance the activity, stability or detection of the oligonucleotide. Such moieties or conjugates include lipids such as chromophores, fluorophores, cholesterol, cholic acid, thioethers, aliphatic chains, phospholipids, polyamines, polyethylene glycol (PEG), partimine moieties, and, for example, US Patent Nos. 5,514,758, 5,574,142. 5,585,481, 5,587,371, 5,597,696 and others as disclosed in 5,958,773. The probe may also be present in the form of PNA (peptide nucleic acid) with particularly favorable pairing properties. Thus, oligonucleotides may contain other added groups such as peptides, hybridization-induced cleavage agents (Krol et al., BioTechniques 6: 958-976, 1988) or intercalating agents (Zon, Pharm Res. 5: 539-549, 1988). For this purpose, the oligonucleotide may be conjugated to another molecule such as, for example, a chromophore, a fluorophore, a peptide, a hybridization-induced cross-linking agent, a transport agent, a hybridization-induced cleavage agent.

  Oligonucleotides may also contain altered sugar and / or base moieties, as recognized by at least one technique, or may contain altered backbones or unnatural intranucleoside linkages.

  Oligonucleotides or oligomers according to certain embodiments of the invention are typically used in a “set”, which is the genomic sequence SEQ ID NO: 1, SEQ ID NO: 149, SEQ ID NO: 150 and SEQ ID NO: 60 to SEQ ID NO: 75 and Contains at least one oligomer for analysis of each of these complementary sequences of CpG dinucleotides or processed by SEQ ID NO: 2-5, SEQ ID NO: 151 to SEQ ID NO: 158 and SEQ ID NO: 76 to SEQ ID NO: 103 The corresponding CpG, TpG, CpA dinucleotides in the sequence of nucleic acids and their complementary sequences. However, for economic or other factors, it can be expected that it is preferable to analyze the limited selection of CpG dinucleotides within the sequence and that the contents of the set of oligonucleotides will be changed accordingly. .

  Thus, in a particular embodiment, the present invention relates to processed genomic DNA (SEQ ID NO: 2-5, SEQ ID NO: 151 to SEQ ID NO: 158 and SEQ ID NO: 76 to SEQ ID NO: 103) or genomic DNA (SEQ ID NO: 1, SEQ ID NO: 149, a set of at least two (2) (oligonucleotide and / or PNA oligomer) useful for detecting the methylation status of SEQ ID NO: 150 and SEQ ID NO: 60 to SEQ ID NO: 75 and their complementary sequences) I will provide a. These probes allow diagnosis and / or classification of genetic and epigenetic parameters of lung cell proliferation disorders. The set of oligomers can be either in processed genomic DNA (SEQ ID NO: 2-5, SEQ ID NO: 151 to SEQ ID NO: 158 and SEQ ID NO: 76 to SEQ ID NO: 103) or genomic DNA (SEQ ID NO: 1, SEQ ID NO: 149, SEQ ID NO: 150). And SEQ ID NO: 60 to SEQ ID NO: 75 and sequences complementary thereto) may also be used to detect single nucleotide polymorphisms (SNPs).

  In a preferred embodiment, at least one of the set of oligonucleotides, and more preferably all members are bound to the solid phase.

  In a further embodiment, the invention amplifies the DNA sequence of SEQ ID NO: 2-5, SEQ ID NO: 151 to SEQ ID NO: 158 and SEQ ID NO: 76 to SEQ ID NO: 103 and their complementary sequences or segments thereof. A set of at least two (2) oligonucleotides for use as a “primer” oligonucleotide is provided.

  Oligonucleotides are expected to constitute all or part of an “array” or “DNA chip” (ie, an arrangement of different oligonucleotides and / or PNA oligomers bound to a solid phase). Such arrays of different oligonucleotide and / or PNA oligomer sequences can be characterized in that they are arranged on a solid phase, for example in the form of a rectangular or hexagonal lattice. The solid surface may be composed of silicon, glass, polystyrene, aluminum, steel, iron, copper, nickel, silver or gold. Plastics such as nitrocellulose as well as nylon may also be used, which can be present in the form of pellets or as a resin matrix. A summary of the prior art in oligomer array manufacturing can be gathered from a special edition of Nature Genetics (from Nature Genetics Supplement, Volume 21, January 1999, and references cited therein). Fluorescently labeled probes are often used for scanning immobilized DNA arrays. Easy attachment of Cy3 and Cy5 dyes to the 5′-OH of specific probes is particularly suitable for fluorescent labeling. Detection of the fluorescence of the hybridized probe may be performed, for example, with a confocal microscope. In addition to many others, Cy3 and Cy5 dyes are commercially available.

  It is also anticipated that oligonucleotides or their specific sequences may constitute all or part of a “virtual array”, where oligonucleotides or their specific sequences include, for example, a complex mixture of analytes. Used as a “specific factor” as part of or in combination with various populations of unique labeled probes for analysis. Such a method is described in, for example, US 2003/0013091 (US Serial No. 09 / 898,743, published on January 16, 2003). In such a method, sufficient label is collected so that each nucleic acid in the complex mixture (eg, each analyte) is uniquely bound by a unique label and thus detected (each label is directly counted). Resulting in a digital readout of each molecular species in the mixture).

  The described invention further provides compositions useful for providing prognosis and / or prediction of outcome of endocrine therapy in breast cancer patients. The composition is selected from the group comprising at least one nucleic acid of 18 base pairs in length of the segment of the nucleic acid sequence disclosed in SEQ ID NOs: 2 to 5 and SEQ ID NOs: 151, 152, 155 and 156, and: Contains one or more substances: magnesium chloride, dNTP, taq polymerase, bovine serum albumin, oligomers, in particular oligonucleotide or peptide nucleic acid (PNA) oligomers, said oligomers are SEQ ID NO: 2 to SEQ ID NO: 5 and SEQ ID NOs: 151, 152 155 and 156 and at least 9 nucleotides that are complementary or moderately stringent or hybridize under stringent conditions to pretreated genomic DNA with one of their complementary sequences In each case at least one base sequence having a length of Preferably, the composition comprises a buffer solution suitable for stabilizing the nucleic acid in aqueous solution and allowing a polymerase-based reaction in the solution. Suitable buffers are known in the prior art and are commercially available.

  Furthermore, an additional aspect of the invention is a kit comprising, for example, a bisulfite-containing reagent, and an 18 base length of the sequences of SEQ ID NOs: 2 to 5 and SEQ ID NOs: 151, 152, 155 and 156. At least one oligonucleotide having a sequence that hybridizes under stringent or highly stringent conditions, each corresponding to a segment in each case. Said kit is complementary, stringent or high, each corresponding to an 18 base long segment of the sequences SEQ ID NO: 2-5, SEQ ID NO: 151-158 and SEQ ID NO: 76-103. It may further comprise at least one oligonucleotide having a sequence that hybridizes under stringent conditions. The kit may further comprise instructions for performing and evaluating the described method. In a further preferred embodiment, the kit may further comprise standard reagents for performing CpG position-specific methylation analysis, said analysis comprising one or more of the following techniques: MS-SNuPE , MSP, MethyLight (R), HeavyMethyl (R), COBRA, and nucleic acid sequencing. However, a kit in accordance with the principles of the present invention can also include only some of the aforementioned components.

  Typical reagents for COBRA analysis (eg, as can be found in a typical COBRA-based kit) include, but are not limited to: a specific gene (or DNA undergoing methylation changes) PCR primers for sequences, or CpG islands); restriction enzymes and appropriate buffers; gene hybridization oligos; control hybridization oligos; kinase labeling kits for oligonucleotide probes; and radioactive nucleotides. In addition, the bisulfite conversion reagent includes: DNA denaturation buffer; sulfonation buffer; DNA recovery reagent or kit (eg, precipitation, ultrafiltration, affinity column); desulfonation buffer; and DNA recovery components.

  Typical reagents for MethyLight® analysis (eg, as can be found in a typical MethyLight®-based kit) include, but are not limited to: a specific gene (or PCR primers for DNA sequences undergoing methylation changes, or CpG islands); TaqMan® probes; optimal PCR buffers and deoxynucleotides; and Taq polymerase.

  Typical reagents for Ms-SNuPE analysis (eg, as can be found in a typical Ms-SNuPE-based kit) include, but are not limited to: PCR primers for specific genes ( Or DNA sequence undergoing methylation change, or CpG island); optimal PCR buffer and deoxynucleotide; gel extraction kit; positive control primer; Ms-SNuPE primer for specific gene; reaction buffer (for Ms-SNuPE reaction) ); And radioactive nucleotides. In addition, the bisulfite conversion reagent includes: DNA denaturation buffer; sulfonation buffer; DNA recovery reagent or kit (eg, precipitation, ultrafiltration, affinity column); desulfonation buffer; and DNA recovery components.

  Typical reagents for MSP analysis (eg, as can be found in a typical MSP-based kit) include, but are not limited to: a specific gene (or DNA sequence that has undergone a methylation change) , Or CpG islands) methylated and unmethylated PCR primers; optimal PCR buffer and deoxynucleotides; and specific probes.

  Although the invention has been described with specificity in accordance with certain preferred embodiments thereof, the following examples and drawings serve only to illustrate the invention and fall within the broadest description and the principles and scope of their equivalent constructions. It is not intended to limit the invention.

  SEQ ID NOs: 6-9 provide primer and probe nucleic acid sequences that are useful in predicting the survival of breast cancer patients according to the present invention as described in Example 4.

  SEQ ID NOs: 10-12 provide primer and probe nucleic acid sequences with the control gene used in Examples 4 and 5.

  SEQ ID NO: 13 provides a subsequence of SEQ ID NO: 1, which represents the nucleic acid sequence of the human gene PITX2.

  SEQ ID NOs: 14-17 provide primer and probe nucleic acid sequences that are useful in predicting the survival of breast cancer patients according to the present invention as described in Example 5.

  SEQ ID NO: 21 provides the nucleic acid sequence of the polypeptide encoded by gene PITX2. The amino acid sequence of the polypeptide encoded by PITX2 is also shown in FIG.

(Drawing)
FIG. 1 shows a simplified model of stage 1-3 breast tumor, where the first treatment is surgery (point 1) followed by adjuvant therapy with tamoxifen as an example of endocrine treatment. The Y axis represents the amount (or size) of the tumor (s), and the line “3” indicates the detectable limit of the tumor amount. The X axis represents time. In the first scenario, patients without recurrence during endocrine therapy (4) are shown to remain below detectable limits during the duration of observation. Patients with recurrence of cancer (5) have a duration of disease-free survival (2), and recurrence continues when the amount of carcinoma reaches a detectable level.

  FIG. 2 shows the results of the analysis described in Example 4 (QM analysis): curve of metastasis-free survival by Kaplan-Meier estimation for three CpG sites of the PITX2 gene by real-time methylation specific probe analysis (QM analysis). The lower curve shows the percentage of patients without metastasis in the population with a methylation level above the median, and the upper curve shows the percentage of patients without metastasis in the population with a methylation level below the median. The X axis shows the time of patient metastasis survival in months, and the Y axis shows the percentage of the patient population with metastasis survival.

  FIG. 3 shows the results of chip hybridization described in Example 2. Metastasis-free survival curve by Kaplan-Meier estimation for two CpG positions of the PITX2 gene by methylation specific detection oligonucleotide hybridization analysis. The lower curve shows the percentage of patients without metastasis in the population with a methylation level above the median, and the upper curve shows the percentage of patients without metastasis in the population with a methylation level below the median. The X axis shows the time of patient metastasis survival in months, and the Y axis shows the percentage of the patient population with metastasis survival.

  FIG. 4 shows a metastasis-free survival curve by Kaplan-Meier estimation for the two CpG positions of the PITX2 gene by methylation-specific detection oligonucleotide hybridization analysis. The lower line shows the percentage of patients with no metastasis in a population of 55 patients with a methylation level above the median, and the upper curve is without metastasis in a population of 54 patients with a methylation level below the median Shows the percentage of patients. The X-axis shows the time of patient metastasis-free survival in years, and the Y-axis shows the percentage of patients with metastasis-free survival in%. This was obtained from the first set of data achieved in the first study.

  FIG. 5 shows metastasis-free survival curves according to Kaplan-Meier estimation for 6 different CpG positions located within the preferred region of the PITX2 gene (SEQ ID NO: 13) by methylation specific detection oligonucleotide hybridization analysis. The lower line shows the percentage of patients with no metastasis in the population of 118 patients with a methylation level above the median, and the upper curve has no metastasis in the population of 118 patients with a methylation level below the median Shows the percentage of patients. The X-axis shows the time of patient metastasis-free survival in years, and the Y-axis shows the percentage of patients with metastasis-free survival in%. This was obtained from the second set of data achieved in the second study.

  FIG. 6 shows metastasis-free survival curves according to Kaplan-Meier estimates for 6 different CpG positions within the preferred region of the PITX2 gene (SEQ ID NO: 13) by methylation specific detection oligonucleotide hybridization analysis. This time, only a sub-population of 148 patients characterized as grade G1 or G2 tumors were analyzed: the lower curve is patients with no metastasis in a population of 74 patients with a methylation level above the median The upper curve shows the percentage of patients with no metastasis in the 74 patient population with a methylation level below the median. The X-axis shows the time of patient metastasis-free survival in years, and the Y-axis shows the percentage of patients with metastasis-free survival in%. This was obtained from a second set of data as shown in Example 2.

  FIG. 7 shows metastasis-free survival curves according to Kaplan-Meier estimation for four different CpG positions within the preferred region of the PITX2 gene (SEQ ID NO: 13) by methylation specific detection oligonucleotide hybridization analysis. This time, only a subpopulation of 224 patients characterized as stage 1 or 2 (T1 or T2) tumors were analyzed: the lower curve is a population of 112 patients with a methylation level above the median The percentage of patients with no metastasis is shown, and the upper curve shows the percentage of patients with no metastasis in the 112 patient population with a methylation level below the median. The X-axis shows the time of patient metastasis-free survival in years, and the Y-axis shows the percentage of patients with metastasis-free survival in%. This was obtained from the second set of data achieved in the second example.

  FIG. 8 shows the disease free survival curve of a combination of two oligonucleotides each from genes TBC1D3 and CDK6 and one oligonucleotide from gene PITX2 covering two CpG sites. The black curve shows the proportion of disease free patients in the population with a methylation score above the median, and the gray curve shows the proportion of disease free patients in the population with a methylation score below the median.

  FIG. 9 shows a plot according to FIG. 8, showing the classification of the samples set by the St. Gallen method. The unbroken line represents the methylation analysis, the black curve shows the proportion of disease free patients in the population with a methylation score above the median, and the gray curve in the population with a methylation score below the median The percentage of patients without disease is shown. The dashed line represents the St. Gallen classification of the sample set, the black curve indicates the disease-free survival time of the high-risk population, and the gray curve indicates the disease-free survival time of the low-risk population.

  FIG. 10 illustrates the amino acid sequence of the polypeptide encoded by the gene PITX2.

  FIG. 11 illustrates the position of the amplification sequenced in Example 7. “A” shows a diagram of the gene annotated with major exons, “B” shows the annotated mRNA transcript variant, and “C” shows the CpG rich region of the gene. The positions of the amplification products 1 to 11 are shown on the right side of the figure.

  FIG. 12 shows sequencing data for 11 amplified products of gene PITX2 according to Example 7. Each column in the matrix of columns “A” and “B” shows the sequencing data for one amplification. Amplification numbers are shown to the left of the matrix. Each row of the matrix represents a single CpG site within the fragment, and each column represents an individual DNA sample. The matrix of the column marked “A” shows methylation below the median value measured by QM analysis, and the matrix of the column marked “B” was measured by QM analysis Shows methylation below median.

  The left vertical bar represents the percentage methylation scale by the degree of methylation represented by the darkness of each position in the light gray column representing black to 0% methylation representing 100% methylation. White positions represent measurements for which no data was available.

  FIG. 13 shows a schematic representation of a transcriptional variant of PITX2 mRNA as annotated in the online Ensembl database.

  FIG. 14 shows a disease-free survival curve by Kaplan-Meier estimation for the CpG position of the ERBB2 gene by real-time methylation-specific probe analysis according to Example 8. The X axis shows the time of disease free survival of patients in years, and the Y axis shows the percentage of patients with disease free survival. The black plot shows the proportion of patients with no disease in the population with the optimal cut-off point methylation level, and the gray plot has no disease in the population with the optimal cut-off point methylation level Shows the percentage of patients.

  FIG. 15 shows a curve of metastasis-free survival by Kaplan-Meier estimation for CpG position of ERBB2 gene by real-time methylation specific probe analysis according to Example 8. The X axis shows the time of disease free survival of patients in years, and the Y axis shows the percentage of patients with disease free survival. The black plot shows the proportion of patients with no disease in the population with the optimal cut-off point methylation level, and the gray plot has no disease in the population with the optimal cut-off point methylation level Shows the percentage of patients.

  FIG. 16 shows a curve of metastasis-free survival by Kaplan-Meier estimation for CpG position of TFF1 gene by real-time methylation specific probe analysis according to Example 8. The X axis shows the time of disease free survival of patients in years, and the Y axis shows the percentage of patients with disease free survival. The black plot shows the proportion of patients with no disease in the population with the optimal cut-off point methylation level, and the gray plot has no disease in the population with the optimal cut-off point methylation level Shows the percentage of patients.

  FIG. 17 shows a metastasis-free survival curve by Kaplan-Meier estimation for the CpG position of the TFF1 gene by real-time methylation specific probe analysis according to Example 8. The X axis shows the time of disease free survival of patients in years, and the Y axis shows the percentage of patients with disease free survival. The black plot shows the proportion of patients with no disease in the population with the optimal cut-off point methylation level, and the gray plot has no disease in the population with the optimal cut-off point methylation level Shows the percentage of patients.

  FIG. 18 shows a disease-free survival curve by Kaplan-Meier estimation for CpG position of PLAU gene by real-time methylation specific probe analysis according to Example 8. The X axis shows the time of disease free survival of patients in years, and the Y axis shows the percentage of patients with disease free survival. The black plot shows the proportion of patients with no disease in the population with the optimal cut-off point methylation level, and the gray plot has no disease in the population with the optimal cut-off point methylation level Shows the percentage of patients.

  FIG. 19 shows a curve of metastasis-free survival by Kaplan-Meier estimation for CpG position of PLAU gene by real-time methylation specific probe analysis according to Example 8. The X axis shows the time of disease free survival of patients in years, and the Y axis shows the percentage of patients with disease free survival. The black plot shows the proportion of patients with no disease in the population with the optimal cut-off point methylation level, and the gray plot has no disease in the population with the optimal cut-off point methylation level Shows the percentage of patients.

  FIG. 20 shows a disease-free survival curve by Kaplan-Meier estimation for the CpG position of the PITX2 gene by real-time methylation specific probe analysis according to Example 8. The X axis shows the time of disease free survival of patients in years, and the Y axis shows the percentage of patients with disease free survival. The black plot shows the proportion of patients with no disease in the population with the optimal cut-off point methylation level, and the gray plot has no disease in the population with the optimal cut-off point methylation level Shows the percentage of patients.

  FIG. 21 shows a curve of metastasis-free survival by Kaplan-Meier estimation for CpG position of PITX2 gene by real-time methylation specific probe analysis according to Example 8. The X axis shows the time of disease free survival of patients in years, and the Y axis shows the percentage of patients with disease free survival. The black plot shows the proportion of patients with no disease in the population with the optimal cut-off point methylation level, and the gray plot has no disease in the population with the optimal cut-off point methylation level Shows the percentage of patients.

  FIG. 22 shows a disease-free survival curve by Kaplan-Meier estimation for the CpG position of the TBC1D3 gene by real-time methylation specific probe analysis according to Example 8. The X axis shows the time of disease free survival of patients in years, and the Y axis shows the percentage of patients with disease free survival. The black plot shows the proportion of patients with no disease in the population with the optimal cut-off point methylation level, and the gray plot has no disease in the population with the optimal cut-off point methylation level Shows the percentage of patients.

  FIG. 23 shows a metastasis-free survival curve by Kaplan-Meier estimation for the CpG position of the TBC1D3 gene by real-time methylation specific probe analysis according to Example 8. The X axis shows the time of disease free survival of patients in years, and the Y axis shows the percentage of patients with disease free survival. The black plot shows the proportion of patients with no disease in the population with the optimal cut-off point methylation level, and the gray plot has no disease in the population with the optimal cut-off point methylation level Shows the percentage of patients.

  FIG. 24 shows a disease-free survival curve by Kaplan-Meier estimation for the CpG position of the ERBB2 gene by real-time methylation-specific probe analysis according to Example 8. The X axis shows the time of disease free survival of patients in years, and the Y axis shows the percentage of patients with disease free survival. The black plot shows the proportion of patients with no disease in the population with the optimal cut-off point methylation level, and the gray plot has no disease in the population with the optimal cut-off point methylation level Shows the percentage of patients.

  FIG. 25 shows a curve of the metastasis-free survival by Kaplan-Meier estimation for the CpG position of the ERBB2 gene by real-time methylation specific probe analysis according to Example 8. The X axis shows the time of disease free survival of patients in years, and the Y axis shows the percentage of patients with disease free survival. The black plot shows the proportion of patients with no disease in the population with the optimal cut-off point methylation level, and the gray plot has no disease in the population with the optimal cut-off point methylation level Shows the percentage of patients.

  FIG. 26 shows a disease-free survival curve by Kaplan-Meier estimation for the CpG position of the TFF1 gene by real-time methylation-specific probe analysis according to Example 8. The X axis shows the time of disease free survival of patients in years, and the Y axis shows the percentage of patients with disease free survival. The black plot shows the proportion of patients with no disease in the population with the optimal cut-off point methylation level, and the gray plot has no disease in the population with the optimal cut-off point methylation level Shows the percentage of patients.

  FIG. 27 shows a metastasis-free survival curve by Kaplan-Meier estimation for the CpG position of the TFF1 gene by real-time methylation specific probe analysis according to Example 8. The X axis shows the time of disease free survival of patients in years, and the Y axis shows the percentage of patients with disease free survival. The black plot shows the proportion of patients with no disease in the population with the optimal cut-off point methylation level, and the gray plot has no disease in the population with the optimal cut-off point methylation level Shows the percentage of patients.

  FIG. 28 shows a disease-free survival curve by Kaplan-Meier estimation for CpG position of PLAU gene by real-time methylation specific probe analysis according to Example 8. The X axis shows the time of disease free survival of patients in years, and the Y axis shows the percentage of patients with disease free survival. The black plot shows the proportion of patients with no disease in the population with the optimal cut-off point methylation level, and the gray plot has no disease in the population with the optimal cut-off point methylation level Shows the percentage of patients.

  FIG. 29 shows a curve of metastasis-free survival by Kaplan-Meier estimation for CpG position of PLAU gene by real-time methylation specific probe analysis according to Example 8. The X axis shows the time of disease free survival of patients in years, and the Y axis shows the percentage of patients with disease free survival. The black plot shows the proportion of patients with no disease in the population with the optimal cut-off point methylation level, and the gray plot has no disease in the population with the optimal cut-off point methylation level Shows the percentage of patients.

  FIG. 30 shows a disease-free survival curve by Kaplan-Meier estimation for the CpG position of the PITX gene by real-time methylation specific probe analysis according to Example 8. The X axis shows the time of disease free survival of patients in years, and the Y axis shows the percentage of patients with disease free survival. The black plot shows the proportion of patients with no disease in the population with the optimal cut-off point methylation level, and the gray plot has no disease in the population with the optimal cut-off point methylation level Shows the percentage of patients.

  FIG. 31 shows a curve of metastasis-free survival by Kaplan-Meier estimation for CpG position of PITX gene by real-time methylation specific probe analysis according to Example 8. The X axis shows the time of disease free survival of patients in years, and the Y axis shows the percentage of patients with disease free survival. The black plot shows the proportion of patients with no disease in the population with the optimal cut-off point methylation level, and the gray plot has no disease in the population with the optimal cut-off point methylation level Shows the percentage of patients.

  FIG. 32 shows a disease-free survival curve by Kaplan-Meier estimation for CpG position of PITX gene by real-time methylation specific probe analysis according to Example 8. The X axis shows the time of disease free survival of patients in years, and the Y axis shows the percentage of patients with disease free survival. The black plot shows the proportion of patients with no disease in the population with the optimal cut-off point methylation level, and the gray plot has no disease in the population with the optimal cut-off point methylation level Shows the percentage of patients.

  FIG. 33 shows a curve of metastasis-free survival by Kaplan-Meier estimation for CpG position of PITX gene by real-time methylation specific probe analysis according to Example 8. The X axis shows the time of disease free survival of patients in years, and the Y axis shows the percentage of patients with disease free survival. The black plot shows the proportion of patients with no disease in the population with the optimal cut-off point methylation level, and the gray plot has no disease in the population with the optimal cut-off point methylation level Shows the percentage of patients.

  FIG. 34 shows a disease-free survival curve by Kaplan-Meier estimation for the CpG position of the ONECUT gene by real-time methylation specific probe analysis according to Example 8. The X axis shows the time of disease free survival of patients in years, and the Y axis shows the percentage of patients with disease free survival. The black plot shows the proportion of patients with no disease in the population with the optimal cut-off point methylation level, and the gray plot has no disease in the population with the optimal cut-off point methylation level Shows the percentage of patients.

  FIG. 35 shows a curve of metastasis-free survival by Kaplan-Meier estimation for CpG position of ONECUT gene by real-time methylation specific probe analysis according to Example 8. The X axis shows the time of disease free survival of patients in years, and the Y axis shows the percentage of patients with disease free survival. The black plot shows the proportion of patients with no disease in the population with the optimal cut-off point methylation level, and the gray plot has no disease in the population with the optimal cut-off point methylation level Shows the percentage of patients.

  FIG. 36 shows a disease-free survival curve by Kaplan-Meier estimation for the CpG position of the TBC1D3 gene by real-time methylation specific probe analysis according to Example 8. The X axis shows the time of disease free survival of patients in years, and the Y axis shows the percentage of patients with disease free survival. The black plot shows the proportion of patients with no disease in the population with the optimal cut-off point methylation level, and the gray plot has no disease in the population with the optimal cut-off point methylation level Shows the percentage of patients.

  FIG. 37 shows a curve of metastasis-free survival by Kaplan-Meier estimation for CpG position of TBC1D3 gene by real-time methylation specific probe analysis according to Example 8. The X axis shows the time of disease free survival of patients in years, and the Y axis shows the percentage of patients with disease free survival. The black plot shows the proportion of patients with no disease in the population with the optimal cut-off point methylation level, and the gray plot has no disease in the population with the optimal cut-off point methylation level Shows the percentage of patients.

  FIG. 38 shows a disease-free survival curve by Kaplan-Meier estimation for the CpG position of ABCA8 gene by real-time methylation specific probe analysis according to Example 8. The X axis shows the time of disease free survival of patients in years, and the Y axis shows the percentage of patients with disease free survival. The black plot shows the proportion of patients with no disease in the population with the optimal cut-off point methylation level, and the gray plot has no disease in the population with the optimal cut-off point methylation level Shows the percentage of patients.

  FIG. 39 shows a curve of metastasis-free survival by Kaplan-Meier estimation for the CpG position of ABCA8 gene by real-time methylation specific probe analysis according to Example 8. The X axis shows the time of disease free survival of patients in years, and the Y axis shows the percentage of patients with disease free survival. The black plot shows the proportion of patients with no disease in the population with the optimal cut-off point methylation level, and the gray plot has no disease in the population with the optimal cut-off point methylation level Shows the percentage of patients.

  FIG. 40 shows a disease-free survival curve by Kaplan-Meier estimation for the CpG position of the combination of TFF1 and PLAU genes by real-time methylation specific probe analysis according to Example 8. The X axis shows the time of disease free survival of patients in years, and the Y axis shows the percentage of patients with disease free survival. The black plot shows the proportion of patients with no disease in the population with the optimal cut-off point methylation level, and the gray plot has no disease in the population with the optimal cut-off point methylation level Shows the percentage of patients.

  FIG. 41 shows a curve of metastasis-free survival by Kaplan-Meier estimation for CpG position of the combination of TFF1 and PLAU gene by real-time methylation specific probe analysis according to Example 8. The X axis shows the time of disease free survival of patients in years, and the Y axis shows the percentage of patients with disease free survival. The black plot shows the proportion of patients with no disease in the population with the optimal cut-off point methylation level, and the gray plot has no disease in the population with the optimal cut-off point methylation level Shows the percentage of patients.

  FIG. 42 shows disease-free survival curves according to Kaplan-Meier estimation for CpG positions of TFF1, PLAU and PITX gene combinations by real-time methylation specific probe analysis according to Example 8. The X axis shows the time of disease free survival of patients in years, and the Y axis shows the percentage of patients with disease free survival. The black plot shows the proportion of patients with no disease in the population with the optimal cut-off point methylation level, and the gray plot has no disease in the population with the optimal cut-off point methylation level Shows the percentage of patients.

  FIG. 43 shows a metastasis-free survival curve according to Kaplan-Meier estimation for the CpG position of the combination of TFF1, PLAU and PITX genes by real-time methylation specific probe analysis according to Example 8. The X axis shows the time of disease free survival of patients in years, and the Y axis shows the percentage of patients with disease free survival. The black plot shows the proportion of patients with no disease in the population with the optimal cut-off point methylation level, and the gray plot has no disease in the population with the optimal cut-off point methylation level Shows the percentage of patients.

  FIG. 44 shows a disease-free survival curve according to Kaplan-Meier estimation for the CpG position of the PITX and TFF1 gene combination by real-time methylation specific probe analysis according to Example 8. The X axis shows the time of disease free survival of patients in years, and the Y axis shows the percentage of patients with disease free survival. The black plot shows the proportion of patients with no disease in the population with the optimal cut-off point methylation level, and the gray plot has no disease in the population with the optimal cut-off point methylation level Shows the percentage of patients.

  FIG. 45 shows a curve of metastasis-free survival by Kaplan-Meier estimation for CpG position of the combination of PITX and TFF1 genes by real-time methylation specific probe analysis according to Example 8. The X axis shows the time of disease free survival of patients in years, and the Y axis shows the percentage of patients with disease free survival. The black plot shows the proportion of patients with no disease in the population with the optimal cut-off point methylation level, and the gray plot has no disease in the population with the optimal cut-off point methylation level Shows the percentage of patients.

  FIG. 46 shows a disease-free survival curve by Kaplan-Meier estimation for the CpG position of the combination of PITX and PLAU genes by real-time methylation specific probe analysis according to Example 8. The X axis shows the time of disease free survival of patients in years, and the Y axis shows the percentage of patients with disease free survival. The black plot shows the proportion of patients with no disease in the population with the optimal cut-off point methylation level, and the gray plot has no disease in the population with the optimal cut-off point methylation level Shows the percentage of patients.

  FIG. 47 shows a metastasis-free survival curve according to Kaplan-Meier estimation for the CpG position of the combination of PITX and PLAU genes by real-time methylation specific probe analysis according to Example 8. The X axis shows the time of disease free survival of patients in years, and the Y axis shows the percentage of patients with disease free survival. The black plot shows the proportion of patients with no disease in the population with the optimal cut-off point methylation level, and the gray plot has no disease in the population with the optimal cut-off point methylation level Shows the percentage of patients.

  FIG. 48 shows a disease-free survival curve by Kaplan-Meier estimation for the CpG position of the combination of TFF1 and PLAU genes by real-time methylation specific probe analysis according to Example 8. The X axis shows the time of disease free survival of patients in years, and the Y axis shows the percentage of patients with disease free survival. The black plot shows the proportion of patients with no disease in the population with the optimal cut-off point methylation level, and the gray plot has no disease in the population with the optimal cut-off point methylation level Shows the percentage of patients.

  FIG. 49 shows a curve of metastasis-free survival by Kaplan-Meier estimation for CpG position of the combination of TFF1 and PLAU gene by real-time methylation specific probe analysis according to Example 8. The X axis shows the time of disease free survival of patients in years, and the Y axis shows the percentage of patients with disease free survival. The black plot shows the proportion of patients with no disease in the population with the optimal cut-off point methylation level, and the gray plot has no disease in the population with the optimal cut-off point methylation level Shows the percentage of patients.

  FIG. 50 shows a disease-free survival curve according to Kaplan-Meier estimation for the CpG position of the combination of TFF1, PLAU and PITX genes by real-time methylation specific probe analysis according to Example 8. The X axis shows the time of disease free survival of patients in years, and the Y axis shows the percentage of patients with disease free survival. The black plot shows the proportion of patients with no disease in the population with the optimal cut-off point methylation level, and the gray plot has no disease in the population with the optimal cut-off point methylation level Shows the percentage of patients.

  FIG. 51 shows a curve of metastasis-free survival by Kaplan-Meier estimation for CpG positions of TFF1, PLAU and PITX gene combinations by real-time methylation specific probe analysis according to Example 8. The X axis shows the time of disease free survival of patients in years, and the Y axis shows the percentage of patients with disease free survival. The black plot shows the proportion of patients with no disease in the population with the optimal cut-off point methylation level, and the gray plot has no disease in the population with the optimal cut-off point methylation level Shows the percentage of patients.

  FIG. 52 shows a disease-free survival curve by Kaplan-Meier estimation for the CpG position of the combination of PITX and TFF1 genes by real-time methylation specific probe analysis according to Example 8. The X axis shows the time of disease free survival of patients in years, and the Y axis shows the percentage of patients with disease free survival. The black plot shows the proportion of patients with no disease in the population with the optimal cut-off point methylation level, and the gray plot has no disease in the population with the optimal cut-off point methylation level Shows the percentage of patients.

  FIG. 53 shows a curve of metastasis-free survival by Kaplan-Meier estimation for the CpG position of the combination of PITX and TFF1 genes by real-time methylation specific probe analysis according to Example 8. The X axis shows the time of disease free survival of patients in years, and the Y axis shows the percentage of patients with disease free survival. The black plot shows the proportion of disease free patients in the population with above the optimal cut-off point methylation level.

  FIG. 54 shows a disease-free survival curve according to Kaplan-Meier estimation for the CpG position of the combination of PITX and PLAU genes by real-time methylation specific probe analysis according to Example 8. The X axis shows the time of disease free survival of patients in years, and the Y axis shows the percentage of patients with disease free survival. The black plot shows the proportion of patients with no disease in the population with the optimal cut-off point methylation level, and the gray plot has no disease in the population with the optimal cut-off point methylation level Shows the percentage of patients.

  FIG. 55 shows a metastasis-free survival curve by Kaplan-Meier estimation for the CpG position of the combination of PITX and PLAU genes by real-time methylation specific probe analysis according to Example 8. The X-axis shows the time of patient disease-free survival in years, and the Y-axis shows the percentage of patients with metastasis-free survival. The black plot shows the proportion of patients without metastasis in the population with the optimal cut-off point methylation level, and the gray plot is disease-free in the population with the optimal cut-off point methylation level Shows the percentage of patients.

  56 is a scatter diagram of matched pair PET and fresh frozen tissue analyzed using PITX2 gene analysis 1 according to Example 8. FIG. Quantitative methylated CT scores for PET samples are shown on the Y axis, quantitative methylated CT scores for fresh frozen samples are shown on the X axis, and the correlation between paired samples is 0.81 (Spearman's rho) . This analysis is based on n = 89 samples.

  FIG. 57 shows the disease free survival (DFS) of a randomly selected ER +, N0, untreated patient population in a Kaplan-Meier survival plot according to Example 8. The percentage of patients without disease is shown on the Y axis and the time in years is shown on the X axis. 139 events were observed (observed event rate = 33%). Disease-free survival after 5 years: 74.5% [70.3%, 78.9%], and after 10 years 59.8% [54.2%, 66%]. 95% confidence intervals are plotted.

  FIG. 58 shows the distribution of follow-up periods in the ER +, N0, untreated population according to Example 8. The frequency is shown on the Y axis and the time in months is shown on the X axis. The figure on the left shows a patient with an event (all types of recurrence). Average follow-up time was 45.8 months (standard deviation = 31), median = 38 (range = [2, 123]). The right figure shows a censored patient. Average follow-up was 93 months (standard deviation = 35.6), median = 94 (range = [1, 190]).

  FIG. 59 shows disease free survival (DFS) of the ER +, N0, TAM treated population in a Kaplan-Meier survival plot according to Example 8. The percentage of patients without disease is shown on the Y axis and the time in years is shown on the X axis. 56 events were observed (observed event rate = 10%). DFS 5 years later: 92.4% [90%, 94.9%], 10 years later: 82.1% [77.3%, 87.2%]. 95% confidence intervals are plotted.

FIG. 60 shows the distribution of follow-up periods in the ER +, N0, untreated population according to Example 8. The frequency is shown on the Y axis and the time in months is shown on the X axis. The figure on the left shows a patient with all events (all types of recurrence). Average follow-up time was 47.9 months (standard deviation = 24.4), median = 45 (range = [2,98]).
The right figure shows a censored patient. The mean follow-up period was 65.3 months (standard deviation = 31.6), median = 64 (range = [0, 158]).

  FIG. 61 shows ROC plots at different times for marker models 3522 (Analysis 1) and 2265 in the ER + N0 TAM treated population according to Example 8. Figure A shows a plot at 60 months, Figure B shows a plot at 72 months, Figure C shows a plot at 84 months, and Figure D shows a plot at 96 months. Only distant metastases are defined as events. Sensitivity (the proportion of all relapsed patients in the poor prognostic group) is shown on the X-axis and specificity (the proportion of all non-relapsed patients in the good prognostic group) is shown on the Y-axis, these are KM estimates And the estimated area (AUC) under the curve is calculated. The values for the median cutoff (triangle) and the best cutoff (diamond, 0.32 quantile) are plotted.

  FIG. 62 shows ROC plots at different times for only marker model 3522 (analysis 1) in the ER + N0 TAM treated population according to Example 8. Figure A shows a plot for 60 months, Figure B shows a plot for 72 months, Figure C shows a plot for 84 months, and Figure D shows a plot for 96 months. Only distant metastases are defined as events. Sensitivity (percentage of all relapsed patients in the poor prognostic group) is shown on the X-axis and specificity (percentage of all non-relapsed patients in the good prognostic group) is shown on the Y-axis, which are based on KM estimates And the estimated area under the curve (AUC) is calculated. The values for the median cutoff (triangle) and the best cutoff (diamond, 0.42 quantile) are plotted.

  FIG. 63 shows ROC plots at different times for marker model 2265 in the ER + N0 TAM treated population according to Example 8. Figure A shows a plot for 60 months, Figure B shows a plot for 72 months, Figure C shows a plot for 84 months, and Figure D shows a plot for 96 months. Only distant metastases are defined as events. Sensitivity (the proportion of all relapsed patients in the poor prognosis group) is shown on the X axis, and specificity (the proportion of all non-relapsed patients in the good prognosis group) is shown on the Y axis, which are different thresholds ( = 5, 6, 7, 8 years) is calculated from the KM estimate, and the estimated area under the curve (AUC) is calculated. The values for the median cutoff (triangle) and the best cutoff (diamond, 0.78 quantile) are plotted.

  FIG. 64 shows ROC plots at different times for marker model 2395 in the ER + N0 TAM treated population according to Example 8. Figure A shows a plot for 60 months, Figure B shows a plot for 72 months, Figure C shows a plot for 84 months, and Figure D shows a plot for 96 months. Only distant metastases are defined as events. Sensitivity (the proportion of all relapsed patients in the poor prognosis group) is shown on the X axis, and specificity (the proportion of all non-relapsed patients in the good prognosis group) is shown on the Y axis, which are different thresholds ( = 5, 6, 7, 8 years) is calculated from the KM estimate, and the estimated area under the curve (AUC) is calculated. The values for the median cutoff (triangle) and the best cutoff (diamond, 0.77 quantile) are plotted.

(Example 1: Study 1)
The first study is based on a population of 109 patients, which includes patients with both N0 and N + lymph node status. All patients were ER + (estrogen receptor positive). All patients received tamoxifen monotherapy immediately after surgery or diagnosis. Samples were analyzed using Applicant's chip technology with two chip panels representing 117 candidate genes. For further details, see the examples of published patent applications WO 04/035803 and EP 03 090 432.0, which are hereby incorporated by reference. In this study, one of the most significant marker genes was PITX2. The methylation status of PITX2, which encodes a transcription factor, was statistically significantly correlated with disease-free survival in patients receiving adjuvant tamoxifen treatment. This was calculated using a Cox regression model taking into account the condition of the patient's lymph nodes at the time of diagnosis.

  The results from this study-for PITX2-are shown in FIG. The X-axis shows the time of patient metastasis-free survival in years, and the Y-axis shows the percentage of patients with metastasis-free survival in%. Among 54 patients with a methylation level below the median (upper line), some patients with a metastasis-free survival time significantly longer than 55 patients with a methylation level above the median (lower line) Have As shown in the results, patients with no metastasis after 10 years of surgery combined with tamoxifen monotherapy are less than 60% of patients with methylation above the median of PITX2, compared with PITX2. More than 75% of patients with methylation below the median.

  Since the survival time of breast cancer patients is known to be related to the patient's lymph node status, the marker discriminating power in this mixed population is expected to be smaller in the homogeneous population.

  Other studies analyzed whether the same marker could be independently identified in a completely different set of patient samples and discriminate against predicting survival for a subgroup of patients all N0 Was done to characterize.

(Example 2: Study 2)
The second study was based on samples from 236 patients from 5 different sample donors, all of whom were N0 (negative for lymph node status), older than 35 years. In all cases, surgery was performed before 1988. All patients were ER + (estrogen receptor positive) and tumors were graded as T1-3, G1-3. In this study, all patients received tamoxifen directly after surgery and results were determined according to the length of disease-free survival. To show as much as possible about the final target group, patients and their tumor samples had to meet the following criteria:
The patient's follow-up range and median were:
Median: 64.5 months Range: 3 to 142 months (calculated based on patients with no disease at the end of the observation period).
Analysis of the methylation pattern of a patient sample (FIG. 1) treated with tamoxifen as an adjunct therapy immediately after surgery is described in the plots according to FIGS. For the amplification, the average methylation over 4 oligo pairs for the amplification is calculated, the population is divided into groups by their average methylation value, and the first group is from individuals with a methylation score higher than the median. And the second group consists of solids with a methylation score lower than the median.

The primer oligonucleotides used to generate the amplification analyzed in the array experiment are:
Array primer PITX2_Q21: GTAGGGGAGGGAAGTAGATGT (SEQ ID NO: 22)
Array primer PITX2_R23: TCCTCAACTCTACAAACCTAAAA (SEQ ID NO: 23)
The corresponding genomic region of the amplified product is given in SEQ ID NO: 13.

The sequence of the oligonucleotides used in this array experiment is as follows:
SEQ ID NO: xx: AGTCGGGAGAGCGAAA
SEQ ID NO: xx: AGTTGGGAGAGTGAAA
SEQ ID NO: xx: AAGAGTCGGGAGTCGGA
SEQ ID NO: xx: AAGAGTTGGGAGTTGGA
SEQ ID NO: xx: GGTCGAAGAGTCGGGA
SEQ ID NO: xx: GGTTGAAGAGTTGGGA
SEQ ID NO: xx: ATGTTAGCGGGTCGAA
SEQ ID NO: xx: TAGTGGGTTGAAGAGT

  Data derived from analyzing six different CpG sites located within the preferred amplified region of the PITX2 gene by methylation-specific detection oligonucleotide hybridization analysis is a curve of metastasis-free survival according to Kaplan-Meier estimation Can be understood that the discriminating power of the marker PITX2 is increased by the selection of N0 patients. This is shown in FIGS. The X-axis shows the time of patient metastasis-free survival in years, and the Y-axis shows the percentage of patients with metastasis-free survival in%. The lower line shows the percentage of patients with no metastasis in the population of 118 patients with a methylation level above the median, and the upper curve has no metastasis in the population of 118 patients with a methylation level below the median Shows the percentage of patients.

  For example, as shown in FIG. 5, after 10 years of surgery alone, only about 65% of 118 patients with hypermethylation have no metastasis, whereas 118 patients with hypomethylation. About 90% of them have no metastasis.

  This distinguishing power increases again, as shown in Figure 6, similar Kaplan-Meier analysis for a sub-population of 148 patients characterized as stage G1 or G2 tumors: 10 years after surgery, Only about 60% of 74 patients with a hypermethylated state have no metastases, while about 95% of 74 patients with a hypomethylated state have no metastases.

  FIG. 7 shows how survival is also correlated with tumor stage in surgery by showing similar Kaplan-Meier analysis for a subpopulation of 150 patients characterized as T1 or T2 tumor stage : The number of patients with 10-year MFS is about 68% of 112 patients with hypermethylation status, while about 95% of 112 patients with hypomethylation status have no metastasis.

(Example 3 :)
The accuracy of the distinction between different groups is further increased by combining multiple oligonucleotides from different genes. As described in this document, it was recognized that adding additional informational markers to the analysis could potentially increase the prognostic power of survival trials. Thus, it is calculated how the combination of each two methylation specific oligonucleotides from genes TBC1D3 and CDK6 and one oligonucleotide from gene PITX2 distinguishes good or bad prognostic groups It was done. The result is shown in FIG. 8 by Kaplan-Meier curve.

  FIG. 9—at the top of FIG. 8—shows the classification of patients from the sample set by the St. Gallen method (a recent method of choice for estimating disease-free survival), which makes it possible to The improvement of methylation analysis after 80 months is shown.

(Example 4: Real-time quantitative methylation analysis)
Genomic DNA was analyzed using real-time PCR after bisulfite conversion. In this analysis, four oligonucleotides were used in each reaction. Two unmethylated specific PCR primers were used to amplify a segment of processed genomic DNA containing a methylated variable oligonucleotide probe binding site. The two oligonucleotide probes competitively hybridize to the binding site, one specific for the methylated form of the binding site and the other specific for the unmethylated form of the binding site. Thus, one of the probes contains CpG at the methylation variable position (ie, anneals to the methylated bisulfite treatment site) and the other contains TpG at that position (ie, at the unmethylated bisulfite treatment site). Anneal). Each type of probe is labeled with a 5 ′ fluorescent reporter dye and a 3 ′ quencher dye in which CpG and TpG oligonucleotides are labeled with different dyes.

  In order to quantify the level of methylation in the sample, the reaction is corrected with reference to a DNA standard of known methylation level. DNA standards are bisulphite treated phi29 amplified genomic DNA (ie unmethylated) and / or phi29 amplified genomic DNA treated with Sss1 methylase enzyme (which methylates each CpG position in the sample) (this Is then treated with a bisulfite solution). Seven different reference standards were used at 0% (ie phi29 amplified genomic DNA only), 5%, 10%, 25%, 50%, 75% and 100% (ie phi29 Sss1 treated genomes only) .

  The amount of sample DNA amplified is quantified with reference to the gene (β-actin (ACTB)) and normalized for input DNA. With respect to normalization, primers and probes for analysis of the ACTB gene lack CpG dinucleotides and thus amplification is possible regardless of the methylation level. Since there is no methylation variable position, only one probe is required.

The following oligonucleotides were used in reactions that amplify the control amplification:
Control primer 1: TGGTGATGGAGGAGGTTTAGTAAGT (SEQ ID NO: 10)
Control primer 2: AACCAATAAAACCTACTCCTCCCTTAA (SEQ ID NO: 11)
Control probe: 6FAM-ACCACCACCCAACACACAATAACAAACACA-TAMRA or Dabcyl (SEQ ID NO: 12)

  The nucleic acid sequence of gene PITX2 is given in (SEQ ID NO: 1), and after bisulfite treatment, two different strands are generated, each of which is represented twice, once before processing the methylated form ( SEQ ID NO: 2 and 3), and once before processing the unmethylated form (SEQ ID NO: 4 and 5), which are free of cytosine bases (adjacent to guanine and methylated before processing) Characterized 5 ').

The following primers are used to generate an amplification within the PITX2 sequence containing the CpG site of interest:
PITX bisulfite amplification product length: 144 bp primer
PITX2: GTAGGGGAGGGAAGTAGATGTT (SEQ ID NO: 6)
PITX2: TTCTAATCCTCCTTTCCACAATAA (SEQ ID NO: 7)

The genomic region of the generated amplified product having a length of 144 bp is given in SEQ ID NO: 18.
probe:
PITX2cg1: FAM-AGTCGGAGTCGGGAGAGCGA-Darquencher (SEQ ID NO: 8)
An alternative quencher TAMRA was also used in additional experiments:
FAM-AGTCGGAGTCGGGAGAGCGA-TAMRA
PITX2tg1: YAKIMA YELLOW-AGTTGGAGTTGGGAGAGTGAAAGGAGA-Darquencher (SEQ ID NO: 9)
The following were also used in further experiments:
VIC- AGTTGGAGTTGGGAGAGTGAAAGGAGA -TAMRA

The degree of methylation at a particular locus was determined by the following formula:
Methylation rate = 100 * I (CG) / (I (CG) + I (TG))
(I = fluorescence intensity of CG-probe or TG-probe)
PCR components were ordered from Eurogentec:
3 mM MgCl2 buffer, 10x buffer, Hotstart TAQ
Program (45 cycles): 95 ° C, 10 minutes; 95 ° C, 15 seconds; 62 ° C, 1 minute

  This analysis was performed on the same 236 samples used in Example 2. The results are shown in FIG. FIG. 2 shows a disease-free survival curve by Kaplan-Meier estimation for the 3CpG position of the PITX2 gene by real-time methylation specific probe analysis as described above. The lower curve shows the proportion of disease free patients in the population with a methylation level above the median, and the upper curve shows the proportion of disease free patients in the population with a methylation level below the median. The X axis shows the time of patient disease free survival in months, and the Y axis shows the percentage of the population of patients with disease free survival. The p-value (the observed distribution is likely to happen by chance) is calculated to be 0.0031, thereby confirming the data obtained by array analysis.

  For comparison, FIG. 3 shows from the array analysis of the gene by the chip hybridization experiment described in Example 2 in which a detection oligo was used (for details see EP 03 090 432.0, which is incorporated herein by reference). The results are shown. The p-value (the observed distribution could be caused by chance) was calculated as 0.0011.

(Example 5)
Another QM analysis was developed and was also performed very well. The following PITX2-specific oligonucleotides were used to generate 164 bp amplifications. The oligonucleotide is specific for three comethylated CpG positions:
Primer for PITX2 bisulfite amplification with a length of 162 bp:
PITX2: AACATCTACTTCCCTCCCCTAC (SEQ ID NO: 14)
PITX2: GTTAGTAGAGATTTTATTAAATTTTATTGTAT (SEQ ID NO: 15)
The genomic region resulting from the 162 bp long amplification is given in SEQ ID NO: 19.

Probe (from ABI):
PITX2-IIcg1: FAM-TTCGGTTGCGCGGT-MGBNQF (SEQ ID NO: 16)
PITX2-IItg1: VIC-TTTGGTTGTGTGGTTG-MGBNQF (SEQ ID NO: 17)

The degree of methylation at a particular locus was determined by the following formula:
Methylation rate = 100 * I (CG) / (I (CG) + I (TG))
(I = fluorescence intensity of CG-probe or TG-probe)
PCR components were ordered from Eurogentec:
2,5 mM MgCl2 buffer, 10x buffer, Hotstart TAQ
Program (45 cycles): 95 ° C, 10 minutes; 95 ° C, 15 seconds; 60 ° C, 1 minute

(Example 6: LOH analysis)
Patient Materials The materials used in this study were fresh frozen healthy breast tissue, fresh frozen breast tumor tissue from untreated breast cancer patients (followed over> 10 years), and tamoxifen treated patients Consisting of samples from (followed> 10 years since tamoxifen treatment). An aqueous solution of DNA from these microdissected lesions was used as a raw template for PCR-based LOH (loss of heterozygosity) analysis. All tumor samples were derived from ER + lymph node negative patients.

LOH analysis DNA from all tissue samples was subjected to PCR-based LOH using two 4q25-26 markers (D4S1284 and D4S406). These markers define a region on chromosome 4 that contains the PITX2 gene, which is a distance greater than 8.5 kbp of the region previously shown to have undergone LOH in breast carcinoma [Cancer Research 59, 3576- 3580, August 1, 1999].

DNA extraction
DNA is extracted from the sample using the Wizzard Kit (Promega).
PCR reactions See Clin. Cancer Res., 5: 17-23, 1999 for further details.
Analyze each sample by single-plex PCR using the following primers:

D4S406
Forward primer: GAAAGGCAGAGTCATAACAGGAAG (SEQ ID NO: 32)
Reverse primer: TAAGGATAGAGTGATTTCCAAGAAAG (SEQ ID NO: 33)
PCR product size: 205 (bp)
GenBank Accession: Z16728

D4S1284
Forward primer: CTTATCTGACAACAAGCGAGTATG (SEQ ID NO: 34)
Reverse primer: CAATTATTGTATTGTAGCATCGGAG (SEQ ID NO: 35)
PCR product size: 172 (bp)
GenBank Accession: L14168
A forward primer having either a fluorescent FAM tag (D4S1284) or a fluorescent TET tag (D4S406) at the 5 ′ end is synthesized.
Prepare a suitable amount of the nucleotide mixture described in Table 2.
Place 1 μl of each DNA sample aqueous solution in separate PCR tubes, add 9 μl of the reaction mixture according to Table 3, and perform thermal cycling according to the following conditions.
Thermal cycling conditions:
95 ° C, 15 minutes; 39 cycles: 95 ° C, 1 minute; 55 ° C, between 0:45; 72 ° C, between 1:15; and 72 ° C, between 10 minutes.

Gel electrophoresis Horizontal ultrathin high throughput fluorescence-based DNA fragment gel electrophoresis (Horizontal ultrathin, high throughput fluorescence- based DNA fragment gel electrophoresis) of the PCR product alleles were separated, and the preferred technique for the analysis. Prior to electrophoresis, 1 microliter of amplification material is mixed with 2 μl formamide loading dye (APB). ROX 350 fluorescence size marker (0.7 μl; ABI) is added to the amplified tumor DNA to allow allelic sizing. Samples are heated to 95 ° C., loaded to 70 μm and electrophoresed in 1 × TBE on a 5% horizontal polyacrylamide gel for 1 hour 15 minutes at 30 W.

Data may be collected as generally known in the art (see, eg, Clin. Cancer Res., 5: 17-23, 1999).
To determine whether allelic detection occurred at individual markers, alleles were calculated and, after normalization, loss of intensity of treated and untreated tumor samples compared to corresponding normal samples As a percentage (D value). If the allele ratio in tumor DNA is reduced by more than 40% (DO.40) than seen in normal DNA, the sample is meant to have LOH at that locus.

(Example 7: Sequencing of gene PITX2)
To confirm that CpG position comethylation is related across all exons, the gene PITX2 was sequenced. For bisulfite sequencing, amplification primers were designed to cover 11 sequences within the gene PITX2. See FIG. 11 for more details. The 16 samples analyzed in Example 4 were used for amplicon generation. Each sample was treated with sodium bisulfite and sequenced. Sequence data was obtained using ABI 3700 sequencing technology. The resulting sequence trace was normalized and the methylation percentage was calculated using the bisulfite sequencing trace analysis program owned by the applicant (see WO 2004/000463 for further information) .

In the analysis using QM analysis as described in Sample Example 4, 8 samples showed hypermethylation and 8 samples showed hypomethylation.

The fragment to be amplified was amplified using the following conditions.

PCR reaction solution:
Taq 5 U / μl 0,2
dNTPs 25mM 0,2 each
10x buffer 2,5
Wed 10,1
Primer (6,25μM) 2
DNA (1ng / μl) 10

Cycle conditions:
15 minutes, 95 ° C; 30s, 95 ° C; 30s, 58 ° C; 1:30 minutes, 72 ° C (40 cycles)

Sequencing
Only the g-rich primer was used to sequence with one exception: amplification number 2 was sequenced using both forward and reverse primers.

ExoSAP-IT reaction solution:
4 μl PCR product + 2 μl ExoSAP-IT
45 minutes / 37 ° C and 15 minutes / 95 ° C

Cycle sequencing:
1μl BigDye v.1.1
1μl water
4μl Sanger buffer
4μl dNTP mix (0,025 mM each)
-----
10 μl
+
5μl primer (2 pmol/μl)
+
6μl ExoSAP-IT product

cycle
2 minutes, 96 ° C, 26 cycles (30s / 96 ° C, 15s / 55 ° C, 4 minutes / 60 ° C)

Purification
96-well MultiScreen (Millipore) plates were filled with Sephadex G50 (Amersham) using an appropriate distributor. 300 μl of water was added to each well and incubated for 3 hours at 4 ° C. Water was removed by spinning at 910 g for 5 minutes. The cycle sequencing product was loaded onto the plate and purified by spinning at 910 g for 5 minutes. 10 μl formamide was added to each eluate.

result:
All PCR produced a product.
FIG. 12 provides a matrix generated from bisulfite sequencing data analyzed by applicant's own software (see WO 2004/000463 for further information). Each column of the matrix of columns “A” and “B” represents sequencing data for one amplification. Amplified numbers are shown to the left of the matrix. Each row of the matrix represents a single CpG site within the fragment, and each column represents an individual DNA sample. The matrix for the column marked “A” indicates methylation below the median value measured by QM analysis (see Example 4) and the matrix for the column marked “B” is QM analysis Methylation below the median value measured by.

  The left vertical bar represents the percentage methylation scale by the degree of methylation represented by the darkness of each position in the light gray column representing black to 0% methylation representing 100% methylation. White positions represent measurements for which no data was available.

  Bisulfite sequencing showed different methylation of CpG sites between the two selected classes of samples, and comethylation was observed across genes. In particular, amplifications 4 to 7 showed a high level of different methylation among the analysis groups.

(Example 8)
Real-time analysis is designed to validate the most promising marker panels from the set of ERBB2, TFF1, PLAU, PITX2, ONECUT, TBC1D3, and ABCA8 and optimized to provide an analysis with optimal accuracy It was done. The analysis was performed in combination with paraffin embedded tissue (hereinafter also referred to as PET) and fresh frozen tissue samples. PET-derived DNA is often “low quality” (eg, high degree of DNA fragmentation and low DNA recovery from the sample), so confirmation of analytical results with PET is robust to the analysis, and Shows increased marker utility.

  Quantitative methylation analysis was designed for genes ERBB2, TFF1, PLAU, PITX2, ONECUT, TBC1D3, and ABCA8, 415 estrogen receptor positive lymph node negative samples, and 541 estrogen receptor Tested using a sample set of tamoxifen treated samples of positive lymph node negative samples. Nearly 100 of these samples were previously analyzed in the microarray study.

QM analysis (Quantitative Methylation Assay) is a real-time PCR-based method for quantitative DNA methylation detection. The analytical principle is based on unmethylated specific amplification of the target region and methylation specific detection by competitive hybridization of two different probes that are specific for CG or TG status. For this study, TaqMan probes were labeled with two different fluorescent dyes (“FAM” for CG-specific probes and “VIC” for TG-specific probes) and further modified with a quencher molecule (“ TAMRA "or" Minor Groove Binder / non-fluorescent quencher ") was used.
Evaluation of the raw data for QM analysis is possible using two different methods:
1. 1. Measurement of absolute fluorescence intensity (FI) in log phase of amplified product Difference in threshold cycle (Ct) of CG and TG specific probes.
By using the Ct method, the results of this study are generated.

  In the following series of quantitative methylation analyses, the amount of sample DNA amplified is quantified by reference to the gene GSTP1 and normalized for input DNA. For normalization, primers and probes for analysis of the GSTP1 gene lack CpG dinucleotides and thus amplification is possible regardless of the methylation level. Since there are no methylation variable positions, only one probe oligonucleotide is required.

Sample set
ER + N0 tumor samples from patients who were not treated with any adjuvant therapy were analyzed to show that the identified markers without ER + N0 treatment have a strong prognostic component. Markers that can show a significant difference in survival in this population are considered prognostic. All 508 samples in this set were obtained from university collaborators as fresh cell pellets (fresh frozen samples). The sample population can be divided into two subsets: one has 415 randomly selected samples (from both censored and recurrent patients) and has a natural distribution of relapses Representing the population, the other 93 samples are from relapsed patients only. The latter sample was used only for sensitivity / specificity analysis.

  FIG. 16 shows disease free survival of a randomly selected population in the Kaplan-Meier plot, and FIG. 17 shows the distribution of follow-up periods for recurrent censored patients in a frequency distribution table. Table 6 lists the number of events interrupted by different types of recurrence. In summary, the survival time of this population is consistent with that predicted from the literature.

ER + N0 TAM treated population One intended target population of the present invention is patients with ER + N0 tumors treated with hormone therapy. To investigate the performance of marker candidates in this population, 589 samples from ER + N0 tumors from patients treated with tamoxifen were analyzed. All samples were received as paraffin embedded tissue (PET). Three to ten 10 μm pieces were provided.

In addition, for 89 PET, patient samples corresponding to fresh frozen samples from the same tumor were included in the study as controls. Since these samples have already been used in Phase 1, they have allowed two identical studies:
・ Chip vs. QM analysis ・ Frozen frozen sample vs. PET sample

  ER +, N0, TAM treated population samples were accepted from 8 different donors. A total of 589 samples were processed, 48 of which had to be excluded from the study for various reasons (eg, two samples from the same tumor, samples from patients who did not meet the inclusion criteria, etc.).

  FIG. 18 shows disease free survival of the entire population in Kaplan-Meier plot, and FIG. 19 shows the distribution of follow-up period for recurrent censored patients in a frequency distribution table. Table 5 lists the number of events interrupted by different types of recurrence. In summary, the survival time of this population (82.1% after 10 years) is consistent with that predicted from the literature (79.2%).

DNA extraction <br/> DNA extraction from fresh frozen samples
> Genomic DNA was isolated using QIAamp Kit (Qiagen, Hilden, Germany) from all 508 fresh frozen samples available as nuclear pellets. Extraction was performed according to the cell culture protocol with protease K with some modifications.
DNA extraction from PET samples
589 provided PET samples were deparaffinized directly in tubes that have been carried by the donor. The tissue was then lysed and DNA extracted using the QIAGEN DNeasy Tissue kit.

Bisulfite treatment Bisulfite treatment is optimized for the applicant's laboratory workflow based on the method disclosed by Olek et al., Nucleic Acids Res. 1996 Dec 15; 24 (24): 5064-6 Has been executed.

Quantitative Standards Reactions were corrected with reference to DNA standards of known methylation level to quantify the methylation level in the sample. DNA standards include bisulphite-treated phi29 amplified human genomic DNA (Promega) (ie unmethylated) and / or phi29 amplified genomic DNA treated with Sss1 methylase enzyme (in this way, each CpG position in the sample). (Which is then bisulfite treated). Seven different reference standards were used at 0% (ie phi29 amplified genomic DNA only), 5%, 10%, 25%, 50%, 75% and 100% (ie phi29 Sss1 treated genome only). A 2000 ng batch of human genomic DNA (Promega) was treated with bisulfite. To produce methylated MDA DNA, 13 tubes of 4.5 μg MDA-DNA (700 ng / μl) were treated with Sss1.

Control analysis
The GSTP1-C3 analytical design is derived from different sources, including fresh / frozen samples, small samples such as plasma or serum, and DNA obtained from specimens in storage such as paraffin embedding materials. It is suitable for quantifying the amount of DNA. The following oligonucleotides were used in the reaction to amplify the control amplification:
Control primer 1: GGAGTGGAGGAAATTGAGAT (SEQ ID NO: 104)
Control primer 2: CCACACAACAAATACTCAAAAC (SEQ ID NO: 105)
Control probe: FAM-TGGGTGTTTGTAATTTTTGTTTTGTGTTAGGTT-TAMRA (SEQ ID NO: 06)
Cycle program (40 cycles): 95 ° C, 10 minutes
95 ° C, 15 seconds
58 ° C, 1 minute

Analytical design and reaction conditions Two analyzes were developed for the analysis of the gene PITX2 (SEQ ID NO: 23).
Analysis 1:
Primer: GTAGGGGAGGGAAGTAGATGTT (SEQ ID NO: 107)
TTCTAATCCTCCTTTCCACAATAA (SEQ ID NO: 108)
Probe: FAM-AGTCGGAGTCGGGAGAGCGA-TAMRA (SEQ ID NO: 109)
VIC-AGTTGGAGTTGGGAGAGTGAAAGGAGA-TAMRA (SEQ ID NO: 110)
Amplicon (SEQ ID NO: 111):
Fragment length: 143 bp

Primer, probe, and CpG dinucleotide positions are highlighted.
PCR components (supplied by Eurogentec): 3 mM MgCl2 buffer, 10x buffer, Hotstart TAQ, 200 μM dNTP, 625 nM each primer, 200 nM each probe

Cycle program (45 cycles): 95 ° C, 10 minutes
95 ° C, 15 seconds
62 ° C, 1 minute

Analysis 2:
Primer: AACATCTACTTCCCTCCCCTAC (SEQ ID NO: 112)
GTTAGTAGAGATTTTATTAAATTTTATTGTAT (SEQ ID NO: 113)
Probe: FAM-TTCGGTTGCGCGGT-MGBNQF (SEQ ID NO: 114)
VIC-TTTGGTTGTGTGGTTG-MGBNQF (SEQ ID NO: 115)
Amplicon (SEQ ID NO: 116):
Fragment length: 164 bp
Primer, probe, and CpG position locations are highlighted.

The probe covers three comethylated CpG positions.
PCR components (supplied by Eurogentec): 2,5 mM MgCl2 buffer, 10x buffer, Hotstart TAQ, 200 μM dNTP, 625 nM each primer, 200 nM each probe

Program (45 cycles): 95 ° C, 10 minutes
95 ° C, 15 seconds
62 ° C, 1 minute

The degree of methylation at a particular locus was determined by the following formula:
Use absolute fluorescence intensity: methylation rate = 100 * I (CG) / (I (CG) + I (TG))
(I = fluorescence intensity of CG probe or TG probe)
Use threshold cycle Ct: methylation rate = 100 * CG / (CG + TG) = 100 / (1 + TG / CG) = 100 / (1 + 2 ^ delta (ct))
(Assuming PCR efficiency E = 2; delta (Ct) = Ct (methylated)-Ct (unmethylated))

Gene PLAU
Primer: GTTAGGTGTATGGGAGGAAGTA (SEQ ID NO: 117)
TCCCTCCCCTATCTTACAA (SEQ ID NO: 118)
Probe: FAM-ACCCGAACCCCGCGTACTTC-TAMRA (SEQ ID NO: 119)
VIC-ACCCAAACCCCACATACTTCCACA-TAMRA (SEQ ID NO: 120)
Amplicon (SEQ ID NO: 121):
Fragment length: 166 bp
Primer, probe, and CpG position locations are highlighted.

PCR components supplied by Eurogentec: 2,5 mM MgCl2 buffer, 10x buffer, Hotstart TAQ, 200 μM dNTP, 625 nM each primer, 200 nM each probe

Program (45 cycles): 95 ° C, 10 minutes
95 ° C, 15 seconds
62 ° C, 1 minute

Gene ONECUT2
Primer: GTAGGAAGAGGTGTTGAGAAATTAA (SEQ ID NO: 122)
CCACACAAAAAATTTCTATACTCCT (SEQ ID NO: 123)
Probe: FAM- ACGGGTAGAGGCGCGGGT -TAMRA (SEQ ID NO: 124)
VIC- ATGGGTAGAGGTGTGGGTTATATTGTTTTG-TAMRA (SEQ ID NO: 125)
Amplicon (SEQ ID NO: 126):
Fragment length: 266 bp
Primer, probe, and CpG position locations are highlighted.

PCR components supplied by Eurogentec: 3 mM MgCl2 buffer, 10x buffer, Hotstart TAQ, 200 μM dNTP, 625 nM each primer, 200 nM each probe


Program (45 cycles): 95 ° C, 10 minutes
95 ° C, 15 seconds
62 ° C, 1 minute

Gene ABCA8
Primer: GTGAGGTATTGGATTTAGTTTATTTG (SEQ ID NO: 127)
CCCTAAATCTCATCCTAAAAACAC (SEQ ID NO: 128)
Probe: FAM- TGAGGTTTCGGTTTTTAACGGTGG-TAMRA (SEQ ID NO: 129)
VIC- TGAGGTTTTGGTTTTTAATGGTGGGAT -TAMRA (SEQ ID NO: 130)
Amplicon (SEQ ID NO: 131):
Fragment length: 168 bp
Primer, probe, and CpG position locations are highlighted.

PCR components supplied by Eurogentec: 3 mM MgCl2 buffer, 10x buffer, Hotstart TAQ, 200 μM dNTP, 625 nM each primer, 200 nM each probe

Program (45 cycles): 95 ° C, 10 minutes
95 ° C, 15 seconds
62 ° C, 1 minute

Gene ERBB2
Primer: GGAGGGGGTAGAGTTATTAGTTTT (SEQ ID NO: 134)
ACTCCCAACTTCACTTTCTCC (SEQ ID NO: 135)
Probe: FAM-TAATTTAGGCGTTTCGGCGTTAGG-TAMRA (SEQ ID NO: 136)
VIC- TAATTTAGGTGTTTTGGTGTTAGGAGGGA -TAMRA (SEQ ID NO: 137)
Amplicon (SEQ ID NO: 138):
Fragment length: 144 bp
Primer, probe, and CpG position locations are highlighted.

PCR components supplied by Eurogentec: 2,5 mM MgCl2 buffer, 10x buffer, Hotstart TAQ, 200 μM dNTP, 625 nM each primer, 200 nM each probe

Program (45 cycles): 95 ° C, 10 minutes
95 ° C, 15 seconds
62 ° C, 1 minute

Gene TFF1
Primer: AGTTGGTGATGTTGATTAGAGTT (SEQ ID NO: 139)
CCCTCCCAATATACAAATAAAAACTA (SEQ ID NO: 140)
Probe: FAM- ACACCGTTCGTAAAA-MGBNFQ (SEQ ID NO: 141)
VIC- ACACCATTCATAAAAT-MGBNFQ (SEQ ID NO: 142)
Amplicon (SEQ ID NO: 143):
Fragment length: 189 bp
Primer, probe, and CpG position locations are highlighted.

PCR components supplied by Eurogentec: 2,5 mM MgCl2 buffer, 10x buffer, Hotstart TAQ, 200 μM dNTP, 625 nM each primer, 200 nM each probe

Program (45 cycles): 95 ° C, 10 minutes
95 ° C, 15 seconds
62 ° C, 1 minute

Gene TBC1D3
Primer: TTTTTAGTTGGTTTTTATTAGGGTTTT (SEQ ID NO: 144)
CCAACATATCCACCCACTTACT (SEQ ID NO: 145)
Probe: FAM-TTTTCGACTAATCTCCCGCCGA-TAMRA (SEQ ID NO: 146)
VIC- TTTCAACTAATCTCCCACCAAATTTACTATCA-TAMRA (SEQ ID NO: 147)
Amplicon (SEQ ID NO: 148):
Fragment length: 142 bp
Primer, probe, and CpG position locations are highlighted.

PCR components supplied by Eurogentec: 4,5 mM MgCl2 buffer, 10x buffer, Hotstart TAQ, 200 μM dNTP, 625 nM each primer, 200 nM each probe

Program (45 cycles): 95 ° C, 10 minutes; 95 ° C, 15 seconds; 62 ° C, 1 minute

Each of the designed analyzes was tested on the following set of samples:
Patients who have relapsed during treatment (all relapses) and have been treated with tamoxifen.
Patients treated with tamoxifen who relapsed during treatment due to distant metastases only.
Patients who have relapsed during treatment (all relapses) and have not been treated with tamoxifen.
・ Patients who have relapsed during treatment due to distant metastases only and have not been treated with Tamoxifen.

Raw data processing All analyzes were based on CT evaluation (evaluation using fluorescence intensity is available upon request). Assuming optimal real-time PCR conditions in the exponential amplification phase, the methylated DNA concentration (C _meth ) is
Determined by
Where CT _CG indicates the threshold cycle of the CG receptor (FAM channel)
CT _TG indicates the threshold cycle of the TG receptor (VIC channel).
The cycle threshold was determined by a human expert after video inspection of the Amplification Plots [ABI PRISM 7900 HT Sequence Detection System User Guide]. Cycle values (CT _CG and CT _TG ) were calculated using these thresholds by ABI 7900 software. Whenever the amplification curve did not exceed the threshold, the cycle value was set to the maximum cycle, ie 50.

Statistical methods Cox regression The relationship between disease-free survival time (DFS) (or metastasis-free survival, MFS) and covariates is the Cox Proportional Hazard model (Cox and Oates, 1984; Harrel, 2001). Hazard, the immediate risk of recurrence, for invariant and multiple regression analysis, respectively
h (t | x) = h ₀ (t) × exp (βx) (3)
as well as
h (t | x ₁ ,…, x _k ) = h ₀ (t) × exp (β ₁ x ₁ +… + β _k x _k ) (4)
Where t is the time measured in months after surgery, h ₀ (t) is the baseline hazard, and x is a covariate vector (eg analytical measure) ) And β are vectors of regression coefficients (model parameters). β is estimated by maximizing the partial likelihood of the Cox proportional hazards model.
A likelihood ratio test is performed to test whether methylation is associated with a hazard. The difference between ^~ Log (likelihood) of the complete model and the zero model is almost a ' ² distribution with k degrees of freedom under the zero hypothesis β ₁ =… = β _k = 0.

  The proportional hazard assumption was confirmed by the scaled Schoenfeld residual (Thernau et al., 2000).

  For the calculation, an analysis and diagnosis of the Cox proportional hazards model in the “survival” package was used.

Stepwise Regression Analysis For multivariable Cox regression models, the stepwise procedure (Venables et al., 1999; Harrel, 2001) was used to find submodels containing only relevant variables. Two effects are usually achieved by these procedures:
-Variables (methylation rate) that are not basically related to dependent variables (DFS / MFS) are excluded because they do not add information relevant to the model.
Only one of the highly correlated sets of variables that are most relevant to the dependent variable is retained.
Inclusion of both types of variables can lead to numerical instability and loss of power. Furthermore, the prediction performance can be lowered by overfitting.

The applied algorithm aims to minimize Akaike's Information Criterion (AIC), which is
AIC = -2 x maximum log likelihood + 2 x number of parameters.
AIC is related to the predictive performance of the model, and smaller values promise better performance. Inclusion of additional variables always improves the suitability of the model and thus increases the possibility, but the second term disadvantageously estimates additional parameters. The best model gives a compromise model with good fit and usually a small or moderate number of variables.

  Stepwise regression calculation by AIC was performed by R function "step".

Kaplan-Meier survival curves and log-rank test survival curves are estimated from DFS / MFS data using the Kaplan-Meier method (Kaplan and Meier, 1958). The Log-rank test was used to test for two survival curves, such as the difference between a hypermethylated group and a hypomethylated group. See (Cox and Oates, 1984) for a description of this study.

  For Kaplan-Meier analysis, the functions "survfit" and "survdiff" in the "survival" package were used.

Independence of markers from other covariates To see if our marker panel gives additional and independent information, we include other relevant clinical factors in the Cox proportional hazards model and weight each factor The p-value was calculated for (Wald test) (Thernau et al., 2000). For the analysis of additional factors in the Cox proportional hazards model, the R function "coxph" was used.

Correlation analysis
Pearson and Spearman correlation coefficients are calculated to estimate the consistency between measurements (eg, methylation in matching fresh frozen and PET samples).

For density estimation numerical variables, nuclear density estimation is performed using the Gaussian kernel and the bandwidth of the variable. Bandwidth is determined using Silverman's "rule-of-thumb" (Silverman, 1986). The R function "density" was used to calculate the density.

Sensitivity and specificity analysis For single analysis and marker panel sensitivity and specificity analysis, ROC was calculated. The calculation of ROC was done in two ways: The first method is to calculate the sensitivity and specificity for a given threshold T _{Threshold with} respect to time. Using that threshold, true positives, false positives, true negatives, false negatives were defined, and values for sensitivity and specificity were calculated for different cutoffs in the model. Patients who were censored before T _Threshold were excluded. ROC was calculated for T _{Threshold at} different times (3 years, 4 years, ..., 10 years). The second method is to calculate sensitivity and specificity by using the Bayesian formula based on Kaplan-Meier estimates for the probability of survival in marker positive and marker negative groups for a given time T _Threshold. (Heagerty et al., 2000). ROC was calculated for T _{Threshold at} different times (3 years, 4 years, ..., 10 years).

A k- fold cross-validation (Hastie et al., 2001) was used for k-fold cross-validation model selection and model robustness analysis. The set of observations was divided into k parts by random numbers. In turn, all parts were used as a test set and the remaining k-1 parts were used as a training set. This procedure was repeated n times.

Population Charts
Population Charts (Moecks et al., 2002) were used to explain the relationship between censoring and covariates. Covariate baselines were calculated including all observations with events. For a given time t, for all censored patients at the risk set at time t, an average (for actual variables such as age) or a fraction (for absolute variables) is calculated and baseline Added to the value of.

Technical performance
Comparison of analysis repeats Each marker was measured at least 3 times and the variation between analysis repeats was found to be higher for PET than for fresh frozen samples.

Fresh frozen sample vs. PET sample consistency study The markers analyzed in this study (Example 2) were first identified on a chip platform using fresh frozen samples (Example 1). The untreated population of ER + N0 was also analyzed for fresh frozen samples in Example 2. Consistency studies must show that the measured methylation ratio is comparable for fresh frozen and PET samples. For this purpose, 89 fresh frozen samples from three different donors already used on the chip were processed again in parallel with corresponding PET samples from the same tumor.

  FIG. 15 shows such a concordance study for marker candidate PITX2 analysis 1 as a scatter plot between freshly frozen and PET samples (using QM analysis). The association between the paired samples is 0.81 (Spearman's rho). This is based on n = 89 samples.

result
Single marker evaluation Each of the eight constructed QM analyzes was used to measure 508 samples from a non-treated patient population of N0, ER + (random selection and further recurrence) in triplicate. It was. After filtering the measurement points that did not meet the qualitative criteria and performing Cox analysis, Kaplan-Meier survival curves and ROC curves for each single marker were generated.
Two different clinical endpoints were used for analysis:
Use disease-free survival, ie, all types of recurrence (distant metastasis, in situ recurrence, recurrence in the opposite breast) as an event.
Treat only metastasis-free survival, ie distant metastases as events.

To analyze the ER +, N0, TAM treated population, five candidate markers were analyzed for 541 samples from a population of patients not treated with N0, ER +. The analysis was measured in triplicate. The three analyzes measured for the untreated population (PITX-2, ONECUT, and ABCA8) were not measured by the limited material available for the TAM treated population. Whether these analyzes are performed poorly in the untreated population (ONECUT and ABCA8) or in PITX2-II when performed significantly worse than other analyzes of this marker (PITX2-I) Because it was, it was rejected. After filtering the measurement points that did not meet the qualitative criteria, Kaplan-Meier survival curves and ROC curves were generated for each single marker.
Two different clinical endpoints were used:
Use disease-free survival, ie, all types of recurrence (distant metastasis, in situ recurrence, recurrence in the opposite breast) as an event.
Treat only metastasis-free survival, ie distant metastases as events.

  The disease-free or metastasis-free survival curves from Kaplan-Meier estimates for each single analysis are shown in FIGS. 14-39 and the analysis combinations are shown in FIGS. 40-55. The X axis shows the time of disease free survival of patients in years, and the Y axis shows the percentage of patients with disease free survival. The black plot shows the proportion of patients with no disease in the population with the optimal cut-off point methylation level, and the gray plot has no disease in the population with the optimal cut-off point methylation level Shows the percentage of patients.

  The following p-values (the observed distribution can occur by chance) were calculated when the cutoff was optimized. For cut-off optimization, the quantiles for both groups varied between 0.2 and 0.8, and the p-value for curve separation was calculated for each quantile. The quantile with the lowest p-value was then the best cutoff. The percentage value is attributed to the methylation rate at the cut-off point.

Single gene analysis
Tamoxifen treatment
TAM treatment (all recurrences) ERBB2 (Figure 14): p-value 0.089; cut-off point: 1.3%
TAM treatment (only distant) ERBB2 (Figure 15): p-value 0.084; cut-off point: 0.1%
TAM treatment (all recurrences) TFF1 (Figure 16): p-value 0.037; cut-off point: 50.9%
TAM treatment (only distant) TFF1 (Figure 17): p-value 0.029; cut-off point: 52.9%
TAM treatment (all recurrences) PLAU (Figure 18): p-value 0.056; cut-off point: 4.8%
TAM treatment (only distant) PLAU (Figure 19): p-value 0.065; cut-off point: 4.8%
TAM treatment (all recurrences) PITX2 (Figure 20): p-value 0.01; cut-off point: 13.1%
TAM treatment (only distant) PITX2 (Figure 21): p-value 0.0012; cut-off point: 14.3%
TAM treatment (all recurrences) TBC1D3 (analysis II) (Figure 22): p-value 0.28; cut-off point: 94.6%
TAM treatment (only distant) TBC1D3 (analysis II) (Figure 23): p-value 0.078; cut-off point: 97%

FIG. 62 shows ROC plots at different times for only marker model PITX2 (Analysis 1) in the ER + N0 TAM treated population. Figure A shows a plot for 60 months, Figure B shows a plot for 72 months, Figure C shows a plot for 84 months, and Figure D shows a plot for 96 months. Only distant metastases are defined as events. Sensitivity (percentage of all relapsed patients in the poor prognosis group) is shown on the X-axis and specificity (percentage of all non-relapsed patients in the good prognostic group) is shown on the Y-axis, which are based on KM estimates And the estimated area under the curve (AUC) is calculated. The values for the median cutoff (triangle) and the best cutoff (diamond, 0.42 quantile) are plotted.
AUC 60 months: 0.6
AUC 72 months: 0.69
AUC 84: 0.69
AUC 96: 0.67

FIG. 63 shows ROC plots at different times for the marker model TFF1 in the ER + N0 TAM treated population. Figure A shows a plot for 60 months, Figure B shows a plot for 72 months, Figure C shows a plot for 84 months, and Figure D shows a plot for 96 months. Only distant metastases are defined as events. Sensitivity (the proportion of all relapsed patients in the poor prognosis group) is shown on the X axis, and specificity (the proportion of all non-relapsed patients in the good prognosis group) is shown on the Y axis, which are different thresholds ( = 5,6,7,8)), the estimated area under the curve (AUC) is calculated. The values for the median cutoff (triangle) and the best cutoff (diamond, 0.78 quantile) are plotted.
AUC 60: 0.7
AUC 72 months: 0.65
AUC 84: 0.61
AUC 96: 0.64

FIG. 64 shows ROC plots at different times for the marker model PLAU in the ER + N0 TAM treated population. Figure A shows a plot for 60 months, Figure B shows a plot for 72 months, Figure C shows a plot for 84 months, and Figure D shows a plot for 96 months. Only distant metastases are defined as events. Sensitivity (the proportion of all relapsed patients in the poor prognosis group) is shown on the X axis, and specificity (the proportion of all non-relapsed patients in the good prognosis group) is shown on the Y axis, which are different thresholds ( = 5,6,7,8)), the estimated area under the curve (AUC) is calculated. The values for the median cutoff (triangle) and the best cutoff (diamond, 0.77 quantile) are plotted.
AUC 60: 0.6
AUC 72 months: 0.63
AUC 84: 0.57
AUC 96: 0.6

No tamoxifen treatment, no tamoxifen treatment (all relapses) ERBB2 (Figure 24): p-value 0.21; cut-off point: 0%;
No tamoxifen treatment (only distant) ERBB2 (Figure 25): p-value 0.23; cut-off point: 0.6%,
No tamoxifen treatment (all relapses) TFF1 (Figure 26): p-value 0.012; cut-off point: 49.6%;
No tamoxifen treatment (only distant) TFF1 (Figure 27): p-value 0.016; cut-off point: 45.4%;
No tamoxifen treatment (all relapses) PLAU (Figure 28): p-value 0.011; cut-off point: 3.2%;
No tamoxifen treatment (only distant) PLAU (Figure 29): p-value 0.0082; cut-off point: 5.5%;
No tamoxifen treatment (all relapses) PITX2 (I) (Figure 30): p-value 1.4e-06; cut-off point: 35.4%;
No tamoxifen treatment (only distant) PITX2 (I) (Figure 31): p-value 1.7 e-05; cut-off point: 41.2%;
No tamoxifen treatment (all relapses) PITX2 (II) (Figure 32): p-value 0.00026; cut off point: 56.1%;
No tamoxifen treatment (only distant) PITX2 (II) (Figure 33): p-value 0.0026; cut-off point: 61.9%;
No tamoxifen treatment (all recurrences) ONECUT (Figure 34): p-value 0.26; cut-off point: 0%;
No tamoxifen treatment (only distant) ONECUT (Figure 35): p-value 0.77; cut-off point: 0%;
No tamoxifen treatment (all relapses) TBC1D3 (Figure 36): p-value 0.004; cut-off point: 98.6%;
No tamoxifen treatment (only distant) TBC1D3 (Figure 37): p-value 0.00022; cut-off point: 98.6%;
No tamoxifen treatment (all relapses) ABCA8 (Figure 38): p-value 0.0065; cut-off point: 60.9%;
No tamoxifen treatment (only distant) ABCA8 (Figure 39): p-value 0.15; cut-off point: 49.2%;

Based on the results of panel single marker evaluation, it was decided to build a model using candidate marker PITX2-analysis I, TFF1 and PLAU. The full potential of these marker combinations was evaluated.

Tamoxifen treatment
TAM treatment (all relapses) TFF1 and PLAU (FIG. 40): p-value 0.023; cut-off point: 0.7 quantile;
TAM treatment (only distant) TFF1 and PLAU (FIG. 41): p-value 0.00084; cut-off point: 0.72 quantile;
TAM treatment (all relapses) TFF1 and PLAU and PITX2 (FIG. 42): p-value 0.037; cut-off point: 0.72 quantile;
TAM treatment (only distant) TFF1 and PLAU and PITX2 (FIG. 43): p-value 0.0014; cut-off point: quartile;
TAM treatment (all relapses) PITX2 and TFF1 (Figure 44): p-value 0.17; cut-off point: 0.78 quantile;
TAM treatment (only distant) PITX2 and TFF1 (Figure 45): p-value 0.0048; cut-off point: 0.32 quantile;
TAM treatment (all relapses) PITX2 and PLAU (Figure 46): p-value 0.1; cut-off point: 0.74 quantile;
TAM treatment (only distant) PITX2 and PLAU (FIG. 47): p-value 0.0081; cut-off point: 0.44 quantile.

FIG. 61 shows ROC plots at different times for marker models PITX2 (analysis 1) and TFF1 in the ER + N0 TAM treated population. Figure A shows a plot for 60 months, Figure B shows a plot for 72 months, Figure C shows a plot for 84 months, and Figure D shows a plot for 96 months. Only distant metastases are defined as events. Sensitivity (percentage of all relapsed patients in the poor prognosis group) is shown on the X-axis and specificity (percentage of all non-relapsed patients in the good prognostic group) is shown on the Y-axis, which are based on KM estimates And the estimated area under the curve (AUC) is calculated. The values for the median cutoff (triangle) and the best cutoff (diamond, 0.32 quantile) are plotted.
AUC 60: 0.62
AUC 72 months: 0.67
AUC 84: 0.63
AUC 96: 0.65

No tamoxifen treatment No tamoxifen treatment (all relapses) TFF1 and PLAU (Figure 47): p-value 0.0015; cut-off point: 0.78 quantile;
No tamoxifen treatment (only distant) TFF1 and PLAU (FIG. 47): p-value 0.003; cut-off point: about 0.8 quantile;
No tamoxifen treatment (all relapses) TFF1 and PLAU and PITX2 (FIG. 47): p-value 8.9e-07; cut-off point: 0.64 quantile;
No tamoxifen treatment (only distant) TFF1 and PLAU and PITX2 (FIG. 47): p-value 5.4e-05; Cut-off point: 0.66 quantile;
No tamoxifen treatment (all relapses) PITX2 and TFF1 (Figure 47): p-value 1.9e-06; cut-off point: 0.72 quantile;
No tamoxifen treatment (only distant) PITX2 and TFF1 (Figure 47): p-value 3.5e-05; cut-off point: 0.76 quantile;
No tamoxifen treatment (all relapses) PITX2 and PLAU (Figure 47): p-value 1.1e-06; cut-off point: 0.68 quantile;
No tamoxifen treatment (only distant) PITX2 and PLAU (Figure 47): p-value 1.5e-05; cut-off point: 0.64 quantile.

Robustness of the marker model In order to evaluate the robustness of the model, cross-validation was repeated 200 times with only the model marker panel PITX2 (analysis 1) and TFF1 and the marker panel PITX2 (analysis 1). The stability of assignment of a particular patient to a bad or good outcome group is shown in FIG. 65, the left hand figure shows the model marker panel PITX2 (analysis 1) and TFF1, and the right hand figure shows the marker panel PITX2 Only (Analysis 1) is shown. The plot shows how many cross-validation iterations each patient is assigned to group 1 (light gray) or group 2 (dark gray).

A simple model of stage 1-3 breast cancer tumor is shown. The result of the analysis (QM analysis) described in Example 4 is shown. FIG. 6 is a diagram showing the results of chip hybridization experiments described in Example 2. It is a figure which shows the curve of a metastasis free survival by Kaplan-Meier estimation regarding two CpG positions of the PITX2 gene by methylation specific detection oligonucleotide hybridization analysis. FIG. 6 shows a curve of metastasis-free survival according to Kaplan-Meier estimation for 6 different CpG positions within a preferred region of the PITX2 gene (SEQ ID NO: 13) by methylation specific detection oligonucleotide hybridization analysis. FIG. 5 shows a Kaplan-Meier estimated metastasis-free survival curve for six different CpG positions within a preferred region of the PITX2 gene (SEQ ID NO: 13) by methylation specific detection oligonucleotide hybridization analysis (grade G1 or G2). Analysis of only a sub-population of 148 patients characterized as tumors. FIG. 5 shows a Kaplan-Meier estimated metastasis-free survival curve for six different CpG positions within a preferred region of the PITX2 gene (SEQ ID NO: 13) by methylation specific detection oligonucleotide hybridization analysis (stage 1 or 2). (Analysis of only a subpopulation of 224 patients characterized as (T1 or T2) tumors). FIG. 5 shows disease-free survival curves for a combination of two oligonucleotides from each of genes TBC1D3 and CDK6 and one oligonucleotide from gene PITX2 covering two CpG sites. It is a figure which shows the plot by FIG. 8, and shows the classification | category of the sample set by St. Gallen method. FIG. 2 is a diagram of the amino acid sequence of a polypeptide encoded by the gene PITX2. FIG. 6 is a diagram of the positions of the amplification products sequenced in Example 7. FIG. 7 shows sequencing data of 11 amplified products of gene PITX2 according to Example 7. Schematic representation of PITX2 mRNA transcript variants as annotated in the online Ensembl database. It is a figure which shows the curve of disease free survival by Kaplan-Meier estimation regarding the CpG position of ERBB2 gene by real-time methylation specific probe analysis by Example 8. It is a figure which shows the curve of a metastasis-free survival time by the Kaplan-Meier estimation regarding the CpG position of ERBB2 gene by real-time methylation specific probe analysis by Example 8. It is a figure which shows the curve of a metastasis-free survival time by the Kaplan-Meier estimation regarding the CpG position of TFF1 gene by the real-time methylation specific probe analysis by Example 8. It is a figure which shows the curve of a metastasis-free survival time by the Kaplan-Meier estimation regarding the CpG position of TFF1 gene by the real-time methylation specific probe analysis by Example 8. It is a figure which shows the curve of a disease free survival time by the Kaplan-Meier estimation regarding the CpG position of a PLAU gene by the real-time methylation specific probe analysis by Example 8. It is a figure which shows the curve of a metastasis-free survival time by the Kaplan-Meier estimation regarding the CpG position of a PLAU gene by the real-time methylation specific probe analysis by Example 8. It is a figure which shows the curve of disease free survival by Kaplan-Meier estimation regarding the CpG position of PITX2 gene by real-time methylation specific probe analysis by Example 8. It is a figure which shows the curve of a metastasis-free survival time by the Kaplan-Meier estimation regarding the CpG position of a PITX2 gene by the real-time methylation specific probe analysis by Example 8. It is a figure which shows the curve of disease free survival by Kaplan-Meier estimation regarding the CpG position of TBC1D3 gene by real-time methylation specific probe analysis by Example 8. It is a figure which shows the curve of a metastasis free survival by Kaplan-Meier estimation regarding the CpG position of TBC1D3 gene by real-time methylation specific probe analysis by Example 8. It is a figure which shows the curve of disease free survival by Kaplan-Meier estimation regarding the CpG position of ERBB2 gene by real-time methylation specific probe analysis by Example 8. It is a figure which shows the curve of a metastasis-free survival time by the Kaplan-Meier estimation regarding the CpG position of ERBB2 gene by real-time methylation specific probe analysis by Example 8. It is a figure which shows the curve of a disease free survival by Kaplan-Meier estimation regarding the CpG position of TFF1 gene by real-time methylation specific probe analysis by Example 8. It is a figure which shows the curve of a metastasis-free survival time by the Kaplan-Meier estimation regarding the CpG position of TFF1 gene by the real-time methylation specific probe analysis by Example 8. It is a figure which shows the curve of a disease free survival time by the Kaplan-Meier estimation regarding the CpG position of a PLAU gene by the real-time methylation specific probe analysis by Example 8. It is a figure which shows the curve of a metastasis-free survival time by the Kaplan-Meier estimation regarding the CpG position of a PLAU gene by the real-time methylation specific probe analysis by Example 8. It is a figure which shows the curve of a disease free survival time by the Kaplan-Meier estimation regarding the CpG position of a PITX gene by the real-time methylation specific probe analysis by Example 8. It is a figure which shows the curve of a metastasis-free survival time by the Kaplan-Meier estimation regarding the CpG position of a PITX gene by the real-time methylation specific probe analysis by Example 8. It is a figure which shows the curve of a disease free survival time by the Kaplan-Meier estimation regarding the CpG position of a PITX gene by the real-time methylation specific probe analysis by Example 8. It is a figure which shows the curve of a metastasis-free survival time by the Kaplan-Meier estimation regarding the CpG position of a PITX gene by the real-time methylation specific probe analysis by Example 8. It is a figure which shows the curve of disease free survival by Kaplan-Meier estimation regarding the CpG position of the ONECUT gene by real-time methylation specific probe analysis by Example 8. It is a figure which shows the curve of a metastasis-free survival time by the Kaplan-Meier estimation regarding the CpG position of the ONECUT gene by the real-time methylation specific probe analysis by Example 8. It is a figure which shows the curve of disease free survival by Kaplan-Meier estimation regarding the CpG position of TBC1D3 gene by real-time methylation specific probe analysis by Example 8. It is a figure which shows the curve of a metastasis free survival by Kaplan-Meier estimation regarding the CpG position of TBC1D3 gene by real-time methylation specific probe analysis by Example 8. It is a figure which shows the curve of a disease free survival by Kaplan-Meier estimation regarding the CpG position of ABCA8 gene by real-time methylation specific probe analysis by Example 8. It is a figure which shows the curve of a metastasis-free survival time by the Kaplan-Meier estimation regarding the CpG position of ABCA8 gene by the real-time methylation specific probe analysis by Example 8. It is a figure which shows the curve of disease free survival by Kaplan-Meier estimation regarding the CpG position of TFF1 and PLAU gene by real-time methylation specific probe analysis by Example 8. It is a figure which shows the curve of a metastasis-free survival time by the Kaplan-Meier estimation regarding the CpG position of TFF1 and PLAU gene by real-time methylation specific probe analysis by Example 8. It is a figure which shows the disease-free survival curve by the Kaplan-Meier estimation regarding the CpG position of the combination of TFF1, PLAU, and PITX gene by real-time methylation specific probe analysis by Example 8. It is a figure which shows the curve of a metastasis free survival by Kaplan-Meier estimation regarding the CpG position of the combination of TFF1, PLAU, and PITX gene by real-time methylation specific probe analysis by Example 8. It is a figure which shows the curve of disease free survival by Kaplan-Meier estimation regarding the CpG position of the combination of PITX and TFF1 gene by real-time methylation specific probe analysis by Example 8. It is a figure which shows the curve of a metastasis-free survival time by the Kaplan-Meier estimation regarding the CpG position of the combination of PITX and TFF1 gene by the real-time methylation specific probe analysis by Example 8. It is a figure which shows the curve of disease free survival by Kaplan-Meier estimation regarding the CpG position of the combination of PITX and PLAU gene by real-time methylation specific probe analysis by Example 8. It is a figure which shows the curve of a metastasis-free survival time by the Kaplan-Meier estimation regarding the CpG position of the combination of a PITX and a PLAU gene by the real-time methylation specific probe analysis by Example 8. It is a figure which shows the curve of disease free survival by Kaplan-Meier estimation regarding the CpG position of the combination of TFF1 and PLAU gene by real-time methylation specific probe analysis by Example 8. It is a figure which shows the curve of a metastasis-free survival time by the Kaplan-Meier estimation regarding the CpG position of the combination of TFF1 and PLAU gene by real-time methylation specific probe analysis by Example 8. It is a figure which shows the disease-free survival curve by the Kaplan-Meier estimation regarding the CpG position of the combination of TFF1, PLAU, and PITX gene by real-time methylation specific probe analysis by Example 8. It is a figure which shows the curve of a metastasis free survival by Kaplan-Meier estimation regarding the CpG position of the combination of TFF1, PLAU, and PITX gene by real-time methylation specific probe analysis by Example 8. It is a figure which shows the curve of disease free survival by Kaplan-Meier estimation regarding the CpG position of the combination of PITX and TFF1 gene by real-time methylation specific probe analysis by Example 8. It is a figure which shows the curve of a metastasis-free survival time by the Kaplan-Meier estimation regarding the CpG position of the combination of PITX and TFF1 gene by the real-time methylation specific probe analysis by Example 8. It is a figure which shows the curve of disease free survival by Kaplan-Meier estimation regarding the CpG position of the combination of PITX and PLAU gene by real-time methylation specific probe analysis by Example 8. It is a figure which shows the curve of a metastasis-free survival time by the Kaplan-Meier estimation regarding the CpG position of the combination of a PITX and a PLAU gene by the real-time methylation specific probe analysis by Example 8. FIG. 10 is a scatter plot of matched pair PET and fresh frozen tissue analyzed using PITX2 gene analysis 1 according to Example 8. FIG. 9 shows disease free survival (DFS) of a population of untreated patients in a randomly selected ER +, N0, Kaplan-Meier survival plot according to Example 8. FIG. 9 shows the distribution of follow-up periods in a population not treated with ER +, N0, Example 8. FIG. 6 shows disease free survival (DFS) of TAM treated population in ER +, N0, Kaplan-Meier survival plot according to Example 8. FIG. 9 shows the distribution of follow-up periods in a population not treated with ER +, N0, Example 8. FIG. 10 shows ROC plots at different times for 2265 in the marker model 3522 (analysis 1) and ER + N0 TAM treated population according to Example 8. FIG. 10 shows ROC plots at different times for only marker model 3522 (Analysis 1) in the ER + N0 TAM treated population according to Example 8. FIG. 10 shows ROC plots at different times for marker model 2265 in the ER + N0 TAM treated population according to Example 8. FIG. 10 shows ROC plots at different times for marker model 2395 in the ER + N0 TAM treated population according to Example 8. FIG. 5 shows the stability of assignment of a particular patient to a bad or good outcome group.

Claims (18)

A method for providing a prognosis and / or prediction of outcome of an endocrine treatment in a subject having breast cell proliferation disorder comprising the following steps:
a) obtaining a biological sample from the subject,
b) determining the expression of the gene PITX2 and / or its regulatory sequences in the sample,
c) determining the prognosis and / or outcome of the subject's endocrine treatment. The method according to claim 1, further comprising determining the expression of genes PITX2, TFF1, and PLAU and / or their regulatory sequences in step b). The method according to claim 1, further comprising determining the expression of the genes PITX2 and PLAU and / or their regulatory sequences in step b). The method according to claim 1, further comprising determining the expression of genes PITX2 and TFF1 and / or their regulatory sequences in step b). The method of claim 1, wherein the subject is estrogen receptor positive. 6. The method of claims 1-5 further comprising d) determining a treatment plan suitable for the subject. The preferred treatment regimen includes chemotherapy, radiotherapy, surgery, biological therapy, immunotherapy, antibodies, molecular targeted drugs, estrogen receptor modulators, estrogen receptor downregulators, aromatase inhibitors, ovariectomy, LHRH analogs , And one or more therapies selected from the group consisting of centrally acting drugs that affect other estrogen production. Non-invasive breast cancer, invasive breast cancer, invasive lobular cancer, non-invasive lobular cancer, comedous cancer, inflammatory cancer, mucinous cancer, hard cancer, colloid cancer, tubular The method according to claim 1, which is selected from the group consisting of cancer, medullary cancer, metaplastic cancer, papillary cancer, and non-invasive papillary cancer, undifferentiated or anaplastic cancer, and Paget's disease of the breast. 9. The method of claims 1-8, wherein the expression is determined by at least one analysis of mRNA, LOH, protein expression. 10. The method of claims 1-9, wherein the expression is determined by determining the methylation status of one or more CpG positions within the gene and / or its regulatory region. A method of providing a prognosis and / or predicting outcome of an endocrine treatment in a subject having breast cell proliferation disorder comprising the following steps:
a) isolating genomic DNA from a biological sample obtained from said subject;
b) treating genomic DNA or fragments thereof with one or more reagents that convert unmethylated cytosine bases at position 5 into uracil or other bases whose hybridization properties are detectably different from cytosine;
c) contacting the processed genomic DNA or a processed fragment thereof with an amplification enzyme and at least two primers, which in each case are SEQ ID NO: 150, 151, 155 and 156 of the sequence listing and their A contiguous sequence of at least 18 nucleotides in length that is complementary to a sequence selected from the group consisting of complementary sequences or hybridizes under moderately stringent or stringent conditions and has been processed DNA or a fragment thereof is amplified to produce one or more amplifications or not amplified;
d) the methylation status of at least one CpG dinucleotide sequence of SEQ ID NO: 149 of the sequence listing, or the average of a plurality of CpG dinucleotide sequences of SEQ ID NO: 149 of the sequence listing, based on the presence, amount or characteristics of the amplification product A value reflecting the methylation status is determined, and e) the prognosis and / or outcome of the subject's endocrine treatment is determined from the methylation status. In step c), in each case, it is complementary or moderate to a sequence selected from the group consisting of SEQ ID NO: 76-103 and 153, 154, 157 and 158 and their complementary sequences in the sequence listing 12. The method of claim 11 further comprising at least two primers comprising flanking sequences of at least 18 nucleotides in length that hybridize to or under stringent conditions. 13. The method of claims 11 and 12, wherein the one or more reagents comprise a solution selected from the group consisting of bisulfite, bisulfate, disulfite, and combinations thereof. The method according to claims 11 and 12, wherein d) is performed by one or more methods selected from the group consisting of oligonucleotide hybridization analysis, Ms-SnuPE, sequencing, real-time detection probe and oligonucleotide array analysis. . One of sequences selected from the group consisting of SEQ ID NO: 2 to 5, SEQ ID NO: 151 to SEQ ID NO: 158, and SEQ ID NO: 76 to SEQ ID NO: 103. A nucleic acid molecule consisting of a sequence of at least 18 bases in length. An oligomer, in particular an oligonucleotide or a peptide nucleic acid (PNA) -oligomer, wherein said oligomer is SEQ ID NO: 2-5, SEQ ID NO: 151-SEQ ID NO: 158, and SEQ ID NO: 76 An oligomer consisting essentially of at least one base sequence having a length of at least 10 nucleotides that hybridizes to or matches one of the nucleic acid sequences set forth in SEQ ID NO: 103 in the Sequence Listing. A composition comprising:
-Chemically described in one of the sequences selected from the group comprising SEQ ID NO: 2 to SEQ ID NO: 5, SEQ ID NO: 5, SEQ ID NO: 151, 152, 155 and 156 and sequences complementary thereto A nucleic acid comprising a sequence of at least 18 bases in length of a segment of pretreated genomic DNA, and a buffer comprising at least one of the following substances: magnesium chloride, dNTP, taq polymerase oligomers, in particular oligonucleotides or peptide nucleic acids (PNA ) An oligomer, said oligomer being complementary to the pretreated genomic DNA according to SEQ ID NO: 2-5, SEQ ID NO: 151, 152, 155 and 156 and one of their complementary sequences In each case at least one base sequence having a length of at least 9 nucleotides that hybridizes under moderately stringent or stringent conditions Hey including by. 18. The method of claim 1-15, the nucleic acid of claim 16, the nucleic acid of claim 17, for providing a prognosis and / or prediction of outcome of endocrine therapy in a subject having a cell proliferation disorder of breast tissue. Use of an oligonucleotide and the composition of claim 18.