Detection of hypermethylated genes for diagnosing pancreatic cancer

Abstract

The present invention relates to a method for diagnosing or identifying pancreatic cancer in a specimen. The inventors have actually identified two DNA methylation biomarkers, which, alone or preferably in combination, can highly specifically contribute to the diagnosis or follow-up of pancreatic cancer patients and can contribute to distinguishing them from other types of cancer. The present invention can then be used to determine and adjust appropriate treatment regimens for specimens diagnosed with pancreatic cancer. The present invention also relates to kits containing primers or probes for detecting, diagnosing, and identifying hypermethylated genes.

Description

[Technical Field]

[0001] The present invention relates to the field of cancer, and more particularly to the early diagnosis of pancreatic cancer. Accordingly, the present invention relates to a method for diagnosing or identifying pancreatic cancer in a specimen, which can be used to monitor the progression of pancreatic cancer in a specimen diagnosed with pancreatic cancer and / or to determine or adjust a treatment regimen appropriate for said specimen. The present invention also relates to a kit comprising primers and / or probes for detecting, determining, or identifying hypermethylated genes. [Background technology]

[0002] Cancer of the pancreas (also known as pancreatic cancer) is cancer that originates in the cells of the pancreas and begins to grow uncontrollably. The pancreas contains two main types of cells: exocrine cells and endocrine cells. Knowing whether a cancer in the pancreas is exocrine or endocrine is very important because exocrine and endocrine cells in the pancreas form different types of tumors. Therefore, these cells have unique risk factors and causes, different signs and symptoms, are diagnosed with different tests, are treated differently, and have different outlooks. Exocrine cancer is by far the most common type of pancreatic cancer. Approximately 95% of cancers in the exocrine pancreas are pancreatic adenocarcinomas. These cancers usually begin in the pancreatic duct. Less commonly, these cancers arise from enzyme-producing cells in the pancreas, which are called acinar cell carcinomas. On the other hand, less common exocrine carcinomas include: adenosquamous carcinomas, squamous cell carcinomas, signet ring cell carcinomas, undifferentiated carcinoma, and undifferentiated carcinoma with giant cells.

[0003] Tumors of the endocrine portion of the pancreas are uncommon, comprising less than 5% of all pancreatic cancers. As a group, they are often referred to as pancreatic neuroendocrine tumors (NETs), or insulinomas. Pancreatic NETs can be both benign and malignant.

[0004] Early symptoms of exocrine pancreatic cancer may include: jaundice and related symptoms, such as dark urine, light or fatty stools, itchy skin; abdominal or back pain, weight loss and loss of appetite, nausea and vomiting, enlarged gallbladder or liver, blood clots, fatty tissue abnormalities, diabetes. As mentioned above, symptoms of endocrine pancreatic cancer are different from exocrine pancreatic cancer and may include: gastrinoma, glucagonoma, insulinoma, somatostatinoma.

[0005] Many risk factors can be cited (e.g., tobacco, overweight and obesity, exposure to certain chemicals used, aging, sex, race, family history, genetics, diabetes, chronic pancreatitis, cirrhosis of the liver, stomach problems).

[0006] Pancreatic adenocarcinoma is the leading cause of cancer death in Western societies. For example, in France, the number of cases is increasing every year (12,000 cases / year) and will constitute the second leading cause of cancer death in Western countries in 2020 [Bouvier et al. 2014]. The 5-year life expectancy, for all stages combined, is barely above 5% [Coleman et al. 2003 & Eheman et al. 2012].

[0007] For over a decade, gemcitabine monotherapy has been the gold standard for advanced pancreatic adenocarcinoma [Burris et al. 1997]. Recently, two randomized phase 3 trials demonstrated that one FOLFIRINOX protocol and the other gemcitabine plus nab-paclitaxel combination were superior to gemcitabine monotherapy in terms of response rate, progression-free survival, and overall survival [Conroy T et al. 2011 & Von Hoff DD et al. 2013]. Despite these advances, pancreatic adenocarcinomas remain particularly radioresistant and chemoresistant, and overall prognosis for patients at all stages has shown little improvement. Combination and targeted therapies are being developed to improve reference treatments. Targeting genes involved in tumor progression is one option for improved treatment.

[0008] As mentioned above, ductal adenocarcinoma is the most common (80% of exocrine carcinomas) and, above all, the most serious. This is because establishing the diagnosis is extremely difficult, due to the lack of early specific clinical signs. At diagnosis, only 10% of patients benefit from radical resection, 30–40% have partially advanced disease for surgery, and 50–60% develop metastatic disease elsewhere [Siegel et al. 2015]. Mutations have been reported in patients with pancreatic cancer [Bardeesy et al. 2002]. The primary abnormalities are associated with the following genes: KRAS, TP53, or SMAD4. Activation of the proto-oncogene KRAS appears early and is found in 80–90% of pancreatic cancers [Almoguera et al. 1988 & Tada et al. 1993 & Morris JPt et al. 2010 & Kanda M et al. 2012]. Mutations in the tumor suppressor gene TP53 are found in 50–75% of pancreatic cancers [Hruban et al. 2008]. TP53 mutations occur late in the progression of pancreatic cancer and play a role in accelerating carcinogenesis [Moore et al. 2003]. SMAD4 / DPC4, a tumor suppressor gene, is inactivated in 48–55% of pancreatic cancers. Loss of SMAD4 function occurs late in pancreatic carcinogenesis. However, further research is needed to better understand and determine the prognostic and predictive value of SMAD4 mutations in pancreatic adenocarcinoma. Carbohydrate antigen 19-9, also known as CA19-9, is a serological marker that has a significant prognostic value in pancreatic adenocarcinoma [Poruk et al. 2013]. Cell-free tumor-derived circulating DNA (ctDNA) has also been evaluated to improve the value of CA19-9 in diagnosing pancreatic cancer. Thus, Dabritz and colleagues were able to diagnose pancreatic adenocarcinoma with a sensitivity of 91% [Dabritz et al. 2009 & Pietrasz et al. 2017].

[0009] The identification of prognostic and / or predictive biomarkers will ultimately allow: to better define the treatment strategy (especially surgery and / or radiotherapy) for a given patient; to search for new therapeutic targets; and to individualize treatment according to the tumor's molecular characteristics.

[0010] Currently, pancreatic cancer is usually diagnosed by the following methods: biopsy, either during an endoscopy or by inserting a needle through the patient's skin into the pancreas; blood tests to detect tumor markers (e.g., CA19-9 test); imaging tests such as CT scans, MRIs, and sometimes PET scans, or endoscopic ultrasound (EUS). However, these methods require expensive medical equipment and incur significant costs.

[0011] In this regard, there remains a need to identify sensitive and specific biomarkers for pancreatic cancer that can be used on body effluents, preferably blood samples, and that can be used to diagnose, monitor cancer progression, and determine appropriate treatment regimens for subjects diagnosed with pancreatic cancer in a simple manner.

[0012] Since the introduction of genome-wide sequencing of DNA from diseased tissues using next-generation sequencing (NGS), there has been an increased effort to identify candidate cancer biomarkers. However, these identifications are only useful for early cancer detection if performed noninvasively in the periphery. Unlike mutations, which are specific changes in DNA sequence and therefore unambiguously recognizable, DNA methylation frequently exists at multiple sites and has a spectrum of quantitative differences between tumors and normal tissues.

[0013] Methylation of the promoter and first exon of tumor suppressor genes, which results in downregulation of gene expression, has been shown to play an important role in cancer development. Aberrant DNA methylation frequently occurs in the early stages of carcinogenesis and therefore represents an attractive potential marker for the early detection of cancer. To develop biomarkers for cancer screening and early detection, increased methylation (hypermethylation) provides a positive readout for well-defined target regions for the development of sensitive and specific assays and is therefore more advantageous than global hypomethylation for clinical trials.

[0014] Despite this compelling promise, translating candidate methylated DNA markers into useful clinical diagnostic markers has been challenging, and several reviews have discussed obstacles that limit the routine use of DNA methylation biomarkers in cancer treatment [Mikeska T, 2012; Delpu Y et al, 2013]. Summary of the Invention [Problem to be solved by the invention]

[0015] These reviews emphasize that analytical sensitivity is a key parameter for the diagnostic application of methylation screening technologies, and that good analytical sensitivity can only be achieved by combining several markers or by supplementing existing screening tests, because single marker approaches are unlikely to achieve sufficient sensitivity (>90%) for screening, unless the marker is linked to the etiology of cancer and represents an extremely well-defined, high-risk group.

[0016] Another important factor for screening tests is high specificity. Low specificity results in a high number of false positives, which can lead to psychological stress for patients and unnecessary medical procedures, undermining the reliability and acceptability of the test in the population. The localized methylation profile requires a sufficient amount of assay to distinguish between baseline methylation and hypermethylation events truly associated with cancer. Furthermore, DNA methylation biomarkers are rarely specific to a single type of cancer, and most are reported to be conserved across multiple tumor types. Therefore, it seems difficult to propose a single DNA methylation change as a biomarker for a specific type of cancer, although this would be ideal for providing patients with appropriate treatment as soon as possible. To compensate for this lack of specificity, combinations of multiple biomarkers are often proposed, but this also reduces the sensitivity of the test.

[0017] Overall, the above review shows that while the principles already demonstrate the great potential of methylation biomarkers for clinical use, the field faces significant challenges in identifying ideal biomarkers that exhibit both high sensitivity and high specificity.

[0018] Furthermore, none of these reviews disclosed studies that enriched methylation biomarkers for diagnosing pancreatic cancer, suggesting that identifying sensitive and specific methylation biomarkers for pancreatic cancer is quite challenging. [Means for solving the problem]

[0019] However, in this scientific context, the present inventors have found that a combination of two specific methylation biomarkers can be used in pancreatic cancer to diagnose and monitor pancreatic cancer with excellent sensitivity and specificity, as will be disclosed in the following Examples section, which is contrary to all the shortcomings suggested by experts in the field in the above review.

[0020] Summary of the Invention

[0021] The present invention relates to a method for diagnosing or identifying pancreatic cancer in a subject, which method relies on detecting abnormal hypermethylation levels of specific genes in a biological sample from the subject. The inventors have surprisingly identified two DNA methylation biomarkers, which, alone or preferably in combination, aid in the diagnosis of pancreatic cancer in patients or in the follow-up of the progression of pancreatic cancer, and highly specifically aid in the differentiation of pancreatic cancer from other types of cancer. The present invention proposes measuring DNA hypermethylation of the genes by dPCR in a body fluid sample, preferably a blood sample, from the subject. The present invention also relates to a kit or other tool for diagnosing pancreatic cancer.

[0022] Thus, in a first aspect, the present invention relates to an in vitro method for diagnosing or identifying pancreatic cancer in a subject, the method comprising determining the level or amount of methylation of the POU4F1 gene and / or the HOXD8 gene in a biological sample of the subject.

[0023] In a second aspect, the present invention relates to an in vitro method for monitoring the progression of pancreatic cancer in a specimen diagnosed for pancreatic cancer, the method comprising the steps of:

[0024] a) determining the level or amount of methylation of at least one gene selected from the group consisting of POU4F1 and HOXD8 in a biological sample of said subject at a first time point;

[0025] b) determining the level or amount of methylation of said at least one gene previously selected in step a) in a biological sample of said subject at a second time point; and

[0026] c) comparing the level or amount of methylation determined in step b) with the level or amount determined in step a) or with a reference value.

[0027] In a third aspect, the present invention relates to a method for determining or adjusting an appropriate treatment regimen for a subject diagnosed with pancreatic cancer, the method comprising the steps of:

[0028] a) determining the level or amount of methylation of the POU4F1 gene and / or the HOXD8 gene in a biological sample of the subject before or during treatment of said subject;

[0029] b) determining the level or amount of methylation of the POU4F1 gene and / or the HOXD8 gene in a biological sample of the subject after administering a treatment to the subject;

[0030] c) comparing the level or amount of methylation determined in step b) with the level or amount of methylation determined in step a) or with a reference value;

[0031] d) adjusting / altering the treatment regimen for the specimen based on the comparison of step c).

[0032] In another aspect, the present invention relates to a method for predicting clinical outcome in a subject suffering from pancreatic cancer, the method comprising the steps of:

[0033] a) determining the level or amount of methylation of the POU4F1 gene and / or the HOXD8 gene in a biological sample of said patient and comparing said level or amount with a reference value;

[0034] b) predicting clinical outcomes based on the comparison in step a).

[0035] In another aspect, the present invention relates to a kit comprising primers and / or probes specifically targeting the nucleotide region of SEQ ID NO: 1 in the POU4F1 gene or the nucleotide region of SEQ ID NO: 3 in the HOXD8 gene, and the kit preferably comprises: as primers, primers of SEQ ID NO: 5 and SEQ ID NO: 6 and / or SEQ ID NO: 7 and SEQ ID NO: 8, and / or as probes, probes of SEQ ID NO: 9 and / or SEQ ID NO: 10.

[0036] In another aspect, the present invention relates to a microarray comprising nucleotides specifically targeting the nucleotide region of SEQ ID NO: 1 in the POU4F1 gene or the nucleotide region of SEQ ID NO: 3 in the HOXD8 gene, and preferably comprising primers of SEQ ID NO: 5 and / or SEQ ID NO: 6, and / or SEQ ID NO: 7 and / or SEQ ID NO: 8.

[0037] In a further aspect, the present invention relates to the use of the kit or the microarray for:

[0038] a) diagnosing or identifying pancreatic cancer in a specimen;

[0039] b) predicting clinical outcome in specimens with pancreatic cancer;

[0040] c) determining a treatment regimen for a subject with pancreatic cancer; and / or

[0041] d) Monitoring the progression of pancreatic cancer in specimens diagnosed for pancreatic cancer.

[0042] The present invention is particularly suitable for detecting and determining hypermethylation of the POU4F1 gene, the HOXD8 gene, alone or preferably in combination, in order to help diagnose pancreatic cancer with high specificity at an early stage and to distinguish it from other types of cancer, or to follow up the progression of pancreatic cancer in patients who have previously been diagnosed. [Brief explanation of the drawings]

[0043] [Figure 1] We disclose the methylation levels of selected biomarkers HOXD8 and POU4F1 in DNA from tumor (n=20) and adjacent non-tumor (n=18) tissues obtained from pancreatic cancer patients by ddPCR. The Mann-Whitney test was used to analyze the differences in hypermethylation between DNA from normal and adjacent tissues. [Figure 2] A comparison of two analyses is presented: one for mutation detection using BPER NGS (MUT_SEQ) of ctDNA, and the other for methylation analysis of two candidate genes (METH_POS). The frequency of mutant sequences observed by BPER-NGS (Seq_Max, vertical axis) and the frequency of methylated sequences (ratio, horizontal axis) are shown. [Figure 3] Figure 1 shows the survival probability of patients according to their methylation status (validation cohort). METH_POS=0 corresponds to patients who are negative for methylated DNA, METH_POS=1 corresponds to patients who are positive. [Figure 4A] PRODIGE 35 cohort: overall and progression-free survival according to methylation DNA status. [Figure 4B] PRODIGE 35 cohort: overall and progression-free survival according to methylation DNA status. [Figure 5A]Overall survival according to treatment arm and methylation DNA status. Panel a: PRODIGE 35 Phase II trial results [Figure 5A1] Overall survival according to treatment arm and methylation DNA status. Panel a1: PRODIGE 35 patients with methylation DNA-negative status [Figure 5A2] Overall survival according to treatment arm and methylation DNA status. Panel a2: Patients with PRODIGE 35 methylation DNA positive status DETAILED DESCRIPTION OF THE INVENTION

[0044] definition

[0045] As intended herein, the term "comprising" has the meaning of "including" or "containing," and when an object "comprises" one or more elements, it means that other elements besides those listed are also included in the object. Conversely, when an object "consist of" one or more elements, it means that the object does not include elements other than those listed.

[0046] According to the present invention, the terms "subject," "individual," and "patient" are used interchangeably herein and refer to a mammal, whether healthy (no signs of pancreatic cancer), believed to be developing pancreatic cancer, suspected of having pancreatic cancer, or suffering from pancreatic cancer. For example, the subject may exhibit at least one of the following symptoms: jaundice and associated symptoms, such as dark urine, light or fatty stools, itchy skin; abdominal or back pain, weight loss and loss of appetite, nausea and vomiting, enlarged gallbladder or liver, blood clots, abnormalities in fatty tissue, or diabetes. The subject may also have previously suffered from pancreatic cancer, been treated, and be monitored for potential disease recurrence. The subject may also appear healthy but have a predisposition to developing pancreatic cancer, such as tobacco use, being overweight, obesity, genetic predisposition, or a family member of the subject who has or has previously suffered from the same disease. The subject is preferably human.

[0047] Typically, the diagnostic and monitoring methods described above involve the use of a biological sample obtained from a patient. As used herein, the term "biological sample" encompasses various sample types obtained from a specimen and can be used in a diagnostic or monitoring assay. Biological samples include, but are not limited to, blood and other liquid samples of biological origin (e.g., body fluids), or solid tissue samples (e.g., biopsy specimens). For example, biological samples include body fluids collected from an individual suspected of having pancreatic cancer, such as urine, pancreatic juice, feces, etc. Thus, biological samples include: clinical samples, urine, pancreatic juice, feces, stool, blood samples, plasma, and tissue samples, preferably blood samples.

[0048] In the field of cancer biomarkers, the most widely studied epigenetic modification is cytosine methylation. DNA methylation in the human genome occurs most frequently at the cytosine residue, at the carbon 5 position of CpG dinucleotides, resulting in 5-methylcytosine. CpG dinucleotides are rare in the human genome (approximately 1%). CpG dinucleotides are often found in clusters of more than 200 bases, where the G+C content is greater than 50% and the CpG frequency ratio is at least 0.6, known as CpG islands. Approximately 60% of human genomic promoters are associated with CpG islands. CpG islands within promoter regions are generally unmethylated in normal cells, with the exception of those involved in tissue differentiation. CpG island methylation is generally associated with transcriptional silencing. In cancer, both global hypomethylation (a reduction in DNA methylation overall) and local hypermethylation (e.g., promoter methylation and first exon methylation of tumor suppressor genes) are observed. DNA hypomethylation occurs in many genetic sequences (e.g., repetitive elements, retrotransposons, introns, and similar elements), resulting in genetic instability and potentially contributing to the activation of several proto-oncogenes, leading to loss of imprinting, as is the case for the IGF2 gene (encoding IGF-2) in Wilms' tumor.

[0049] To our knowledge, no studies have demonstrated DNA methylation biomarkers specifically for the diagnosis of pancreatic cancer.

[0050] However, the present inventors have found that specific regions within the promoter, gene body, or introns of two specific genes (i.e., POU4F1 and HOXD8) are specifically hypermethylated in pancreatic cancer tumor samples. Interestingly, these two genes have not previously been associated with liver cancer.

[0051] The "POU4F1 gene" of the present invention is a gene encoding a protein, POU domain, class 4, transcription factor 1 (POU4F1), also known as brain-specific homeobox / POU domain protein 3A (BRN3A), homeobox / POU domain protein RDC-1, or Oct-T1 protein. This DNA sequence is located on chromosome 13 (78,598,362-78,603,560), more precisely, on 13q31.1. This DNA sequence is also referred to as NC_000013.11. The present inventors have identified a DNA region, preferably contained between 79,176,072-79,177,072 (SEQ ID NO: 1), more specifically between 79,176,472-79,176,672 (SEQ ID NO: 2), which is significantly hypermethylated specifically in pancreatic tumor samples, in contrast to healthy samples and tumor samples from other cancers. Therefore, the present inventors propose to utilize this gene as a non-invasive, sensitive, and reliable method for diagnosing or identifying pancreatic cancer in specimens.

[0052] The "HOXD8 gene" of the present invention is a protein-coding gene encoding the homeobox protein Hox-D8 (HOXD8). This DNA sequence is located on chromosome 2 (176,129,694-176,132,695), more precisely at 2q31.1. This DNA sequence is designated NC_000002. The inventors have identified a DNA region, preferably contained between 176,994,063-176,995,088 (SEQ ID NO: 3), more specifically between 176,994,363-176,995,088 (SEQ ID NO: 4), that is significantly hypermethylated specifically in pancreatic tumor samples, in contrast to healthy samples and tumor samples from other cancers. Therefore, the present inventors propose to utilize this gene as a non-invasive, sensitive, and reliable method for diagnosing or identifying pancreatic cancer in a specimen.

[0053] The present inventors have shown that hypermethylation of these genes can be used as sensitive and specific biomarkers for pancreatic cancer in patients with pancreatic cancer, even at early stages. Hereinafter, the POU4F1 and HOXD8 genes are referred to as "biomarkers of the present invention." They can be used individually or in combination.

[0054] [Table 1-1] [Table 1-2] [Table 1-3]

[0055] Table 1: Useful sequences of interest in the POU4F1 and HOXD8 genes

[0056] Methods of the Invention

[0057] In a first aspect, the present invention relates to an in vitro method for diagnosing or identifying pancreatic cancer in a specimen, the method comprising: determining the level or amount of methylation of the POU4F1 gene or the HOXD8 gene in a biological sample of the specimen. In a preferred embodiment, the method comprises determining the level or amount of methylation in the nucleotide region of SEQ ID NO: 1 in the POU4F1 gene or the nucleotide region of SEQ ID NO: 3 in the HOXD8 gene. More preferably, the method comprises determining the level or amount of methylation in the nucleotide region of SEQ ID NO: 2 in the POU4F1 gene or the nucleotide region of SEQ ID NO: 4 in the HOXD8 gene.

[0058] In another embodiment, the biological sample according to the present invention is a body fluid of said subject, such as a urine, pancreatic juice, stool, or preferably a blood sample.

[0059] The method of the present invention also includes the following step: comparing the level or amount of methylation of the gene(s) with the level or amount of methylation of the same gene(s) determined in a reference sample. If the POU4F1 gene and / or the HOXD8 gene are significantly hypermethylated in the biological sample of the tested specimen compared to the same biomarkers in the reference sample, the tested specimen has pancreatic cancer or is at an elevated risk of having pancreatic cancer. An elevated risk of having pancreatic cancer can be confirmed by biopsy or imaging test.

[0060] Thus, in the methods of the present invention, if the gene is hypermethylated compared to a reference sample, the specimen is diagnosed or identified as suffering from pancreatic cancer. In other words, a high level or amount of methylation is considered an indicator of pancreatic cancer, aggressive high-grade pancreatic cancer, or recurrence of pancreatic cancer.

[0061] According to the present invention, a "reference sample" used to detect "hypermethylation" for the purpose of diagnosing pancreatic cancer or following up on the progression of pancreatic cancer is a biological sample. Specifically, the reference sample is a biological sample from a subject who does not have pancreatic cancer or a subject who has previously been diagnosed with pancreatic cancer. The biological sample may consist of normal or healthy cells (or both), tissues, body fluids, or a dataset generated using information from normal or healthy cells, tissues, or body fluids. For example, the reference sample may be genomic DNA, extracted from total blood or circulating DNA, which may be extracted from a healthy subject, where the sample is the same sample (e.g., plasma) as the sample being tested for the patient.

[0062] Preferably, the reference sample is a biological sample from a healthy subject with no signs or predisposition to pancreatic cancer. Alternatively, the reference sample is a biological sample from a subject previously diagnosed with pancreatic cancer, and the reference sample is a non-cancerous sample, for example, a biological sample from a patient with pancreatic cancer collected by biopsy near the tumor (without tumor cells). More preferably, the control biological sample is composed of a series of healthy subject samples, where the samples are the same samples (i.e., plasma or blood) as the samples tested for the patient. More preferably, the reference sample is blood or plasma obtained from a healthy subject. As used herein, a reference sample may be a previously collected sample from a subject with pancreatic cancer, and such a sample may be used to track the progression of pancreatic cancer from the subject by comparing it with the sample collected during evaluation.

[0063] In a more preferred embodiment, the reference sample according to the present invention is obtained from a healthy subject.

[0064] Several techniques have been proposed to detect DNA methylation. Delpu et al. (2013) review some of these techniques.

[0065] In methylation-specific PCR (MS-PCR), two sets of PCR primers are specifically designed to amplify methylated and unmethylated DNA regions of interest. Detection of PCR products was originally achieved by gel electrophoresis. This technique has been replaced by quantitative MS-PCR (qMS-PCR), in which PCR amplification is monitored in real time by the incorporation of fluorescent molecules. This modification allows for accurate quantification of DNA methylation levels in multiple specific regions and avoids the time-consuming electrophoresis step. Quantitative multiplex MS-PCR (QM-MS-PCR) and one-step MS-PCR (OS-MS-PCR) are also available, which coamplify specific genes in tissues from different sources or determine the DNA methylation levels of specific regions without DNA extraction procedures. qMS-PCR technology is simple, fast, inexpensive, highly sensitive, and easily standardized. It is currently one of the most frequently used techniques for cancer diagnosis in clinical applications. Methylation-sensitive high-resolution melting (MS-HRM) is based on the following fact: the nucleotide sequence of PCR products from bisulfite-treated DNA varies depending on the methylation status of the DNA region of interest. Methylation levels are determined by comparing melting dissociation curves with standard PCR products from the same region containing known methylated CpG sites. COBRA, an abbreviation for combined bisulfite restriction analysis, utilizes the ability of bisulfite conversion to create new restriction enzyme sites or maintain consensus sites for MSREs. After amplification, PCR products are cleaved with the appropriate MSRE. The proportion of cleaved PCR products is compared to uncleaved PCR products via polyacrylamide gel electrophoresis and image quantification software. This technique is reliably applied to DNA obtained from formalin-fixed, paraffin-embedded (FFPE) tissue samples. Furthermore, this approach allows for the assessment of DNA methylation in a large number of biological samples. More recently, high-throughput approaches have been developed.For example, Methyl Light is a high-throughput quantitative methylation assay that uses fluorescence-based real-time PCR (TaqMan®, Applied Biosystems, Forster City, CA, USA) in combination with bisulfite treatment. Also in combination with bisulfite treatment, pyrosequencing is a quantitative DNA sequencing method in which light is emitted as a result of an enzymatic reaction each time a nucleotide is incorporated into a growing DNA strand. These quantitative techniques detect small amounts of methylated DNA in heterogeneous DNA preparations. Easily standardized, rapid, and inexpensive, these techniques are increasingly being used for clinical purposes.

[0066] More recently, next-generation sequencing (NGS) technology has significantly improved the resolution of DNA methylation profiles. NGS can also be used to immunoprecipitate DNA fragments (also known as methyl-DNA immunoprecipitation sequencing, or MeDIP-seq). Finally, NGS allows for whole-genome sequencing after bisulfite conversion. In addition to these NGS approaches, high-throughput single-nucleotide polymorphism (SNP) genotyping systems are suitable for analyzing DNA methylation from bisulfite-converted genomic DNA. Other methods well known to those skilled in the art can also be used, including direct analysis of unmodified DNA (e.g., nanopore sequencing), specific restriction enzymes, methylation sequencing enrichment methods (e.g., EpiMark Methylation Enrichment Kit, New England), and immunoprecipitation.

[0067] All these methods for detecting DNA methylation are known in the art, e.g., qMS-PCR, MS-HRM, COBRA, MSRE, Methyl Light, NGS, SNP genotyping, pyrosequencing, microarray, ICE-cold PCR, etc., and can be used to determine the level or amount of methylation of the markers of the present invention.

[0068] The method of the present invention requires detecting the "level of methylation," "methylation level," or "amount of methylation," and depends on the detection technique used. According to the present invention, the term "level of methylation" or "amount of methylation" refers to that determined by a quantitative method. Thus, the terms "level of methylation" and "amount of methylation" can be used interchangeably.

[0069] "Hypermethylation" is determined, for example, when the methylation value (level or amount) of any of the biomarkers of the present invention in the biological sample of the specimen being tested is significantly higher than the methylation value of the corresponding biomarker measured in the reference sample. A significantly higher amount or level of methylation of at least one of the biomarkers of the present invention in the biological sample of the specimen compared to the normal amount or level of methylation in the reference sample suggests that the specimen being tested has pancreatic cancer or is at high risk of having pancreatic cancer.

[0070] "Amount or level of methylation is significantly higher" means that the amount or level of methylation is greater than the standard error of the assay used to assess the amount or level.

[0071] In a preferred embodiment of the invention, the methylation level is determined by next generation sequencing (NGS) or qPCR, preferably dPCR.

[0072] As used herein, the term "dPCR" is an abbreviation for "digital PCR" and can be used interchangeably with the term "ddPCR," which is an abbreviation for "droplet digital PCR." This PCR is an improved version of the traditional PCR method, in which a sample is separated into multiple partitions and a PCR reaction is performed individually in each partition. This separation allows for more reliable collection and sensitive measurement of nucleic acid abundance. This method has shown utility in studying variations in gene sequences (e.g., copy number variations and point mutations) and is routinely used for clonal amplification of samples for next-generation sequencing. More precisely, the PCR solution is divided into smaller reactions through water-oil emulsion technology and then run separately for PCR. For example, PCR samples can be partitioned into pico- to nanoliter-sized samples and encapsulated in oil droplets. The oil droplets are generated using a droplet generator, which applies a vacuum to each well. Depending on the system used, it is possible to generate 5-10 million picoliter droplets (25-50 μl samples) to 20,000 nanoliter droplets of oil (20 μl samples). It is also possible to use single-tube limited dilutions as well as other types of fractionation (e.g., microfractionated).

[0073] In a more preferred embodiment, the level or amount of methylation of the POU4F1 gene and / or the HOXD8 gene is determined using dPCR by using any one of the following: primers targeting the above-mentioned nucleotide region of the POU4F1 gene of SEQ ID NO: 1 or SEQ ID NO: 2, or the HOXD8 gene of SEQ ID NO: 3 or SEQ ID NO: 4; more specifically, primers of SEQ ID NO: 5, SEQ ID NO: 6, SEQ ID NO: 7, or SEQ ID NO: 8; or a combination of primers of SEQ ID NO: 5 and SEQ ID NO: 6, and / or a combination of primers of SEQ ID NO: 7 and SEQ ID NO: 8.

[0074] In another preferred embodiment, the level or amount of methylation of the POU4F1 gene and / or the HOXD8 gene is determined using dPCR by using any one of the following: a probe targeted to the above-mentioned nucleotide region of the POU4F1 gene of SEQ ID NO: 1 or SEQ ID NO: 2, or the HOXD8 gene of SEQ ID NO: 3 or SEQ ID NO: 4; more specifically, a probe of SEQ ID NO: 9 and / or SEQ ID NO: 10.

[0075] In another embodiment, the level or amount of methylation of the POU4F1 gene or the HOXD8 gene is determined using dPCR by using primers of SEQ ID NO: 5 and SEQ ID NO: 6, and / or SEQ ID NO: 7 and SEQ ID NO: 8; and using a probe targeted to the above nucleotide region of the POU4F1 gene of SEQ ID NO: 1 or SEQ ID NO: 2, or the HOXD8 gene of SEQ ID NO: 3 or SEQ ID NO: 4, more specifically, the probe of SEQ ID NO: 9 and / or SEQ ID NO: 10.

[0076] In a more preferred embodiment, the level or amount of methylation of the POU4F1 gene or the HOXD8 gene is determined using dPCR by using primers of SEQ ID NO: 5 and SEQ ID NO: 6, and / or SEQ ID NO: 7 and SEQ ID NO: 8, and using probes of SEQ ID NO: 9 and / or SEQ ID NO: 10.

[0077] According to a preferred embodiment of the present invention for enhancing the sensitivity and / or specificity of a diagnostic method, the inventors demonstrate that the use and combination of two biomarkers of the present invention can significantly enhance the sensitivity and / or specificity of the method, which preferably comprises simultaneously or sequentially determining the level or amount of methylation of the POU4F1 and HOXD8 genes in a biological sample of said subject.

[0078] Exemplary primers that can be used to determine hypermethylation of the POU4F1 gene are the primers of SEQ ID NO: 5 and / or SEQ ID NO: 6. Exemplary primers that can be used to determine hypermethylation of the HOXD8 gene are the primers of SEQ ID NO: 7 and / or SEQ ID NO: 8.

[0079] Primers that can be used in this preferred embodiment are shown in Table 2 below and SEQ ID NOs: 5 to 8.

[0080] [Table 2]

[0081] Table 2: Useful primers of interest according to the present invention for determining the biomarkers of the present invention by dPCR

[0082] In one exemplary embodiment, the method of the present invention may be carried out by using a gene expression assay (e.g., dPCR), which may use a fluorescent probe that allows for the detection of PCR products that accumulate during PCR. Such an assay may be, for example, a TaqMan® gene expression assay, which may involve detecting the T between a matched probe and a mismatched probe. m Probes containing minor groove binding (MGB) sites are used to enhance the difference between the two. Furthermore, these MGB probes may contain non-fluorescent quenchers (NFQs), which can enhance spectral resolution when multiple dyes are used in a reaction.

[0083] In accordance with a preferred embodiment of the present invention for enhancing the sensitivity and / or specificity of a diagnostic method, the inventors demonstrate that the sensitivity and / or specificity of the method can be significantly enhanced by using probes targeted to the POU4F1 gene and / or the HOXD8 gene.

[0084] An exemplary probe that can be used to determine hypermethylation of the POU4F1 gene is SEQ ID NO: 9. An exemplary probe that can be used to determine hypermethylation of the HOXD8 gene is SEQ ID NO: 10.

[0085] The above-mentioned probes that can be used in a preferred embodiment are shown in Table 3 and SEQ ID NOs: 9 to 10 below.

[0086] [Table 3]

[0087] Table 3: Probes of interest useful according to the present invention.

[0088] In another aspect, the biomarkers of the present invention can be used to monitor the progression of pancreatic cancer in vitro in a specimen diagnosed for pancreatic cancer, the method comprising the steps of:

[0089] a) determining the level or amount of methylation of at least one gene selected from the group consisting of POU4F1 and HOXD8 in a biological sample of the subject at a first time point;

[0090] b) determining the level or amount of methylation of said at least one gene previously selected in step a) in a biological sample of said specimen at a second time point; and

[0091] c) comparing the level or amount of methylation determined in step b) with the level or amount determined in step a) or with a reference value.

[0092] In a preferred embodiment of this aspect, two biomarkers of the invention can be used together, simultaneously or sequentially, to track the progression of pancreatic cancer in a specimen diagnosed for pancreatic cancer. Thus, the POU4F1 gene and the HOXD8 gene can be used to practice this aspect of the invention.

[0093] It can be concluded that the pancreatic cancer malignancy has worsened if the level or amount of methylation determined in step b) is significantly higher than the level or amount determined in step a), in other words, the tested specimen has a disease that is progressing in a negative direction, even if the specimen may have already been treated.

[0094] If the subject is undergoing treatment, the first sample may be taken from the subject before the treatment for pancreatic cancer, and the second sample may be taken from the subject after the treatment for pancreatic cancer, so the first time point is preferably before the subject is treated and the second time point is after the subject has been treated for pancreatic cancer.

[0095] In such cases, the methods of the present invention can be used to assess the efficacy of said treatment.

[0096] In the context of the present invention, if methylation of the HOXD8 gene or the POU4F1 gene, or a combination of these genes, is significantly reduced in a biological sample obtained after administration of a treatment compared to the same biomarker in a sample obtained before the treatment, then the tested subject is likely to have responded effectively to the tested treatment (and is therefore a "responder").

[0097] Conversely, if methylation of the HOXD8 gene or the POU4F1 gene, or a combination of these genes, is significantly elevated in a biological sample obtained after administration of a treatment compared to the same biomarker(s) in a sample obtained before the treatment, then the tested specimen likely did not respond effectively to the tested treatment (and is therefore a "non-responder").

[0098] As used herein, a "non-responder" is considered to be a patient with progressive or stable disease as defined according to RECIST 1.1 criteria.

[0099] As used herein, "treatment" can include some combination of surgery, chemotherapy, radiation therapy, and targeted therapy. In a preferred embodiment, the treatment is surgery.

[0100] In certain embodiments, the sample in step a) is obtained before treatment for pancreatic cancer and the sample in step b) is obtained after the subject has received treatment for pancreatic cancer.

[0101] In another specific embodiment, the subject has previously been treated for pancreatic cancer, and the biomarkers of the present invention can be used to assess the efficacy of the treatment. Thus, the level or amount of methylation is determined once to determine whether the subject's level or amount of methylation has decreased compared to the previous level or amount of methylation determined before treatment. In another aspect, the level or amount of methylation is determined once to determine whether hypermethylation of POU4F1 or HOXD8, alone or in combination, has occurred in the subject. If the level or amount of methylation has decreased in the subject, the treatment has been successful. Alternatively, if hypermethylation has not occurred in the subject, the subject no longer has pancreatic cancer and has been treated.

[0102] In such a case, according to a particular embodiment, the subject is treated for pancreatic cancer in step a), and the level or amount of methylation is determined only once in step b) after the subject has been treated for pancreatic cancer in step a), so that the level or amount of methylation is determined only once in step b), and determining the level or amount of methylation for previously diagnosed subjects is optional.

[0103] In certain embodiments, the reference value is obtained in healthy subjects.

[0104] In another aspect, the biomarkers of the present invention can be used in a method for determining or adjusting an appropriate treatment regimen for a subject diagnosed with pancreatic cancer, the method comprising the steps of:

[0105] a) determining the level or amount of methylation of the POU4F1 gene and / or the HOXD8 gene in a biological sample of said subject before or during treatment of said subject;

[0106] b) determining the level or amount of methylation of the POU4F1 gene and / or the HOXD8 gene in a biological sample of the subject after administering a treatment to the subject;

[0107] c) comparing the level or amount of methylation determined in step b) with the level or amount of methylation determined in step a) or with a reference value;

[0108] d) adjusting / altering the treatment regimen for the specimen based on the comparison of step c).

[0109] In a preferred embodiment of this aspect, two biomarkers of the present invention may be used together, simultaneously or sequentially, to determine or adjust a treatment regimen for a subject diagnosed with pancreatic cancer. Thus, the POU4F1 gene and the HOXD8 gene may be used to practice this aspect of the present invention.

[0110] Specifically, the treatment regimen is likely to be highly effective if the level or amount of the biomarker(s) is significantly lower than the level or amount of the biomarker(s) determined before the treatment.

[0111] Specifically, the treatment regimen should be modified if: the level or amount of the biomarker(s) remains significantly higher than the level or amount of the biomarker(s) determined before the treatment.

[0112] In another aspect, the biomarkers of the present invention can be used to predict the outcome of a patient with pancreatic cancer. These biomarkers can also be used to assist skilled cancer practitioners in selecting an appropriate treatment to maximize patient survival. Suitable treatments include, for example, chemotherapy, immunotherapy, radiation therapy, and / or surgery. Specifically, the patient has been or will be treated with surgery.

[0113] Thus, the biomarkers of the present invention can be used in a method for predicting clinical outcome in a subject suffering from pancreatic cancer, the method comprising the steps of:

[0114] a) determining the level or amount of methylation of the POU4F1 gene and / or the HOXD8 gene in a biological sample of the patient and comparing said level or amount with a reference value;

[0115] b) predicting clinical outcomes based on the comparison in step a).

[0116] If the HOXD8 gene or the POU4F1 gene, or a combination of these genes, is significantly hypermethylated in the biological sample of the test subject compared to the same biomarker in the reference sample, the test subject is more likely to have an adverse clinical outcome.

[0117] Conversely, if the HOXD8 gene or the POU4F1 gene, or a combination of these genes, shows a similar level of methylation in the biological sample of the tested specimen compared to the same biomarker in a reference sample (from a healthy specimen or from a patient not suffering from gastric cancer), the tested specimen is more likely to have a favorable clinical outcome.

[0118] Conversely, if the HOXD8 gene or POU4F1 gene, or a combination of these genes, in the biological sample of the tested specimen is not significantly different in degree of hypermethylation compared to the same biomarkers in a reference sample of the same specimen, the tested specimen is likely to have a favorable clinical outcome.

[0119] Kits of the Invention

[0120] Furthermore, the present invention provides diagnostic and monitoring tools for determining the hypermethylated biomarkers of the present invention for the purpose of diagnosing or monitoring pancreatic cancer.

[0121] First, the present invention relates to nucleic acids useful for detecting the hypermethylated regions. Within the scope of the present invention, "nucleic acids" refers to mRNA, genomic DNA, or cDNA derived from mRNA. These nucleic acids are preferably primers or probes.

[0122] In a preferred embodiment, the present invention relates to a kit, the kit comprising a primer that specifically targets the nucleotide region of SEQ ID NO: 1, more preferably that specifically targets the nucleotide region of SEQ ID NO: 2 in the POU4F1 gene, or that specifically targets the nucleotide region of SEQ ID NO: 3, more preferably that specifically targets the nucleotide region of SEQ ID NO: 4 in the HOXD8 gene, and the kit preferably comprises a primer of SEQ ID NO: 5 or SEQ ID NO: 6, or SEQ ID NO: 7 or SEQ ID NO: 8.

[0123] According to a preferred embodiment, the kit comprises a combination of primers, more preferably selected from the group of SEQ ID NO: 5, SEQ ID NO: 6, SEQ ID NO: 7, SEQ ID NO: 8. The combinations preferably used according to the present invention are SEQ ID NO: 5 and SEQ ID NO: 6 or SEQ ID NO: 7 and SEQ ID NO: 8.

[0124] In a specific embodiment, the kit comprises a primer specifically targeting the nucleotide region of SEQ ID NO: 1 in the POU4F1 gene or the nucleotide region of SEQ ID NO: 3 in the HOXD8 gene, preferably a primer of SEQ ID NO: 5 or SEQ ID NO: 6, or SEQ ID NO: 7 or SEQ ID NO: 8.

[0125] In another aspect, the present invention relates to a kit comprising a probe, the probe comprising a probe that specifically targets the nucleotide region of SEQ ID NO: 1, more preferably a probe that specifically targets the nucleotide region of SEQ ID NO: 2 in the POU4F1 gene, or a probe that specifically targets the nucleotide region of SEQ ID NO: 3, more preferably a probe that specifically targets the nucleotide region of SEQ ID NO: 4 in the HOXD8 gene, the kit preferably comprising probes of SEQ ID NO: 9 and / or SEQ ID NO: 10.

[0126] In another aspect, the present invention relates to a kit comprising primers and a probe that specifically target the nucleotide region of SEQ ID NO: 1 in the POU4F1 gene or the nucleotide region of SEQ ID NO: 3 in the HOXD8 gene. In a preferred embodiment, the kit comprises primers and a probe that specifically target the nucleotide region of SEQ ID NO: 2 in the POU4F1 gene or the nucleotide region of SEQ ID NO: 4 in the HOXD8 gene.

[0127] As used herein, the term "primer" refers to an isolated nucleic acid molecule capable of specifically hybridizing or annealing to the 5' or 3' region of a targeted genomic region (positive and negative strands, respectively, or vice versa). Typically, primers are about 10-30 nucleotides in length and anneal to both ends of a region comprising about 50-200 nucleotides in length. Under appropriate conditions and with appropriate reagents, such primers enable amplification of a nucleic acid molecule comprising the nucleotide sequence flanked by the primers. In another aspect of the present invention, primers can be used in pairs, which are often referred to as "primer pairs" or "primer sets."

[0128] As used herein, the term "probe" refers to a molecule that can specifically hybridize to a genomic region of interest. Probes are useful for highlighting the presence of that genomic region in a biological sample. These probes can contain at least one non-natural nucleotide, such as peptide nucleic acid (PNA), phosphate-containing peptide nucleic acid (PHONA), bridged or locked nucleic acid (BNA or LNA), and morpholino nucleic acid. Non-natural nucleotides also include chemically modified nucleic acids or nucleic acid analogs, such as methylphosphonate DNA or RNA, phosphorothioate DNA or RNA, phosphoramidate DNA or RNA, and 2'-O-methyl DNA or RNA.

[0129] In a preferred embodiment, the probes of the present invention comprise at least 15 consecutive nucleotides, which are complementary to their bisulfite equivalents or fragments. In a further preferred embodiment, molecules that can be used as probes according to the present invention have a minimum total size of 15 nucleotides. In a further preferred embodiment, these molecules comprise 15-30 nucleotides (total).

[0130] For specific uses, the probes and / or primers of the present invention may be directly or indirectly labeled with a detectable label. The label may be of any type, depending on the experiment being performed. The label may be a radioisotope (e.g., 32 P, 33 P, 35 S, 3 H or 125 The label may be a radioactive material selected from the group consisting of radioactive materials (e.g., biotin, avidin, or streptavidin), digoxigenin, haptens, colorants, and fluorescent agents (e.g., radioluminescent, chemiluminescent, bioluminescent, fluorescent dyes, or phosphorescent agents). Preferably, 6-carboxyfluorescein (FAM) and VIC are used. Unlabeled polynucleotide sequences can also be used directly as probes or primers, for example, in PCR-based processes (e.g., quantitative PCR).

[0131] In a particular aspect, the probes of the present invention are TaqMan probes of SEQ ID NOs: 9-10.

[0132] In another aspect, the present invention relates to a microarray comprising nucleotides that specifically target the nucleotide region of SEQ ID NO: 1 in the POU4F1 gene or the nucleotide region of SEQ ID NO: 3 in the HOXD8 gene, and preferably comprising primers of SEQ ID NO: 5, and / or SEQ ID NO: 6, and / or SEQ ID NO: 7, and / or SEQ ID NO: 8, and / or SEQ ID NO: 9, and / or SEQ ID NO: 10. In a preferred embodiment, the microarray comprises nucleotides that specifically target the nucleotide region of SEQ ID NO: 2 in the POU4F1 gene or the nucleotide region of SEQ ID NO: 4 in the HOXD8 gene, and preferably comprising primers of SEQ ID NO: 5, and / or SEQ ID NO: 6, and / or SEQ ID NO: 7, and / or SEQ ID NO: 8, and / or SEQ ID NO: 9, and / or SEQ ID NO: 10.

[0133] These exemplary primers and probes are summarized in Tables 2 and 3 above.

[0134] The present invention also relates to kits and / or microarrays that can be used to:

[0135] a) diagnosing or identifying pancreatic cancer in a specimen;

[0136] b) predicting clinical outcome in subjects with pancreatic cancer;

[0137] c) determining or adjusting a treatment regimen for a subject suffering from pancreatic cancer; and / or

[0138] d) Monitoring the progression of pancreatic cancer in specimens diagnosed for pancreatic cancer.

[0139] Further aspects and advantages of the present invention are disclosed in the following examples, which should be considered as illustrative only. [Example]

[0140] Materials and Methods

[0141] 1. Identification of DNA methylation biomarkers and hypermethylation in tissue sample DNA and circulating cell-free DNA (ccfDNA) from healthy plasma

[0142] 1.1 Patients and healthy control individuals

[0143] Twenty patients with pancreatic cancer were included. Matched tumor and adjacent non-tumor tissue biopsies were collected from each patient. The local ethics committee approved the study, and written informed consent was obtained from all patients.

[0144] 1.2 Tumor sample preparation, storage, DNA extraction, and quantification

[0145] Tumor and adjacent nontumor biopsies were rapidly flash-frozen in liquid nitrogen after resection until further analysis. Each tumor was reviewed by a pathologist, and tumor cell content was assessed by hematoxylin-eosin-saffron staining. DNA was extracted using the QIAamp DNA Mini Kit (Qiagen) according to the manufacturer's instructions. DNA concentration was measured using the dsDNA BR Assay (Invitrogen) with a Qubit 2.0 fluorometer (Invitrogen, Life Technologies). Extracted DNA was stored at -20°C.

[0146] 1.3 Plasma sample preparation, storage, DNA extraction, and quantification

[0147] Plasma samples from healthy individuals were received on dry ice, aliquoted, and quickly frozen at -80°C. Prior to extraction, plasma samples were centrifuged at 3000 g for 10 minutes and then extracted using a QIAmp Circulating Nucleic Acid Kit (Qiagen) or Maxwell according to the manufacturer's instructions. DNA content was measured using the dsDNA HS Assay (Invitrogen) with a Qubit 2.0 fluorometer (Invitrogen, Life Technologies). Extracted DNA was stored at -20°C prior to testing.

[0148] 1.4 Selection of candidate biomarkers for pancreatic cancer patients based on the public methylation database TCGA

[0149] Methylation data from pancreatic cancer patients from The Cancer Genome Atlas (TCGA) (http: / / cancergenome.nih.gov / ) were analyzed using a homemade R script (see above). A list of 485,577 CpGs associated with a gene and the methylation levels of that gene in healthy and tumor tissues was generated; differences in methylation levels (fold changes) between healthy and tumor tissues; correlations between tumor and normal cell abundance in cancer patients; and Wilcoxon tests and statistics. Based on this information, we selected two candidate genes (HOXD8 and POU4F1) containing CpGs that were significantly differentially methylated between tumor and healthy pancreatic tissues.

[0150] 1.5 Bisulfite conversion of tissue DNA and plasma ccfDNA

[0151] Tissue or plasma DNA was modified with bisulfite using the EZ DNA Methylation-Gold Kit (Zymo Research). Briefly, the bisulfite reaction was carried out in a thermocycler at 98°C for 12 minutes and 64°C for 2 hours and 35 minutes. Cleanup of the bisulfite-converted DNA was performed according to the manufacturer's recommendations, and the converted DNA was eluted with M-Elution Buffer and stored at -20°C.

[0152] 1.6 Detection of methylation changes of selected biomarkers by droplet-based digital PCR

[0153] Hypermethylation of selected biomarkers in tumor DNA was verified by ddPCR (Raindrop or QX-200 systems, BIO-RAD Technologies) as previously described [Garrigou, S., et al. 2016]. Duplex format was used to analyze hypermethylation relative to albumin for DNA normalization purposes. Primers and probes are listed in Tables 2 and 3. Briefly, 12.5 μL of Kapa probe Fast qPCR master mix (Kapa Biosystems) was mixed into an assay solution containing 0.75 μL of 40 mM dNTP Mix (New England BioLabs), 0.5 μL of 25 mM MgCl2, 1 μL of 25x Droplet Stabilizer (RainDance Technologies), 1.25 μL of 20x Assay Mix (containing 8 μM forward and reverse primers, 4 μM 6-FAM, and 12 μM VIC Taqman®-labeled probe), and the target modified DNA template in a final reaction volume of 25 μL. When possible, a minimum of 10 ng of modified DNA was used in each reaction.

[0154] The limit of blank (LOB) was calculated as previously described [Taly, V., et al. 2013]. The LOB was defined as the frequency of positive droplets measured in normal control DNA samples with the presence of non-hypermethylated DNA. The calculated LOB was subtracted from each sample to calculate its methylation level.

[0155] Sample analysis was performed according to a previously described procedure [Taly, V., et al. 2013]. A sample was considered positive if the number of droplets observed was greater than the LOB value. The methylation level of each sample was calculated as the ratio of the number of droplets containing methylated sequences to the number of droplets containing albumin sequences.

[0156] To confirm that the modification process was properly performed, two DNA controls were used (positive control: universal hypermethylated DNA, and negative control: normal human genomic DNA).

[0157] 1.7 Measurement of methylation levels of selected biomarkers in plasma circulating cell-free DNA (ccfDNA) from healthy individuals

[0158] DNA methylation of selected biomarkers in plasma from healthy individuals was measured by ddPCR (Raindrop or QX-200 systems, BIO-RAD Technologies) using the same reaction conditions as above. To normalize DNA content, a duplex format was used to analyze hypermethylation relative to albumin. Primers and probes were the same as those listed in Tables 2 and 3.

[0159] 1.8 Calculating Sensitivity and Specificity of Detection

[0160] The sum of the mean and standard deviation of methylation levels in non-tumor tissue DNA was used as the threshold to calculate the sensitivity and specificity of each selected biomarker. Sensitivity is the percentage of patients showing methylation levels in tumor tissues higher than the threshold. Specificity is the percentage of patients showing methylation levels in non-tumor tissues lower than the threshold.

[0161] 2. Validation of the two cohorts

[0162] 2.1 The first cohort, called the "validation cohort"

[0163] 2.1.1 Patient Plasma Sample Preparation, Storage, DNA Extraction, and Quantitation

[0164] Between January 2011 and June 2018, plasma samples were prospectively collected in our oncology department from all consecutive patients with histologically proven metastatic PAC who underwent the first chemotherapy protocol (n = 100). Blood samples were collected immediately before the first cycle of chemotherapy. All patients signed informed consent forms, and approval was obtained from the ethics committee (CPP Ile-de-France 2014 / 59NICB). The following data were collected in the prospective database: clinical and pathological characteristics (gender, age, medical history, date of diagnosis, location of initial tumor, diameter of initial tumor, tumor differentiation grade, disease stage), follow-up data (date of initial resection, date and type of recurrence, date of metastatic disease diagnosis, date and type of chemotherapy regimen, date of death or last follow-up), and biological data (CEA, CA 19-9, albuminemia, bilirubinemia).

[0165] Patient blood samples (9 mL) were drawn from the central catheter and placed in EDTA tubes. Collected samples were centrifuged at 3500 rpm for 15 minutes at 4°C within 3 hours of blood drawing. Plasma was stored at -80°C until further use. DNA was extracted from plasma using the QIAamp® Circulating Nucleic Acid kit (Qiagen, Hilden, Germany) according to the manufacturer's instructions. Incubation with proteinase K was performed for 30 minutes at 68°C. DNA extracted from 2 mL of plasma was eluted with 50 μL of buffer AVE and stored at -80°C. DNA content was assessed using the Qubit™ dsDNA HS (High Sensitivity) Assay kit (Thermo Fisher).

[0166] 2.1.2 Droplet-based digital PCR

[0167] All patient plasma samples were screened for two methylation-selective genes by ddPCR using the BIO-RAD ddpcr Quantasoft® system (Biorad Technologies). Using this system, single target DNA molecules were compartmentalized in droplets with validated fluorescent TaqMan™ probes, as described above. After testing all samples, analysis was performed using Quantasoft Software (version 1.7, Biorad Technologies) according to standard procedures, as described above. Samples were considered positive for ctDNA if they tested positive for two genetic markers. Samples positive for only one marker gene were further analyzed by next-generation sequencing. Samples that were also positive for the presence of cancer-associated mutations by NGS were further considered positive for ctDNA.

[0168] 2.1.3 Next Generation Sequencing (NGS)

[0169] All patient plasma samples were tested by next-generation sequencing (NGS) to correlate methylation and mutation status. Sequencing libraries were prepared from circulating free DNA using the Ion AmpliSeq™ Colon and Lung Cancer Research Panel v2 (Thermo Fisher). Ten nanograms of DNA per sample was used as input for library preparation using the Ion AmpliSeq™ Library Kit 2.0 (Thermo Fisher) according to the manufacturer's protocol. Pooled barcoded libraries (up to 96) were processed using the Ion PI Hi-Q Chef Kit (A27198) on the Ion Chef™ System and sequenced using the Ion PI Chip Kit v3 (A26771) on the Ion Proton™ System. The NGS assay was specifically developed to detect low-frequency allelic mutations, with sensitivity and specificity validated in positive and negative controls. Samples were analyzed using the BPER procedure, which was developed by the inventors (see Pecuchet et al. Clinical Chemistry 2016 & Pietrasz et al. Clinical Cancer Research 2017).

[0170] 2.2 Validation of the prospective cohort (Prodige 35)

[0171] This study aimed to evaluate the prognostic value of methylated tumor-derived circulating DNA (Met-DNA) in metastatic pancreatic adenocarcinoma (mPAC). The prognostic value of Met-DNA was evaluated in a prognostic cohort (validation cohort) of mPAC, correlated by NGS, and in one independent prognostic validation cohort from two randomized phase II trials (PRODIGE 35).

[0172] PRODIGE 35 - PANOPTIMOX: Patients were randomized to receive either: 6m FOLFIRINOX (arm A), 4m FOLFIRINOX followed by LV5FU2 maintenance therapy for control patients (pts) and treatment re-initiated at disease progression (arm B), or sequential treatment alternating gemcitabine and FOLFIRI.3 every 2m (arm C).

[0173] [Table 4]

[0174] Table 4: Details of included cohorts

[0175] statistical analysis

[0176] Statistical analysis was performed using SPSS software version 21.0 (SPSS Inc., Chicago, IL) and R Studio (RStudio: Integrated Development for R. RStudio, Inc., Boston, MA). A P value of ≤0.05 was considered significant. The Mann-Whitney test was used to analyze differences in hypermethylation between normal and adjacent tissues.

[0177] result

[0178] Identification of DNA methylation biomarkers based on TCGA analysis and validation

[0179] Based on the results of the database analysis, we were able to select 10 CpG sites that met the established criteria (i.e., DNA methylation levels in tumor tissues, differences in DNA methylation levels between tumor and normal tissues, correlation between the observed frequency of DNA methylation and the tumor cell content of the sample, and significance (P value) of differences in DNA methylation levels between tumor and normal tissues). These CpG sites were present in the POU4F1, HOXD8, RYR2, XKR4, KCNA3, and PITX2 genes. To validate our method, we selected the POU4F1 and HOXD8 genes. HOXD8 was selected because it contained five CpG sites among the 10 found. The POU4F1 gene was selected because it contained the CpG site with the lowest average methylation in DNA extracted from adjacent pancreatic tissue.

[0180] Validation of DNA methylation changes of selected biomarkers by dPCR

[0181] We examined the differences in DNA methylation of several potential biomarkers in tumor and non-tumor tissues from pancreatic cancer patients by ddPCR (n = 20). DNA methylation in tumor tissues was significantly increased compared with that in non-tumor tissues (Figure 1).

[0182] The sensitivity and specificity of detecting the biomarkers HOXD8 and POU4F1 alone by ddPCR in pancreatic cancer was greater than 70% (Table 5).

[0183] [Table 5]

[0184] Table 5: Sensitivity and specificity of biomarkers of the invention when evaluated as single markers by ddPCR

[0185] Combining different biomarkers to obtain higher sensitivity and specificity

[0186] The sensitivity and specificity of each biomarker by ddPCR are shown in Table 5. The highest sensitivity when using a single biomarker is 80%. To obtain higher detection sensitivity and specificity, different biomarkers can be combined (Table 6). By combining HOXD8 and POU4F1, the detection sensitivity can be further improved to 89%.

[0187] [Table 6]

[0188] Table 6: Detection sensitivity and specificity of selected biomarker combinations

[0189] Methylation profiles by ddPCR of selected biomarkers in plasma ccfDNA from healthy individuals

[0190] The purpose of these biomarkers is to detect DNA hypermethylation in pancreatic cancer patients, but not in healthy individuals. To validate these biomarkers, methylation levels were examined by ddPCR in plasma ccfDNA from healthy individuals (n = 12). No positive results were detected when two markers were used in combination. One sample showed low positivity with a single marker (HOXD8, 3 positive droplets (lower than the LOD), corresponding to 0.4% methylated DNA).

[0191] Additionally, DNA extracted from whole blood (commercial DNA from Promega) was also used to validate the assay and confirm that the marker was not positive in cells contained in blood, thereby ensuring that no false positives would occur in the event of hemolysis of blood cells (e.g., due to sample handling before analysis).

[0192] Comparison of ctDNA detection using two biomarkers and optimized NGS for high sensitivity (BPER analysis)

[0193] Sixty-four samples from the validation cohort, previously analyzed by BPER-NGS, were tested by dPCR for the detection of ctDNA by detecting the methylation of two candidate genes. Of the 64 patients, 20 were negative by both methods, 41 were positive by both methods, two were positive only by dPCR methylation analysis, and one was positive only by BPER-NGS. It is important to note that this last result relates to samples containing very low amounts of ctDNA. Therefore, the negative results may likely be due to the loss of ctDNA by bisulfite treatment, and therefore, more DNA may be involved in the analysis here (hence, leading to a positive result). However, these results clearly demonstrate that the results obtained by analyzing two candidate genes are comparable to those obtained using targeted and optimized NGS for the analysis of multiple gene mutations.

[0194] [Table 7]

[0195] Table 7: Comparison of analysis of ctDNA by mutation detection using BPER NGS (MUT_SEQ) or methylation analysis with two gene candidate analysis (METH_POS). 0: negative samples, 1: positive samples

[0196] Finally, we also show the correlation between the frequency of ctDNA determined by mutation-based BPER-NGS analysis (targeted sequencing) and the frequency in methylation-based dPCR analysis (using two marker genes by the inventors) (Figure 2).

[0197] Monitoring the progression of pancreatic cancer

[0198] The methylation status of circulating cell-free DNA, as determined using the two developed markers, was also shown to correlate with overall survival (Figure 3, validation cohort; Figure 4, Prodige 35 cohort). For the first cohort (Figure 3), 100 patients (validation cohort) were tested for the presence of methylated tumor DNA using the methods described above. 64% were positive. For MetDNA (METH_POS=0)-negative patients, the overall survival rate was 17.1 months, and for MetDNA (METH_POS=1)-positive patients, the overall survival rate was 5.5 months. Similarly, for the Prodige 35 cohort, 177 patients were tested, and the presence of ctDNA, as detected by the developed methylation markers, was correlated with overall survival. The relationship between these markers was further validated using univariate and multivariate analyses. Furthermore, in multivariate analysis, adjusted for sex, age, CA19.9 > 40 UI / mL, treatment arm, number of metastatic sites, and stratified on center, Met-DNA was independently associated with little to no OS in the Prodige 35 cohort (see Table 8 and also Figure 5).

[0199] [Table 8]

[0200] Table 8: Univariate and multivariate analyses of components of overall survival in the Prodige 35 cohort

[0201] conclusion

[0202] These studies (validation and Prodige 35 cohorts) demonstrate that Met-DNA is an independent and powerful prognostic marker in mPAC. These results argue for stratifying patients by ctDNA status for further randomized trials.

[0203] <References> Bouvier AM, Remontet L, Jougla E, et al. Incidence of gastrointestinal cancers in France. Gastroenterologie clinique et biologique. 2004;28(10 Pt 1):877---881. Coleman MP, Gatta G, Verdecchia A, et al. EUROCARE---3 summary: cancer survival in Europe at the end of the 20th century. Annals of oncology : official journal of the European Society for Medical Oncology / ESMO. 2003;14 Suppl 5:v128---149. Eheman C, Henley SJ, Ballard---Barbash R, et al. Annual Report to the Nation on the status of cancer, 1975---2008, featuring cancers associated with excess weight and lack of sufficient physical activity. Cancer. 2012;118(9):2338---2366. Burris HA, 3rd, Moore MJ, Andersen J, et al. Improvements in survival and clinical benefit with gemcitabine as first---line therapy for patients with advanced pancreas cancer: a randomized trial. Journal of clinical oncology: official journal of the American Society of Clinical Oncology. 1997;15(6):2403---2413. Conroy T, Desseigne F, Ychou M, et al. FOLFIRINOX versus gemcitabine for metastatic pancreatic cancer. The New England journal of medicine. 2011;364(19):1817---1825. Von Hoff DD, Ervin T, Arena FP, et al. Increased survival in pancreatic cancer with nab---paclitaxel plus gemcitabine. The New England journal of medicine. 2013;369(18):1691---1703. Siegel RL, Miller KD, Jemal A. Cancer statistics, 2015. CA: a cancer journal for clinicians. 2015;65(1):5---29. Bardeesy N, DePinho RA. Pancreatic cancer biology and genetics. Nature reviews. Cancer. 2002;2(12):897---909. Almoguera C, Shibata D, Forrester K, Martin J, Arnheim N, Perucho M. Most human carcinomas of the exocrine pancreas contain mutant c---K---ras genes. Cell. 1988;53(4):549---554. Tada M, Omata M, Kawai S, et al. Detection of ras gene mutations in pancreatic juice and peripheral blood of patients with pancreatic adenocarcinoma. Cancer research. 1993;53(11):2472---2474. 41 Morris JPt, Wang SC, Hebrok M. KRAS, Hedgehog, Wnt and the twisted developmental biology of pancreatic ductal adenocarcinoma. Nature reviews. Cancer. 2010;10(10):683---695. Kanda M, Matthaei H, Wu J, et al. Presence of somatic mutations in most early---stage pancreatic intraepithelial neoplasia. Gastroenterology. 2012;142(4):730---733 e739. Hruban RH, Maitra A, Schulick R, et al. Emerging molecular biology of pancreatic cancer. Gastrointestinal cancer research : GCR. 2008;2(4 Suppl):S10---15. Moore PS, Beghelli S, Zamboni G, Scarpa A. Genetic abnormalities in pancreatic cancer. Molecular cancer. 2003;2:7. Poruk KE,Gay DZ, Brown K, Mulvihill JD, BoucherKM, Scaife CL, et al. The clinical utility of CA 19-9 in pancreatic adenocarcinoma: diagnostic and prognostic updates. Curr Mol Med 2013;13:340-51. Dabritz J, Preston R, Hanfler J, Oettle H. Follow-up study of K-ras mutations in the plasma of patients with pancreatic cancer: correlation with clinical features and carbohydrate antigen 19-9. Pancreas 2009;38:534-41. Mikeska T, et al. DNA methylation biomarkers in cancer: progress towards clinical implementation. Expert Rev Mol Diagn. 2012 Jun; 12(5):473-87. Delpu Y et al, DNA methylation and cancer diagnosis. Int J Mol Sci. 2013 Jul 18;14(7):15029-58 Pecuchet et al. Clinical Chemistry 2016 & Pietrasz et al. Clinical Cancer Research 2017

Claims

1. 1. An in vitro method for determining a level or amount of methylation, the method comprising: determining the level or amount of methylation of the POU4F1 gene in a biological sample of a subject; wherein said biological sample is derived from a subject suspected of having pancreatic cancer.

2. 10. The method of claim 1, further comprising determining the level or amount of methylation of the HOXD8 gene in the subject's biological sample.

3. 3. The method of claim 1 or 2, wherein the biological sample is a bodily fluid of the subject.

4. The method according to any one of claims 1 to 3, wherein the biological sample is blood, urine, pancreatic juice, or stool.

5. The method according to any one of claims 1 to 4, wherein the biological sample is blood.

6. 6. The method of any one of claims 1 to 5, wherein the methylation is determined by next generation sequencing (NGS) or quantitative PCR (qPCR).

7. The method of any one of claims 1 to 6, wherein the methylation is determined by digital PCR (dPCR).

8. The method according to any one of claims 1 to 7, wherein the level or amount of methylation is determined in the nucleotide region of SEQ ID NO: 1 in the POU4F1 gene.

9. 9. The method according to any one of claims 1 to 8, wherein the level or amount of methylation of the POU4F1 gene is determined by dPCR by using primers of SEQ ID NO: 5 and SEQ ID NO: 6 and by using a probe of SEQ ID NO:

9.

10. 10. The method according to any one of claims 2 to 9, wherein the level or amount of methylation is determined in the nucleotide region of SEQ ID NO: 1 in the POU4F1 gene and in the nucleotide region of SEQ ID NO: 3 in the HOXD8 gene.

11. 10. The method according to any one of claims 2 to 9, wherein the level or amount of methylation of the POU4F1 gene and the HOXD8 gene is determined using dPCR by using primers of SEQ ID NO:5 and SEQ ID NO:6, SEQ ID NO:7 and SEQ ID NO:8; and using probes of SEQ ID NO:9 and SEQ ID NO:

10.

12. 12. The method according to any one of claims 2 to 11, comprising simultaneously or sequentially determining the level or amount of methylation of the POU4F1 gene and the HOXD8 gene in a biological sample of the subject.

13. 1. An in vitro method for analyzing a level or amount of methylation, comprising: a) determining the level or amount of methylation of the POU4F1 gene in a biological sample at a first time point; b) determining the level or amount of methylation of the POU4F1 gene in the biological sample at a second time point; and c) comparing the level or amount of methylation determined in step b) with the level or amount determined in step a) or with a reference value; Here, the biological sample is derived from a subject suspected of having or developing pancreatic cancer.

14. 14. The method of claim 13, further comprising determining the level or amount of methylation of the HOXD8 gene, wherein the level or amount of methylation is determined simultaneously or sequentially in the POU4F1 gene and the HOXD8 gene in the subject's biological sample.

15. 15. The method of claim 13 or 14, wherein the reference value is obtained from a healthy subject.

16. 1. A method for determining a level or amount of methylation, comprising the steps of: a) determining the level or amount of methylation of the POU4F1 gene in a biological sample of the subject and comparing said level or amount with a reference value; Here, the biological sample is derived from the subject suffering from pancreatic cancer, the subject being evaluated with a view to predicting clinical outcome.

17. 17. The method of claim 16, further comprising determining the level or amount of methylation of the HOXD8 gene in a biological sample of the subject and comparing said level or amount with a reference value.

18. A kit for determining the level or amount of methylation in pancreatic cancer, comprising a nucleic acid useful for determining the level or amount of methylation in the POU4F1 gene.

19. 20. The kit of claim 18, wherein the kit comprises a primer specifically targeting the nucleotide region of SEQ ID NO: 1 in the POU4F1 gene.

20. 20. The kit of claim 19, wherein the kit comprises a primer of SEQ ID NO:5 or SEQ ID NO:

6.

21. The kit according to any one of claims 18 to 20, further comprising a nucleic acid useful for determining the level or amount of methylation in the HOXD8 gene.

22. 22. The kit of claim 21, further comprising a primer specifically targeted to the nucleotide region of SEQ ID NO: 3 in the HOXD8 gene.

23. 23. The kit of claim 22, wherein the kit comprises a primer of SEQ ID NO:7 or SEQ ID NO:8.

Images