DNA methylation biomarkers for early detection of cervical cancer

Abstract

To provide an in vitro method to obtain biomarkers for early detection and diagnosis of human cervical cancer.SOLUTION: Cervical cancer is predicted by examining progression of DNA methylation alterations in three stages of premalignant lesions (cervical intraepithelial neoplasia (CIN)), progressing from CIN1 to CIN3. Provided are combinations of DNA methylation for detecting with high specificity and sensitivity cervical cancer by measuring their DNA methylation status and deriving a "methylation score", which is useful as a biomarker for cervical cancer; as well as kits for predicting cervical cancer using such DNA methylation using multiplexed next generation sequencing methylation assays, pyrosequencing assays and methylation specific PCR.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] [CROSS-REFERENCE TO RELATED APPLICATIONS] This application claims priority under Section 119(e) of U.S. Provisional Application Serial No. 62 / 774,994, entitled "DNA methylation markers for early detection of cervical cancer," filed December 4, 2018, the contents of each of which are incorporated herein by reference.

[0002] [Sequence table] This application contains a Sequence Listing that has been submitted electronically in ASCII format and is incorporated herein by reference in its entirety. The above ASCII copy, created on January 12, 2020, is named TPC53811 Seq List_ST25.txt and is 36,864 bytes in size.

[0003] [Technical field] The present invention relates generally to DNA methylation signatures in human DNA, particularly in the field of molecular diagnostics. More specifically, the present invention relates to DNA methylation biomarkers in the form of panels, one and a combination of polygenic DNA methylation biomarkers for the early detection and screening of cervical cancer, and their use as diagnostic kits for the early and accurate detection of cervical cancer. [Background technology]

[0004] Cancer has become a leading cause of death in humans. Early detection of cancer can significantly improve cure rates and reduce tremendous personal and economic costs to patients, their families, and healthcare systems. At the same time, screening healthy individuals to assess the expression and changes of biomarkers in precancerous stages is useful in population-wide screening methodologies, helping to identify healthy individuals at risk and prone to cancer. Cervical cancer is no exception. Screening can identify cancer early, before it causes symptoms. If cervical cancer is detected at the earliest stage, the chances of survival are approximately 93%, dropping to 15% at the later stages. https: / / www.cancer.org / cancer / cervical-cancer / detection-diagnosis-staging / survival.html Current screening methods include Pap smears, liquid-based cytology, HPV testing, and visual inspection, but there is a lack of a stable, accurate, and sensitive method for the early detection of cervical cancer.

[0005] Biomarkers are one of the most important areas in cancer diagnostics. Cancer biomarkers are particularly useful for early detection or diagnosis of the disease. Biomarkers can be used to screen patients, classify different stages or grades of cancer, and predict prognosis and resistance to treatment.

[0006] The established discovery of human papillomavirus (HPV) as the causative agent of cervical dysplasia has revolutionized the prevention and management modalities of this gynecological disease from the perspective of secondary (molecular HPV testing) (1). Knowledge of HPV genotypes is practically useful for clinical prediction, as HPV types 16 and / or 18 pose a higher risk of lesion progression than other oncogenic types. However, persistent infection with oncogenic HPV genotypes is a necessary precursor and driver for cervical cancer development. The latter can lead to a precancerous stage (cervical intraepithelial neoplasia). It represents the stepwise progression from cervical neoplasia (CIN) to invasive cervical cancer. While low-grade CIN (CIN1) is highly reversible, high-grade CIN grades 2 and 3 (i.e., CIN2 and CIN3, respectively) carry a non-negligible risk of progressing to invasion, i.e., cervical cancer. This is especially true for CIN3.

[0007] Managing women with CIN in the clinic continues to pose a significant dilemma for gynecologists. Aggressive excision or ablative treatments can cause immediate complications when female patients decide to conceive, potentially increasing the risk of subsequent miscarriage or premature birth. Recent evidence suggests that epigenetic alterations in specific genes may mediate or predict carcinogenic progression. Early cancer detection biomarkers can categorically distinguish rare cells with lesions at asymptomatic and precancerous stages through prominent changes, including biochemical alterations at the epigenetic level. These epigenetic alterations as biomarkers are often abnormally abundant in cancerous tissues, often interfering with the manifestation of the disease itself. The development of molecular biomarkers is crucial to identify molecular alterations well before disease onset and progression. One such epigenetic biomarker, DNA methylation levels at specific CpG sites in viral and host genes, has been shown to increase with the severity of the underlying cervical lesions (2-7).

[0008] The most studied and targeted host genes with epigenetic alterations associated with cervical cancer and its precursors are cell adhesion molecule 1 (CADM1); death-associated protein kinase 1 (DAPK1); myelin and lymphocyte, T-cell differentiation protein (MAL); paired box 1 (PAX1); telomerase reverse transcriptase (TERT); erythrocyte membrane protein band 4.1-like 3 (EPB41L3), Ras association domain family member 1 (RASSF1); SRY-box 1 (SOX1); cadherin 1 (CDH1); LIM homeobox transcription factor 1 alpha (LMX); cyclin A1 (CCNA1); family of sequence similarity 19 member A4, CC motif chemokine-like (FAM19A4); and retinoic acid receptor beta (RARβ)8. In addition to panels containing two (i.e., CADM1 and MAL (3,4,10), MAL and miR124-2 (11-14), three (i.e., CADM1, MAL, and miR124-2) (13,15), four (i.e., JAM3, EPB41L3, TERT, and C130RF18) (16,17), and five (i.e., PAX1, DAPK1, RARβ, WIFI, and SLIT2) (14) markers, as well as various combinations of SOX1, PAX1, LMX1A, and NKX6-1 markers, single methylation markers (9) were investigated and achieved sufficiently high sensitivity for advanced disease (18).

[0009] However, only one previous study used a genome-wide methylation approach to identify three methylation panels (JAM3 / ANKRD18CP, C13ORF18 / JAM3 / ANKRD18CP, and JAM3 / GFRA1 / ANKRD18CP) with the highest combined diagnostic accuracy for the detection of CIN2+ in cervical samples. Sensitivity was reported to be 72%, 74%, and 73%, respectively, with corresponding specificities of 79%, 76%, and 77% (2). Therefore, improved methods for identifying DNA methylation biomarkers, panels of DNA methylation biomarkers relevant to early detection and risk prediction of cervical cancer, and kits based on such biomarkers for population-wide screening of apparently healthy women for early detection and susceptibility to cervical cancer and risk assessment in women with precancerous conditions are needed.

[0010] Currently, there is a lack of single or combined methylation markers with adequate diagnostic performance for early stage cervical cancer risk prediction. Therefore, the present invention aims to predict the risk of cervical cancer by using DNA methylation biomarkers as single, combined and panel-based biomarkers. This paper provides a solution to the problem associated with the lack of early detection markers for cancer. The present invention discloses a method for obtaining early biomarkers of the progression from precancerous lesions to cervical cancer, which can be used for general screening in asymptomatic women as well as those with CIN1 to CIN3 disease stages.

[0011] [Objective of the Invention] The main object of the present invention relates to biomarkers for the early detection and diagnosis of human cervical cancer.

[0012] A further object of the present invention relates to an in vitro method disclosed herein called "analysis of progressive DNA methylation alterations (APDMA)," which comprises examining genome-wide profiles of DNA methylation in specimens from women with different CIN grades (CIN1 to CIN3) compared to healthy control specimens from women to obtain CGIDs that, when combined as DNA methylation biomarkers in known methylation profiles of cervical cancer, predict cervical cancer with greater than 95% sensitivity and specificity, using the linear regression model disclosed herein.

[0013] Another object of the present invention relates to molecular biomarkers as indicators for population-wide screening of women for the early detection of cervical cancer and for risk assessment of women with CIN1 to CIN3 disease stages.

[0014] Yet another object of the present invention relates to a chip / array useful for the early detection and diagnosis of cervical cancer.

[0015] Yet another object of the present invention is to provide a cheaper, accurate, stable, sensitive, specific and high-throughput diagnostic kit for accurate early diagnosis of human cervical cancer that can be used by those skilled in the art. Summary of the Invention

[0016] Thus, the present invention provides methods and materials related to DNA methylation CGID biomarkers useful for examining DNA methylation changes and for the early detection and diagnosis of human cervical cancer, where progression of precancerous cervical lesions (cervical intraepithelial neoplasia, CIN grades 1-3) correlates with increasing frequency of DNA methylation at CG positions in the human genome, in the form of Illumina probe IDs or DNA methylation numbers or CG identifiers (CGIDs), obtained using the in vitro method of "analysis of progressive DNA methylation changes" (APDMA) disclosed herein. As described in detail below, these biomarkers are typically based on variables that are useful for predicting a woman's risk of developing CIN1 to CIN3 disease states, as well as for population-wide screening for developing cervical cancer. As a result, these biomarkers are useful as early detection and diagnostic biomarkers. The present disclosure provides that the CGID biomarker positions are nearly uniformly methylated in cervical cancer and nearly uniformly unmethylated in normal cervical specimens. Thus, the present invention discloses a series of "categorically" distinct DNA methylation profiles that create a binary differentiation between cervical cancer and non-malignant tissues in the form of DNA methylation at these CGID sites, whereby these sites are only methylated in cervical cancer and completely unmethylated in non-malignant tissues. Furthermore, as disclosed herein, these biomarker sites show an increasing frequency of DNA methylation with the progression of precancerous cervical lesions from CIN1 to CIN3. Thus, the present invention provides targeted amplification and deep next-generation bisulfite amplification of the CGID biomarkers. The present invention provides an in vitro method for early detection and diagnosis that uses DNA sequencing to detect even a few molecules of cervical cancer cells, or even cells that are in the process of transforming from precancerous lesions to cervical cancer in a background of a mostly normal cervical cell profile. Thus, the present invention is useful for the still inaccessible early detection of cervical cancer cells in a high background of non-malignant tissue, particularly by using cervical specimens such as Pap smears as a simple and user-friendly method of early detection that can be used by those skilled in the art.

[0017] One embodiment of the present invention relates to an in vitro method for obtaining highly predictive sites of cervical cancer for early detection at asymptomatic and even precancerous stages, called "Analysis of Progressive DNA Methylation Alterations (APDMA) method," which uses different sources of genome-wide DNA methylation data obtained in the form of DNA methylation CGID biomarker signatures derived by next-generation sequencing, including MeDIP arrays, MeDIP sequencing, etc. The present invention provides a "categorical" CGID biomarker combination for detecting cervical cancer in a discovery set of genome-wide data from specimens of advanced precancerous lesions ranging from CIN1 to CIN3.

[0018] Previous analyses prior to the present invention using classical "case-control" designs and logistic regression revealed DNA methylation CGID biomarkers that detect cancer with lower sensitivity and specificity. Therefore, another embodiment of the present invention relates to a computer-implemented method for obtaining candidate DNA methylation biomarkers for early detection of cervical cancer diagnosis, called the APDMA method, which reveals the earliest methylation profile of cancer that is primary and essential for the cancer state and present in all cervical cancer specimens tested in this disclosure.

[0019] One embodiment of the present invention discloses an in vitro method for accurately detecting cervical cancer by simultaneously measuring DNA methylation in a polygenic set of CGID biomarkers in hundreds of individuals using sequential amplification with target-specific primers followed by barcoded primers and multiplex sequencing, data extraction, and methylation quantification in a single next-generation Miseq sequencing reaction.

[0020] One embodiment of the present invention discloses an in vitro method for measuring the methylation of said DNA methylation CGID biomarkers using pyrosequencing assay or methylation-specific PCR. The present invention discloses the calculation of a polygenic weighted methylation score that is predictive of cervical cancer.

[0021] One embodiment of the present invention discloses a panel of DNA methylation biomarkers for the screening, diagnosis, early detection and prediction of cervical cancer in samples of DNA isolated from specimens from women, including women with no other clinical evidence of cervical cancer from cervical specimens.

[0022] One embodiment of the present invention discloses a panel of DNA methylation biomarkers in the form of a chip for screening, diagnosis, early detection and prediction of cervical cancer in samples of DNA isolated from specimens from women, including women with no other clinical evidence of cervical cancer from cervical specimens.

[0023] One embodiment of the present invention discloses an in vitro non-invasive method using a panel of DNA methylation biomarkers for the screening, diagnosis, early detection and prediction of cervical cancer in samples of DNA isolated from specimens from women, including women with no other clinical evidence of cervical cancer from cervical specimens.

[0024] One embodiment of the present invention discloses the use of the DNA methylation biomarkers disclosed herein for the screening, diagnosis, early detection and prediction of cervical cancer in samples of DNA isolated from specimens from women, including women with no other clinical evidence of cervical cancer from cervical specimens.

[0025] The present invention provides stable DNA methylation biomarkers identified using CGID locations in the human genome, providing highly accurate, specific, and sensitive risk assessment that can guide early intervention and treatment of cervical cancer even in asymptomatic and precancerous women. The present invention provides a simple yet efficient method that can be used by those skilled in the art to detect cervical cancer. The present invention relates to the use of the DNA methylation CGID biomarkers described herein for population-wide screening of healthy women for cervical cancer and for monitoring and assessing cancer risk in women with HPV infection and CIN precancerous lesions. The present invention demonstrates the utility of the disclosed DNA methylation biomarkers in detecting cervical cancer in CIN samples using a polygenic score based on the DNA methylation measurement method disclosed herein. The present invention also discloses the utility of the disclosed methods for obtaining "polygenic" categorical DNA methylation CGID biomarkers for cervical cancer using any method available to one skilled in the art for performing the APDMA method disclosed herein, following genome-wide bisulfite sequencing, such as next-generation bisulfite sequencing, MeDip sequencing, Ion Torrent sequencing, Illumina 450K array, Epic microarray, etc. The APDMA method is for discovering specific and sensitive markers useful for early and very early detection of cervical cancer due to categorical differences in DNA methylation profiles between healthy control specimens and cervical cancer specimens, which show a gradation of increasing frequency as specimens progress from precancerous stage specimens of CIN1 to CIN3.

[0026] Other objects, features, and advantages of the present invention will become apparent to those skilled in the art from the following detailed description. It should be understood, however, that the detailed description and specific examples, while indicating some embodiments of the present invention, are given by way of illustration and not limitation. Many changes and modifications within the scope of the present invention may be made without departing from the spirit thereof, and the invention includes all such modifications. [Brief explanation of the drawings]

[0027] [Figure 1]Figure 1 shows a roadmap for developing an analysis of progressive DNA methylation changes (APDMA) method to obtain early detection DNA methylation biomarkers. The roadmap shows analytical procedures for developing a DNA methylation profile-based APDMA method using Illumina assay probe identification (CGID) that categorically distinguishes normal profiles of cervical specimens from DNA methylation profiles of cervical cancer specimens to obtain "categorical" DNA methylation CGID biomarkers for early detection, diagnosis, and screening of cervical cancer. In step 1, DNA methylation measurements are obtained from cervical specimens at precancerous stages CIN1 to CIN3 compared with healthy control specimens. The DNA methylation measurements are obtained by performing an Illumina Beadchip 450K or 850K assay on DNA extracted from the specimen, or by performing DNA pyrosequencing on DNA extracted from the sample, or by performing a mass spectrometry-based (Epityper®) or PCR-based methylation assay, and target amplification of regions spanning the target CGIDs disclosed herein from the bisulfite-converted DNA, followed by a second set of barcode amplification and indexed multiplex sequencing on an Illumina next-generation sequencer. In step 2, statistical analysis methods are performed on the DNA methylation measurements from step 1. These statistical analyses include receiver operating characteristic (ROC) assays, hierarchical clustering analysis assays, or neural network CK analysis. In step 3, currently developed and disclosed "progressive DNA methylation alteration analysis" (APDMA) methods are performed to identify CGID locations whose methylation levels are early predictors or biomarkers of cervical cancer. In step 4, the present disclosure further refines and shortlists the combination of polygenic DNA methylation CGIDs into a biomarker set of 16 CGIDs. This method allows us to obtain "categorical" differences, rather than quantitative differences, in the methylation profiles between normal cells and cervical cancer cells.As a result, the switch characteristics of DNA methylation profiles in selected CGIDs provide DNA methylation biomarkers for early detection, diagnosis, and screening of cervical cancer, enabling early detection. These serve as a panel of candidate CGID biomarkers for early detection of cervical cancer in women, especially those who are asymptomatic or have precancerous lesions. [Figure 2] Figure 2 shows how we identified sites whose methylation frequency progressively increases across precancerous CIN stages. DNA prepared from cervical specimens from individuals with CIN1, CIN2, and CIN3 histology; as well as nontransformed healthy control specimens; was subjected to genome-wide DNA methylation analysis using Illumina Epic arrays. Methylation levels in 7715 CGIDs were significantly correlated with progression from CIN1 to CIN3 (q > 0.05). A. The IGV browser shows genome-wide differences in methylation of these sites from control cervical specimens. The top track indicates the chromosomal location. The second track indicates the genome-wide location of Refseq genes. The next tracks (△CIN1-Ctrl, △CIN2-Ctrl, △CIN3-Ctrl) show the average methylation difference between each CIN stage and the control. Progressive hypermethylation is observed across stages. [Figure 3] Figure 3 shows that sites derived by the APDMA method are categorically different between normal cervical specimens and cervical cancer. A. Using DNA methylation data from 270 patients (GSE68339), this heatmap shows that the 79 top CGIDs, whose methylation frequency increases during the progression of cervical precancerous stages, are useful for detecting cervical cancer. CGIDs exhibit categorically different methylation profiles between cancer and normal cervix. They are completely unmethylated in normal tissue and highly methylated in cancer tissue. B. Average methylation for each group of normal, precancerous, and cervical cancer (CIN1 to CIN3) (blue indicates 0% methylation, dark red indicates 100% methylation). [Figure 4]Figure 4 shows the specificity of the bigenic DNA methylation score discovered using the APDMA method for detecting cervical cancer DNA in an independent cohort. A. Calculation of effect size, penalized regression, and multivariate linear regression with a short list of two CGID subsets, and the calculated linear regression equation for predicting cervical cancer. B. The threshold for cancer detection calculated by ROC. C. Using this threshold, the sensitivity and specificity of this marker combination set are 1, and the AUC is 1. [Figure 5] Figure 5 shows the cancer methylation scores in individual specimens from control, CIN1 to CIN3, and cervical cancer patients. A. Methylation scores (cervical cancer prediction) calculated using the formula shown in Figure 4A for each individual specimen from control, CIN1 to CIN3, and cervical cancer showing increased methylation scores in advanced precancerous lesions. B. Scatter plot showing the mean methylation scores for the control, precancerous, and cancer groups. [Figure 6] Figure 6 shows the correlation between bigenic methylation score and progression from control to precancerous stages to cervical cancer. Cervical cancer samples are from GSE68339. CIN1 to CIN3 are from the McGill cohort described in this application (assigned Spearman rank: control: 0, CIN1 to CIN3: 1-3, cervical cancer: 4). [Figure 7]Figure 7 shows the validation of cervical cancer methylation markers (n=312) using DNA methylation data from TCGA. Because cervical data were available for only one CGID (cgl3944175) in TCGA, a linear regression equation was used to calculate the cervical cancer methylation score using DNA methylation data from CGID cgl3944175 only. Pearson correlations were calculated between cancer stage and methylation score (see statistics in A and correlation chart in B). CIN1 to CIN3 were from the McGill cohort described in the application. The assigned scales were: control: 0, CIN1 to CIN3: 1-3, cervical cancer: 4. [Figure 8] Figure 8 shows the utility of the present invention: prediction of cervical cancer in CIN1 to CIN3 specimens. While not all CIN1-3 patients develop cervical cancer, a higher proportion of CIN3 patients develop cervical cancer than CIN1 patients. The present invention tested whether the methylation score developed in Figure 3 could be used to identify individual patients with cervical cancer methylation scores as a demonstration of the utility of the present invention. The AX axis lists individual patients, with groups indicated by lines below the X axis. The Y axis indicates the predicted cancer (1) and predicted cancer-free (0). B. Number of individuals with predicted cancer in each group. The predicted cancer increases from CIN1 to CIN3, as expected. DETAILED DESCRIPTION OF THE INVENTION

[0028] Invention Details In describing the embodiments, reference may be made to the accompanying drawings, which form a part hereof, and which show, by way of illustration, specific embodiments in which the invention may be practiced. It is understood that other embodiments may be utilized and structural changes may be made without departing from the scope of the present invention. Many of the techniques and procedures described or referenced herein are well understood and commonly used by those skilled in the art. Unless otherwise defined, all technical terms, notations, and other scientific terms or terms used herein are intended to have the meaning commonly understood by those skilled in the art to which the invention pertains. In some instances, terms with commonly understood meanings are defined herein for the sake of clarity and / or ready reference, and the inclusion of such definitions herein should not necessarily be construed as representing a substantial departure from what is commonly understood in the art.

[0029] All figures in the drawings are intended to illustrate selected versions of the invention and are not intended to limit the scope of the invention.

[0030] All publications mentioned herein are incorporated by reference to disclose and describe the aspects, methods, and / or materials in connection with which the publications are cited.

[0031] DNA methylation refers to the chemical modification of DNA molecules. Technology platforms such as the Illumina Infinium microarray and DNA sequencing-based methods have been shown to provide highly robust and reproducible measurements of DNA methylation levels in humans. There are over 28 million CpG loci in the human genome. Consequently, specific loci are given unique identifiers, such as those found in the Illumina CpG Locus Database (see, for example, Technical Note: Epigenetics, Identification of CpG Loci, ILLUMINA Inc. 2010). These CpG locus-specific identifiers are used herein.

[0032] Definition: As used herein, the terms "CG" or "CpG," used interchangeably, refer to dinucleotide sequences in DNA that contain cytosine and guanosine bases. These dinucleotide sequences can be methylated in the DNA of humans and other animals. The CGID identifies its location in the human genome as defined by the Illumina 450K manifest or the Illumina EPIC manifest (annotations of the CGs listed here are publicly available at: https: / / bioconductor.org / packages / release / data / annotation / html / IlluminaHumanMethylation450k.db.html or https: / / bioconductor.org / packages / release / data / annotation / html / IlluminaHumanMethylationEPICmanifest.html ), installed as the R package IlluminaHumanMethylation450k.db (R package version 2.0.9.) or IlluminaHumanMethylationEPICmanifest (R package version 0.3.0.).

[0033] As used herein, the term "beta value" refers to the calculation of the methylation level at a CGID position derived by normalization and quantification on an Illumina 450K or Epic array using the intensity ratio of the methylated probe to the unmethylated probe and the formula: beta value = methylated C intensity / (methylated C intensity + unmethylated C intensity). Beta values ​​are between 0 and 1, with 0 being fully unmethylated and 1 being fully methylated.

[0034] As used herein, the term "penalized regression" refers to a statistical method that aims to identify the minimum number of predictors required to predict an outcome from a larger list of biomarkers taken, as described, for example, in the R statistical package, where "penalized" is Goeman JJ, L1 penalized estimation in Cox proportional hazards models, Biometrical Journal 52(1), 70-84.

[0035] As used herein, the term "clustering" refers to grouping a set of objects such that objects within the same group (called a cluster) are more similar (in some sense) to each other than to other groups (clusters).

[0036] As used herein, the term "hierarchical clustering" refers to the methodology described in, for example, Kaufman, L.; Rousseeuw, PJ (1990) Finding Groups in Data: An Introduction to Cluster Analysis (1st ed.) New York: John Wiley. ISBN 0-471-87876-6 As explained in , it refers to a statistical method of constructing a hierarchy of "clusters" based on how similar (close) or dissimilar (distant) the clusters are to each other.

[0037] As used herein, the term "receiver operating characteristic (ROC) assay" refers to a statistical method for generating a graphical plot illustrating the performance of a predictor. For example, as described in Hanley, James A., McNeil, Barbara J. (1982) "The Meaning and Use of the Area Under the Receiver Operating Characteristic (ROC) Curve," Radiology 143(1):29-36, the true positive rate of a prediction is plotted against the false positive rate at various threshold settings (i.e., different percentages of methylation) of the predictor.

[0038] As used herein, the term "multivariate or polygenic linear regression" refers to a statistical method for estimating the relationship between multiple "independent variables" or "predictors," such as the methylation percentages of multiple CGIDs, and a "dependent variable," such as cancer. This method assigns a "weight" or coefficient to each CGID in predicting an "outcome" (dependent variable, such as cancer) when several "independent variables," such as CGIDs, are included in the model. Make a decision.

[0039] As used herein, the term "epigenetic" means chemical modification of DNA molecules, relating to, or involving chemical modification of DNA molecules. Epigenetic factors include the addition or deletion of methyl groups, which results in changes in DNA methylation levels. New molecular biomarkers for the early detection, diagnosis, or prediction of cervical cancer that monitor genomic DNA methylation patterns, such as those disclosed herein as CGID-based biomarkers, can predict cervical cancer risk and susceptibility even at very early stages, when women are asymptomatic or in the precancerous stages of CIN1 to CIN3, and are useful in clinics, epidemiologists, and medical professionals. These new molecular biomarkers are the subject of this disclosure, making them accessible and usable by anyone skilled in the art. While clinical biomarkers, such as Pap smears and histological identification, have a long and successful history in diagnosing cervical cancer, they are highly variable and cannot be used for early detection of cervical cancer. In contrast, molecular biomarkers, such as epigenetic markers in the form of DNA methylation biomarkers, are still largely unused.

[0040] As used herein, the term "DNA methylation biomarker" refers to a potentially methylated CpG position. Methylation usually occurs in CpG-containing nucleic acids. CpG-containing nucleic acids can be, for example, present in the CpG island, CpG doublet, promoter, intron, or exon of a gene. For example, in the gene region provided herein, the potential methylation site encompasses the promoter / enhancer region of the indicated gene. Thus, the region can begin upstream of the gene promoter and extend downstream of the transcribed region.

[0041] The currently disclosed method demonstrates that the frequency of cells exhibiting the DNA methylation profile of cervical cancer increases with progression from CIN1 to CIN3 disease, and that these methylation profiles are characteristic of the earliest stages of cervical cancer. Second, because cells that transform into cancer are rare in early malignant tumors, the DNA methylation profile must be categorically distinct from the normal profile of cervical cells, so that it is detected in a background of mostly non-malignant cells at the earliest stages. Third, if these DNA methylation profiles are a primary and important feature of cervical cancer, they must be present in all fully developed cervical cancer specimens. Taking the aforementioned three prerequisites into consideration, the presently disclosed in vitro method, termed "Analysis of Progressive DNA Methylation Alterations (APDMA)," involves using an Infinium Methylation EPIC array to examine genome-wide DNA methylation profiles of isolated specimens obtained from women with different CIN grades (CIN1 to CIN3) compared with healthy, non-transformed, healthy control cervical specimens after well-characterized HPV genotyping. The present invention also discloses an in vitro method for obtaining Illumina probe IDs, or DNA methylation counts, or CG identifiers (CGIDs), which, when combined as DNA methylation biomarkers with known cervical cancer methylation profiles, predict cervical cancer with >95% sensitivity and specificity using the linear regression model disclosed herein. The present invention also provides a panel of DNA methylation biomarkers for screening and early detection of cervical cancer, each panel comprising a CGID having a sequence selected from the group consisting of SEQ ID NO: 1 to SEQ ID NO: 79 listed in Table 1, and combinations thereof, such as a shortlisted subset of the Table 1 sequences listed in Table 2 and a shorter subset of the CGIDs listed in Table 3, as disclosed below. Thus, the present invention provides two CGIDs that are minimally sufficient to detect cervical cancer in published DNA methylation data with a sensitivity and specificity approaching 1. The present invention also discloses a kit for in vitro measurement of DNA methylation biomarkers, such as DNA methylation levels of the disclosed CGIDs in DNA isolated from cervical specimens, to be used in population-wide screening of women for early detection of cervical cancer and risk assessment of women with CIN1 to CIN3 disease stages.

[0042] The invention disclosed herein has many embodiments. In one embodiment, the present invention provides polygenic DNA methylation CGID biomarkers of cervical cancer in cervical smears for the early detection of cervical cancer, wherein the polygenic DNA methylation biomarker panel is derived using the "Analysis of Progressive DNA Methylation Changes (APDMA) method" disclosed in the present invention for genome-wide DNA methylation derived by mapping methods such as Illumina 450K or 850K arrays, genome-wide bisulfite sequencing using various next-generation sequencing platforms, methylated DNA immunoprecipitation (MeDIP) sequencing, hybridization with oligonucleotide arrays, etc.

[0043] In one embodiment, the present invention provides a method for obtaining DNA methylation biomarkers for detecting cervical cancer, comprising the steps of performing statistical analysis and the "Progressive DNA Methylation Change Analysis (APDMA)" method disclosed in the present invention on DNA methylation measurements obtained from cervical specimens with precancerous lesions CIN1 to CIN3.

[0044] In one embodiment, the disclosed method comprises performing a statistical analysis and "Progressive DNA Methylation Change Analysis (APDMA)" method on DNA methylation measurements obtained from cervical specimens, wherein the DNA methylation measurements are obtained by running an Illumina BeadChip 450K or 850K assay on DNA extracted from the specimen. In another embodiment, the DNA methylation measurements are obtained by performing DNA pyrosequencing on DNA extracted from the sample, or by mass spectrometry-based (Epityper®) or PCR-based methylation assays and target amplification of regions spanning target CGIDs disclosed herein from bisulfite-converted DNA, followed by a second set of barcode amplification and indexed multiplex sequencing on an Illumina next-generation sequencer. In a further embodiment, the statistical analysis comprises a receiver operating characteristic (ROC) assay. In yet another embodiment, the statistical analysis comprises a hierarchical clustering analysis assay. In an additional embodiment, the statistical analysis comprises neural network analysis.

[0045] In one embodiment of the present invention, an in vitro method for obtaining an early predictor of cervical cancer is disclosed, which includes the following steps: (a) measuring DNA methylation from a cervical specimen sample; (b) performing statistical analysis of the DNA methylation measurements obtained in step a; (c) determining the DNA methylation status of a number of independent genomic CG locations, referred to as CG identifiers (CGIDs), by performing an analysis of progressive DNA methylation changes (APDMA) of the genome-wide DNA methylation profile obtained in step b; (d) classifying the CGIDs based on the frequency of their DNA methylation associated with the progression of precancerous stages of cervical cancer; and (e) obtaining candidate CGIDs from the classification in step d to obtain an early predictor of cervical cancer as a DNA methylation biomarker.

[0046] In another embodiment of the present invention, an in vitro method for obtaining an early predictor of cervical cancer is disclosed, which method comprises the following steps: (a) measuring DNA methylation from cervical specimen samples; (b) performing a statistical analysis of the DNA methylation measurements obtained in step a; (c) determining the DNA methylation status of a number of independent genomic CG locations, called CG identifiers (CGIDs), by performing an analysis of progressive DNA methylation changes (APDMA) of the genome-wide DNA methylation profile obtained in step b; and (d) identifying a pre- or post-cervical cancer predictor. (e) obtaining candidate CGIDs from the classification of step d to obtain early predictors of cervical cancer as DNA methylation biomarkers, wherein the DNA methylation measurements are performed using methods including genome-wide bisulfite sequencing on platforms including Illumina 27K, 450K, or 850K arrays, HiSeq, MiniSeq, MiSeq, or NextSeq sequencers, Torrent sequencing, methylated DNA immunoprecipitation (MeDIP) sequencing, hybridization with oligonucleotide arrays, DNA pyrosequencing, mass spectrometry-based (Epityper®), or PCR-based methylation assays.

[0047] In yet another embodiment of the present invention, an in vitro method for obtaining an early predictor of cervical cancer is disclosed, comprising the steps of: (a) measuring DNA methylation from a cervical specimen sample; (b) performing statistical analysis of the DNA methylation measurements obtained in step a; (c) determining the DNA methylation status of multiple independent genomic CG locations, referred to as CG IDs, by performing an analysis of progressive DNA methylation changes (APDMA) of the genome-wide DNA methylation profile obtained in step b; (d) classifying the CGIDs based on the frequency of their DNA methylation associated with progression to precancerous stages of cervical cancer; and (e) obtaining candidate CGIDs from the classification in step d to obtain an early predictor of cervical cancer as a DNA methylation biomarker, wherein the statistical analysis of the DNA methylation measurements comprises Pearson correlation, receiver operating characteristic (ROC) assay, and hierarchical clustering analysis.

[0048] In a further embodiment of the present invention, an in vitro method for obtaining an early predictor of cervical cancer is disclosed, comprising the steps of: (a) measuring DNA methylation from a cervical specimen sample; (b) performing statistical analysis of the DNA methylation measurements obtained in step a; (c) determining the DNA methylation status of multiple independent genomic CG locations, referred to as CGIDs, by performing an analysis of progressive DNA methylation changes (APDMA) of the genome-wide DNA methylation profile obtained in step b; (d) classifying the CGIDs based on the frequency of their DNA methylation in relation to progression of precancerous stages of cervical cancer; and (e) obtaining candidate CGIDs from the classification in step d to obtain an early predictor of cervical cancer as a DNA methylation biomarker, wherein the progression of precancerous stages of cervical cancer includes cervical intraepithelial neoplasia lesions of stages CIN1, CIN2, and CIN3.

[0049] In another embodiment of the present invention, an in vitro method for obtaining an early predictor of cervical cancer is disclosed, comprising the steps of: (a) measuring DNA methylation from a cervical specimen sample; (b) performing statistical analysis of the DNA methylation measurements obtained in step a; (c) determining the DNA methylation status of a number of independent genomic CG sites, referred to as CGIDs, by performing a progressive DNA methylation change (APDMA) analysis of the genome-wide DNA methylation profile obtained in step b; (d) classifying the CGIDs based on their DNA methylation frequencies associated with the progression of precancerous stages of cervical cancer; and (e) obtaining candidate CGIDs from the classification in step d to obtain an early predictor of cervical cancer as a DNA methylation biomarker, wherein the CGIDs based on their DNA methylation frequencies associated with the progression of precancerous stages of cervical cancer are selected from the group consisting of CGIDs set forth in SEQ ID NO: 1 to SEQ ID NO: 79 and combinations thereof. In a supplementary embodiment of the present invention, the 79 CGID sites are delineated as DNA methylation biomarkers for the early detection of cervical cancer and are useful, alone or in combination, as early predictors of cervical cancer.

[0050] Table 1: Selected DNA fragments with CG methylation sites (CGIDs) useful in embodiments of the present invention 79 polynucleotides identified. The sequences of the Illumina probe IDs for the 79 selected CGIDs used in various embodiments herein can be found in Table 1 included in this application and include the following: cg08272731 described in SEQ ID NO: 1 (GAAGGAGGCTGCGCGCCAGCCCGCCCGCGGCGCCCGGGCTCAGGCGCCGTGACGGCTGCACGCGCTGCCCCGCACTCTGAGGCCTTCATTAGCTCGCTCCCCGCGCCGAGGCTGGGGCGGG), cg19598567 described in SEQ ID NO: 2 (CCTCCCGCAGCTCATTGCAGCCCCGAGGAAATCACCGGGGGAGGGCTCGGGAGTGCGGCGCGGCAGCCCCATAATTTCCAGGGCCCTTCTCCTACACTGACACGTAATTGTCAGATTGTTTT), cg13944175 described in SEQ ID NO: 3 (CCGCCGCGGGTTCCCAGGGCTGGTGGTAGTTGCCGTCCCACACGTACGTGGCGGGGTCCTCGTCAGCGAAGACCTCGCGGAACATGTCGACCATGTAGAGGTCCTCGGCGCGGTTGCCATCC), cg19717586 described in SEQ ID NO: 4 (GGGGAGGAATATTAGACTCGGAGGAGTCTGCGCGCTTTTCTCCTCCCCGCGCCTCCCGGTCGCCGCGGGTTCACCGCTCAGTCCCCGCTCGCTCCGCACCCCACCCACTTCCTGTGCTCG), cg22721334 described in SEQ ID NO: 5 (CAGGCCGGTCCCAGCCGCCCGGAGCCCCAGTGCGCGATGGCGGCCGGCAAACTGCGCCTGCGCACTGGGCCTCACCGCGGACTACGACTCCCACAATGCCGCGAGGCTGTGCCGCGCACCGG), cg13985485 described in SEQ ID NO: 6 (GTGACGCGCGGCCGCAGCTGCCCGCGGGCGGAGCGCTCTCAGACCCCGGAGCGCACACCGCGGGGCCATCGGTGCCATCGCGGATCTCCAGGCTCCTCATCAGTCCGCCGGGGCCGCAGCAG), cg11358689 described in SEQ ID NO: 7 (GAGGAATATTAGACTCGGAGGAGTCTGCGCGCTTTTCTCCTCCCCGCGCCTCCCGGTCGCCGCGGGTTCACCGCTCAGTCCCCGCGCTCGCTCCGCACCCCACCCACTTCCTGTGCTCGCCC) cg01944624 described in SEQ ID NO: 8 (ATCTACCGTCTCCAATCTCCATCTCCGAAGTTATGCCCACTTCCTCGAAGTTTGGAGCCACGCGAACTACACTGCCCAGAAGGCGCCGCGCCGTGAGCCGCGAT GCTTGGCCAATGAAAAGA), cg04864807 described in SEQ ID NO: 9 (GGGAGGGCTCGTGAGAGCCAATGAGAGCGCGGAAGGCGGCGAGCGAGCCAATGGACGCGGCGGTGGGGCAGGGGGCGGGGCCTGGGCGAGGCCGGGGGCGGAATGGGCTGAGTGCCCTGTCT), cg13849378 described in SEQ ID NO: 10 (CGGCAAGCGGAGCAGCGAGGCAGGGTAGCTTCATCACACTCGCGGCGGATGCGGATTCCGCGCCGCCCCGGCTCTAGCTGCTCAGGCGACCGCCACCCTCGCCTCGCCGCCGCCCGTGCACA) cg19274890 described in SEQ ID NO: 11 (GCGGACGGCGGCTCCATCCGCGGCAATCACCGTAGTGCTTGTTTGTGGAAGCCGAGCGTGCGTGCGCGCGCGCGCACCCAGTCCAGCGCGGAGTGGGCGTCTACCCGAGGAGGGGTGTCTG), cg06783737 described in SEQ ID NO: 12 (TGGGGAATTAGCTCAGGCGGTGGAGCGCTCGCTTAGCTATGCGAGAGGTAGCGAGATCGACGCCCGCATTCTCCAGTTTCTTGTCTGGTTTATGTCTCTTAGTTTGTATTCCCCGTTGTTTC), cg19429281 described in SEQ ID NO: 13 (GAAGTCCCAGGGACCTGCGGAGCGCAGACATAACACAACACAGAGCAAAACTCACCGCTGCGGTGACTTTCACTCCACGCGATCCGCTTCCCGGTTTACGCTAAACTGGGCGCTCGGGACAG), cg00064733 described in SEQ ID NO: 14 (GGCTGCGGACGGCGGCTCCATCCGCGGCAATCACCGTAGTGCTTGTTTGTGGAAGCCGAGCGTGCGTGCGCCGCGCGCGCACCCAGTCCAGCGCGGAGTGGGCGTCTACCCGAGGAGGGGTG), cg25258740 described in SEQ ID NO: 15 (CCCCCGCCGGCCGCCGGCCGCGCTCCCCGCCTTCATTCTGTGATCTGCGGATTTGCCAGTCGCCAACCTCCGCGCCAGAGTCACCATCGCGCAGGGTTGGGCAAACCATGGAGCTCGGGGC), cg08087594 described in SEQ ID NO: 16 (AACTCCTGCACAAATCATTTCAAACGCGGTCGGCTTCTAATCGGGAAGTAATCTCAGTGACGCTGGCGGTGCAGAGAACCGAGTCTGGACGCACACACACAAACACACCGCGGCCTCCGCA), cg17233763 described in SEQ ID NO: 17 (GTGTGCTCAGCCTCAGCGTGAGGGGCACCTGCTCGTCTGGGCTCACAGCGAAGGCAGCCTCGCCGCGAGCTGCCGCTGCCGCTGCTGCCGCCACTGGTGTTGCCGCTCTCAGGCGCCAGGCT), cg11372636 described in SEQ ID NO: 18 (GCCGGGAGCCTGACGTCACCACGCCCTGCCTGTCAATCTGCAGCGCGCCGCCGCTCGCAGCCGCCTTTTCTGCCACCAACTGTATCTCTCACTCGCGGAGCCGGCACAGCGACAGGCGCCCCG), cg01650149 described in SEQ ID NO: 19 (GCGGCGGCGGGCGGGGAGCCAGGCCCGAGCTGCGTTCTGCGCAGCCATTGGTGGGCGCCGCGCTCTGCACTGAGCATGTTCGCGCCCCGCCGGCCCCTAGCCGCAGCCGCAGCCGCAGCGAC), cg17445666 described in SEQ ID NO: 20 (CAACCGGTTCCGCCGCGTTTGTGGGCTGGTAGCCCGGAATACATTTCCCAGAGGCCTTCGCGGCCGACGTGCTTCGCGCAGGAACGCAGCCGCCTCCCGACTGGAGGACGCGGTAGCGGAGC) cg24415208 described in SEQ ID NO: 21 (GCTGCCCGTGGTCAAACTGGAGTCGCTGAAGCGCTGGAACGAAGAGCGGGGCCTCTGGTGCGAGAAGGGGGTGCAGGTGCTGCTGACGACGGTGGGCGCCTTCGCCGCCTTCGGCCTCATGA), cg24221648, as set forth in SEQ ID NO: 22 (CTTCCCGGCTCCCCGCGGTGCGCACCCGCTGGCCACTCTGCGCACGCGCGCCGGGTGCCCCGGCCTAAGGCCGTTGACCTCGGGTTCTCCCCGGCACAGTCGAATCCACGCCAGGGCCCTCA), cg09017434 described in SEQ ID NO: 23 (GCGGGGGAGGTTGCGGGGGAGGCTCGGCGTCCCCGCTCTCCGCCCCGCGACACCGACTGCCGCCGTGGCCGCCCTCAAAGCTCATGGTTGTGCCGCCGCCGCCCTCCTGCCGGCCCGGCTGG), cg15814717 described in SEQ ID NO: 24 (TGTACTACTTCCTCTGCCACCTGGCCTGGTAGACGCGGGCTTCACTACTAGCGTGGTGCCGCCGCTGCTGGCCAACCTGCGCGGACCAGCGCTCTGGCTGCCGCGCAGCCACTGCACGGCC), cg23619365 described in SEQ ID NO: 25 (AAAAAAAAAAAAAAGCAATGAGCCGCAAGCCTTGGACTCGCAGAGCTGCCGGTGCCCGTCCGAGAGCCCCACCAGCGCGGCTCACGCCTCAGTCTCGCCGCCCCAAGGTGGGATCCGACGCC), cg20457275 described in SEQ ID NO: 26 (CGAGAGGGCCCGGTCCAGCAGCCTCTGGGGCCCAGTGCGCAGGGCACTGCGGGCCGATTGCGCCCCGGGGCCAGGAGGCGCCGAGAAAGCAAAAGCAAGCCGGCGGCGGGTGGAGGTCAA), cg22305167 described in SEQ ID NO: 27 (CGGCCGCAGTGTGCCGCCCGCTGCGCTATGCGGGGCTCGTCTCCCCGCGCCTATGTCGCACGCTGGCCAGCGCCTCCTGGCTAAGCGGCCTCACCAACTCGGTTGCGCAAACCGCGCTCCTG), cg16664405 described in SEQ ID NO: 28 (CCTGGCGCGACCGCCAGCAGACCCAGCGCGGGGCCGGGAGCTGCTGGGGGCCCAGGCTCCGCTCTCCCCACCGCTCTGCACCGCTGCCGGCTGCGGACAGACCCGATGCGCCACCACCACC), cg16585333 described in SEQ ID NO: 29 (CCGGAGCGCGCTGCTGCCCTCTACCGGTCATCCGTGCGGCCGGACACCGTGTCAGGCCCGCGAGGAGGGCTCTGCCGCAGTCCCGGGGAACAGCACCCAGCAGCGCCACTGGGAGAGGAAAC), cg05057720 described in SEQ ID NO: 30 (AGTCCAGAGCGGCGCTGTGCAGCTGGAAGGGCGCGCGATAGCTCAAGTTAGAGGCGGCCCCGGGGCGCGGCGCAGGACACAAGACCTCAAACTGGTACTTGCACAGGTAGCCGTTGGCGCGC), cg03419058 described in SEQ ID NO: 31 (GGCGGTGCGAGCTCCCCGCCTGCGGGACGCACGGAGACCGCGGTCAGCGCGCCGCCTGGCCGGCCCAGCGCGCCAGCCCGCGCCCAGCCCCGTCCACTCCCGTCCAGCCCCGCCGCCCGGC), cg02473540 set forth in SEQ ID NO: 32 (CGGTAGAGTTTCCAACACGAAAGCCCGTGTGGTCGCGCCGGGAGCTCACGGCGTTCCAAGCGGCACTTATCCCGCGTTGATGCCCAGGCACCCCGCGCGCCTGTTTCACCAGGCCCAGTCA), cg01758512 set forth in SEQ ID NO: 33 (CCAGCGGCAGTAGCTGTAGCAGCTTCAGCGAAGCCGGAGATGGGCAGAGAGCGCGCGGCGCAGCAGCTCCAGATTCACTGCTCTCCCCTGCAGCTCCCCGCGCCCCCGCCGCTGTCGCTG), cg18897632 described in SEQ ID NO: 34 (GTGTTCTCTGCGGCGGGCCGCGTCCCCGCTGAGCCTCGCGGTGACAGCCGCCTTTGGCAGCGAGCGCTCGGGGCACTTCTATCCCCGCCTCTCAAAGGGTGGGGACAGCCGTTTCCAGATTT), cg09568464 set forth in SEQ ID NO: 35 (CGGCCGCCCCCGGCAGCCCAGGGCGCGCTTCCACCACGGTACCGGTGGATTCGCCGTGCGCAGCCGGAAGATGGCGCAGACGCACAAAGCACACCGATGCTGCGCCATGATAGGGCCGGC), cg15811515 described in SEQ ID NO: 36 (TCTCGCGGCGCAGGCGGCGGCGGCAGAGGTGGGGTCGCGCAGCGGAGGCAGCTCGAGCTTCGGGATGCGCGCTCGCTTCTTGGGCTCCTCGCTCGATCTTACTGCCCCCTTTTTTCTCTCCC), cg00884040 described in SEQ ID NO: 37 (TCCTCCAGCCAGAGTCGGTGGGACTGGCTGCGCTGCCCTGAAGTGGTTCTCCAAGCAGCGCGGAGGGTGGCGGACGGCGGACGGAGCCCAGGGGCCGCGTCGGGTGGGGAAACCCGAACTCG), cg21632158 described in SEQ ID NO: 38 (TGCGCATCGCTGGCTCTGGGTTCCGCCGAATGCGTCCTCCTGGGCGGTGATGGCTCTGGACCGCGCGGCCGCAGTGTGCCGCCCGCTGCGCTATGCGGGGCTCGTCTCCCCGCGCCTATGTCG), cg18343957 described in SEQ ID NO: 39 (AGGGGAGCTGCGAGGCGAAGTGTTCTTCAGGGAAGCGGGCTCGAGTCTCCGCAGCTGCGGCGGCGGCGGCGGCGCGCTGGGCCGGCGGCGGGCGCGGGCAGGGGGCCGGGGGTGCCGCGCGG), cg23883696 described in SEQ ID NO: 40 (CCTCCCCCCCGGGGGGTTCCTGCGCACTGAAAGACCGTTCTCCGGCAGGTTTTTGGGATCCGGCGACGGCTGACCGCGCGCCGCCCCACCGCCCGGTTCCACGATGCTGCAATACAGAAAGT), cg24403845 described in SEQ ID NO: 41 (AGAGAGGGGTCCCAGAACGAAGGTGGCGGCACGAGCTCTGCGCTGGCGGCTGTGGGGGGCCGGCCTCAGGACCCCAACTCCATCCAAGTTGCGCCGCGGTGGGGGCGGGCGGAGGCGGCGC), cg20405017 described in SEQ ID NO: 42 (AATCTCCCCTCGGGCTCGACGGATGTGCGCCCCAGATGTGCTGACACATGTCCGATGCCTCGCTGCCTTGGAGGTCTCCCCGCTCGCGTGTCTCTTCTCTTCGCACCAGCGGCGGAAACCGC), cg21678377 described in SEQ ID NO: 43 (GCTCCGCTTCTCCGGGTTTTAGCGGAAGCCTGCGGGGGGCGGGGTAACCGCGGAAGCCGGCGGCCGTGGGCGCGCGGGTTGGGGGCTCTCGCGCCGCTCCGGGCTCCCCCCCCCCGGCTG), cg03753331 set forth in SEQ ID NO: 44 (CGCGCTCCGCTTCTCCGGGTTTTAGCGGAAGCCTGCGGGGGGCGGGGTAACCGCGGAAGCCGGCGGCCGTGGGCGCGCGGGTTGGGGGCTCTCGCGCCGCTCCGGGCTCTCCCCCCCCGG), cg16587616 described in SEQ ID NO: 45 (GCGAGGGATCCTCTGTGCGTCCTCACTGGCCCATGCACCCAGCACCTGCGACTCCCGCCGTCGGGCTGCGTGGCCCCGCGCCCACACCTGCCCGTCCCTTCCGTCGTCCCTCGCTCGCGCAGA), cg25730685 described in SEQ ID NO: 46 (GGGGAGGTGTGGGGAGCGGAAGGCCGCAGGAGCATCTTTGCGGAGAAAGTACTTTGGCTGCGGCGGGCGCAGGGCGGGCCGGCTAGCCCCGCGCCCCACCTGTTCTGTGCGTCGCGCTCGCC), cg20019985 described in SEQ ID NO: 47 (TAGGGCTGGAAACCCGCCGCCACAGCGGGCTAGAGGTCGTCCCCGCCCGCAACATATGCGCGAAGGAAAGTGCTACGAACGTCAAATGGCCGCCCCCCGCCGACGCCATCTGCTCTGCGAAG), cg03730428 set forth in SEQ ID NO: 48 (CGCCCGCAACATATGCGCGAAGGAAAGTGCTACGAACGTCAAATGGCCGCCCCCCGCCGACGCCATCTGCTCTGCGAAGCAGAAACGGCGGCAGCTGCGCGCCCAGTCCCTCCGCCCGCGCC), cg18384778 described in SEQ ID NO: 49 (CCCCCTGTTCAAGGTCTGTCACCGTAGGGGGCGGGGGGGCGCGTGGAGCCGCTGGGGGTTCGGCCCACCCCGCGAACCG AGCTCCCGGCCCTGTGCGCCCTCAGCTCTGCCGCGGGCGTTGG), cg22010052 set forth in SEQ ID NO: 50 (GCTGTGGCCGCAGCTGAGGCCCGACGAGCTTCCGGCCGGGTCTTTGCCCTTCACTGGCCGCGTGAACATCACGGTGCGCTGCACGGTGGCCACCTCTCGACTGCTGCTGCATAGCCTCTTCC), cg19688250 described in SEQ ID NO: 51 (GTGTGCGTGTGCGTGTGCTCAGCCTCAGCGTGAGGGGCACCTGCTCGTCTGGGCTCACAGCGAAGGCAGCCTCGCCGCGAGCTGCCGCTGCCGCTGCTGCCGCCACTGGTGTTGCCGCTCTC), cg04701034 set forth in SEQ ID NO: 52 (TGGGGCAGCGGCGTTTGCAGGAGATGAGCTCAGCGCAAAGGGAACCCCGCAGCGGCGAGTGCGGCTGCTGGCCTGCGCGCTGTGGCCCACAGGCTGGCAGGGCGCGGGCGGGTGGCGGGGT), cg20505704 set forth in SEQ ID NO: 53 (AGAGTCGGTGGGACTGGCTGCGCTGCCCTGAAGTGGTTCTCCAAGCAGCGCGGAGGGTGGCGGACGGCGGACGGAGCCCAGGGGCCGCGTCGGGTGGGGAAACCCGAACTCGCGGAGGGGAA), cg15124215 described in SEQ ID NO: 54 (AAAGCCCTGGCAGGTAAAGAGAGGACCCGCGCAGGCTGGGAGCTCCCACTCCTCCTCCAGCGTCACGCTCGCCCTCCGCCGCTGCCTCGCGTCCGGGTCTGTTTATATAGCGTCTGGAGGCC), cg07143083 set forth in SEQ ID NO: 55 (CTGGCCAAGTGCCGGCCCATCGCGGTGCGCAGCGGAGACGCCTTCCACGAGATCCGGCCGCGCGCCGAGGTGGCCAACCTCAGCGCGCACAGCGCCAGCCCCATCCAGGATGCGGTCCTGAA), cg00688962 set forth in SEQ ID NO: 56 (GGCGCCGGCAGCTTCGCGCCGGCGGCTGGAAGCGGGCGGGCTGCACGGGCGGCTCGAGTGCGGGGACCCCAGCCCCTCGCCCTCGTGAGCGCCGCCCTGCCACCTGCTGCCAAGTCACCGG), cg00027083 set forth in SEQ ID NO: 57 (CCCCGGCCGCGCCGGGGCGGGGCTCGGGATTCGGGAGACCGCGCGGCGCCGAAGCCACGCGTCAGCCCCACTGTCCCGCGCCTCGCCCCAGGCCTCGGGCTCTTCCTCCGCACCTCGTA), cg08305436 set forth in SEQ ID NO: 58 (ACGCGGGGACTGGAAAGGGCGCCTGGGTGGGAAGAGGCGCTGGCGGGTGATCGTCCCCACCGGGCCAGTCCCCGGGATCTGCTGCCGCCCCTCTCCGAAATTCACAGCCAGAGCGGGCGCAC), cg1463883 set forth in SEQ ID NO: 59 (TCTGAGAAGTGTCCTCCTCGCTCTCTTATAAAAACAGGACTTGTTGCCGAGGTCAGCGCGCGCATCGAGTGTGCCAGGCGTGTGCGTGGTTTCTGCTGTGTCATTGCTTTCACGGAAGGTGG), cg09907509 set forth in SEQ ID NO: 60 (GCGCCCAGACTGCGCGCCGCGCCGCTGCGCCCAACATTCCCGAGGACGGCTTCGCGGGCGCGTATCGTCCAGACCGGAGCACCGCCCCACCGCTAGCGCAGGAGACCTGCCGGGGAAGTCGC), cg20707222 set forth in SEQ ID NO: 61 (AAAGGCCGTACTCTGCCCCCCGGGACCCAGGTCCCCGCCTGCTGCAGAGCGCACTCTGCGCACGTCGAGCCGCGAAAGGTTCACAGAAGAAAACAAGAGAAAGAAGTAGCAGGCACTGAG), cg17056618 set forth in SEQ ID NO: 62 (GGAATCCATTCTTTTAAGCCAGGGTTTAAAACTCTTCAAGCAAGTCATCTGCAAAGGTACCGCTTCTACCATTTTAAAGATAGGATTATGTTCCCTAGGACAACTGGATGAGCCCTAGGAAC), cg18058689 set forth in SEQ ID NO: 63 (GAGGAGCGCGCCGCTGCCTCTGGCGGGCTTTCGGCTTGAGGGGCAAGGTGAAGAGCGCACCGGCCGTGGGGTTACCGAGCTGGATTTGTATGTTGCACCATGCCTTCTTGGATCGGGGCTG), cg22620221 set forth in SEQ ID NO: 64 (CCCTGTGCGTGCCGCCGCGCTGTTGCTCGCAGTGTGCTGGCGCCGAGCTCGGTGGACACGCGCGCAGTCAGAGCTGCCTCTCGCCCTCGCTAGCTGGGCTCGCAGCCTCTTCCTCCCTCCCT), cg02547394 set forth in SEQ ID NO: 65 (CTCTTTGGCAAGTGGTTTGTGCATCAGGAGAAACTTTCCACCTGCGAGCCGAACCGGCGCCGAGTGCGTGTGTTTCTGCCTTTTTTTGTTGTCGTTGCCTCCACCCCTCCCCATTCTTCTCT), cg09469566 set forth in SEQ ID NO: 66 (TGGCTGCCAGAGCGAGTGAGGGGCGCAGAGGCGGCAGAGAGCGGAGAGCCCCGGTGTCTCCGCGAGGGCGGCGGCGGCCAGCAGACGGCGATCGAGGCGCGCCACGGCACGGCCAGCGCA), cg26609631 described in SEQ ID NO: 67 (AAGCGCGTGGAGAGCCGAAAGGTGCGGTGGGCGCAGAGGGCGGGCTGGCTGCGGGGCGACCGCGCGCCGGGGCCATGCCGCGCTCCTTCCTGGTGGACTCGCTAGTGCTGCGCGAGGCGGGC), cg10132208 described in SEQ ID NO: 68 (GGGGTCGCCATGACCGAGTGGCCCAGGCCCGAGCGAAGCCCGCGCGGTGAGTCCGCCGCGGCCCATCCGTCCCTCCGCCCGCCAGAGCGTCCATCGGGACGCCCACCCGGGAGGGTCTCG), cg06000994 set forth in SEQ ID NO: 69 (CCGAGCGCTGCCCCCGCCGGCCCGCGGCTGCCAGCCGGCCCTGCCCGCGCCCGGGCCCGCGAGCGGCCGCACTTCACCTTACGGAGGGGAGATAATGAGATCAATTAGAGGCGCCGTCACC), cg10182317 described in SEQ ID NO: 70 (GGCAACCCTGACTCGGACCGCTCGGGAGAGCCCCAGGAGAGGCCAGCGCCGCGCAGCAGCCGCCCCGCTGCGCCCACCTCCCCGGCTGCTCCCGGAGGGCTCACAAAGGCGGTGGCCGCCCG), cg14222229 set forth in SEQ ID NO: 71 (GCGGGCGGCAGCCGCAAGCGAGGAATCCAGCGCAGGGAAAGTAGCCCCAGTGGGGCCCGGCGCGTCAGCCCCACTCGCGTGGCAAAACTTGCGGGGGCCCCCGCGTGCCGCGCCTCAGCCCA), cg04596005 set forth in SEQ ID NO: 72 (TCCTCGCCGTCGGGGTCCTCCTCCTCTGCCGACGAGTTGTCACTGGGCGAGGCGTAGCTGCGCTCTACGCCGCGGAGGGGCGGCCTCTTGGAGGCGGGGACCGGGTACTCCCGCTGCAGCCC), cg11592503 described in SEQ ID NO: 73 (GCTGCTCGCGCTCCGCCGCCCGGGAGATGCTTCCTCGCGCGGCGCAGCGCTGAGGCCGTGCGTGCGCCCGGCTGCGCTGCGCGCTCCCACATACACAAGCTCTCCATGTGAGCTGACAGG), cg05008595 set forth in SEQ ID NO: 74 (CTTCTCTTGAAAAGGAGGAGAATCAACACTGGGCTCACAACTCATCAGAGCTGAGTCATACGTACATCAGCAGGACCTACGTGGGAACCAAATAGCAAACTCAAATTGGGAAATTTGAGGAA), cg04999026 set forth in SEQ ID NO: 75 (CCGAGAGCCCCGCCTGCAGGCGGTGTAGATACATGTAGATACTGTAGATACTGTAGATACCGCCCCGGCGCCGACTTGATAAACGGTTTCGCCTCTTTTGGAAGCCGCCTGCGTGTCCATTT), cg04546413 set forth in SEQ ID NO: 76 (TGAGGAGTGAGGAGGCAGAAAGGACCGAGAACAAGGGGACCCGGTTCCATTTCTGGACCCCGTCCGCAGGCTGCTCGCCCGACTTGGGGTCGCTCTGCCCCGGACGATCAGGACAGCTGCGT), cg27254667 set forth in SEQ ID NO: 77 (CAAATCTATATGAAGGATCGAATTGCATTGAACTAGCAAACACACACACACACACGCACACGCAAAAACTGATGAAAGCTGAACAAGGTCTGTAGTCTAGTCAACAGTACTGCACTATGTGA), cg18902440 described in SEQ ID NO: 78 (ACAGTCTCTCGCCTCAAAGATCTCCGCCATTAGTGGTAGCCATTTAAGAAAACAGAATTACGATGAATAATGATTTGAAGCCAAAAAGTCAAAATATCTTATTTCGCAACTGTAATTGCTGG), and cg01315092 set forth in SEQ ID NO: 79 (CCACACAGGCCTCTCCCTCGGTGCGGTAGCGAGGGTTGCGGGCCCAAACGCCCGCGCCCACGGAGGCGCCTGCGACGACTAGAAGCTTCCACAGCCATATGGGGGCAAAGACGGCCCAGTAG). The biomarkers were shortlisted as progressively methylated CGIDs, with background methylation (<10%) in normal cells and an average increase in methylation of 10% or greater than 10% decrease during the transition from CIN1 to CIN3 stages, using the assumptions of the APDMA method disclosed herein. The Illumina method utilizes sequences flanking the CG loci to generate unique CG locus cluster IDs in a manner similar to dbSNP's NCBI refSNP ID (rs#).

[0051] [Table 1] [Table 2] [Table 3] [Table 4] [Table 5] [Table 6] [Table 7] [Table 8]

[0052] In one embodiment of the present invention, a panel of DNA methylation biomarkers for screening and early detection of cervical cancer is disclosed, the panel comprising CGIDs derived by APDMA method having sequences selected from the group consisting of SEQ ID NO: 1 to SEQ ID NO: 79 and combinations thereof, and optionally the panel is used in combination with other biomarkers as an early predictor of cervical cancer.

[0053] In one embodiment of the present invention, the polygenic DNA methylation biomarker is a combination of CGIDs listed in Table 2 below or a short subset of this list, such as the examples listed in Table 3 below, for early detection of cervical cancer and risk of cervical cancer in women with CIN1 to CIN3 precancerous lesions.

[0054] Therefore, in a further embodiment of the present invention, an in vitro method for obtaining an early predictor of cervical cancer is disclosed, which comprises the following steps: (a) measuring DNA methylation from a cervical specimen sample; (b) performing statistical analysis of the DNA methylation measurements obtained in step (a); (c) determining the DNA methylation status of multiple independent genomic CG locations, referred to as CGIDs, by performing an analysis of progressive DNA methylation changes (APDMA) of the genome-wide DNA methylation profile obtained in step (b); (d) classifying the CGIDs based on the frequency of their DNA methylation associated with the progression of precancerous stages of cervical cancer; and (e) obtaining candidate CGIDs from the classification in step (d) to obtain an early predictor of cervical cancer as a DNA methylation biomarker, wherein the candidate CGIDs as a DNA methylation biomarker for the early predictor of cervical cancer are selected from the group listed below. SEQ ID NO: 3 (CCGCCGCGGGTTCCCAGGGCTGGTGGTAGTTGCCGTCCCACACGTACGTGGCGGGGTCCTCGTCAGCGAAGACCTCGCGGAACATGTCGACCATGTAGAGGTCCTCGGCGCGGTTGCCATCC), SEQ ID NO:4 (GGGGAGGAATATTAGACTCGGAGGAGTCTGCGCGCTTTTCTCCTCCCCGCGCCTCCCGGTCGCCGCGGGTTCACCGCTCAGTCCCCGCGCTCGCTCCGCACCCCACCCACTTCCTGTGCTCG)、 SEQ ID NO:7 (GAGGAATATTAGACTCGGAGGAGTCTGCGCGCTTTTCTCCTCCCCGCGCCTCCCGGTCGCCGCGGGTTCACCGCTCAGTCCCCGCGCTCGCTCCGCACCCCACCCACTTCCTGTGCTCGCCC)、 SEQ ID NO:17 (GTGTGCTCAGCCTCAGCGTGAGGGGCACCTGCTCGTCTGGGCTCACAGCGAAGGCAGCCTCGCCGCGAGCTGCCGCTGCCGCTGCTGCCGCCACTGGTGTTGCCGCTCTCAGGCGCCAGGCT)、 SEQ ID NO:19 (GCGGCGGCGGGCGGGGAGCCAGGCCCGAGCTGCGTTCTGCGCAGCCATTGGTGGGCGCCGCGCTCTGCACTGAGCATGTTCGCGCCCCGCCGGCCCCTAGCCGCAGCCGCAGCCGCAGCGAC)、 SEQ ID NO:31 (GGCGGTGCGAGCTCCCCGCCTGCGGGACGCACGGAGACCGCGGTCAGCGCGCCGCCTGGCCGGCCCAGCGCGCCCAGCCCGCGCCCAGCCCCGTCCACTCCCGTCCAGCCCCGCCGCCCGGC)、 SEQ ID NO:34 (GTGTTCTCTGCGGCGGGCCGCGTCCCCGCTGAGCCTCGCGGTGACAGCCGCCTTTGGCAGCGAGCGCTCGGGGCACTTCTATCCCCGCCTCTCAAAGGGTGGGGACAGCCGTTTCCAGATTT)、 SEQ ID NO:39 (AGGGGAGCTGCGAGGCGAAGTGTTCTTCAGGAAGCGGGCTCGAGTCTCCGCAGCTGCGGCGGCGGCGGCGCGCTGGGCGGCGGCGGGGCGCGGGCAGGGGCCGGGGTGCCGCGCGG)、 sequence number 42 (AATCTCCCCTCGGGCTCGACGGATGTGCCCCCCAGATGTGCTGACACATGTCCGATGCCTCGCTGCCTTGGAGGTCTCCCCGCTCGCGTGTCTCTTCTCTTCGCACCAGCGGCGGAAACCGC)、 sequence number 43 (GCTCCGCTTCTCCGGGTTTTTAGCGGAAGCCTGCGGGGGGCGGGGTAACCGCGGAAGCCGGCGGCCGTGGGCGCGCGGGTTGGGGCTCTCGCGCCGCTCCGGGCTCTCCCCCCCCGGCTG)、 sequence number 49 (CCCCCTGTTCAAGGTCTGTCACCGTAGGGGGCGGGGGGGCGCGTGGAGCCGCTGGGGGTTCGGCCCACCCCGCGAACCGAGCTCCCGGCCCTGTGCCCCTCAGCTCTGCCGCGGGCGTTGG)、 sequence number 56 (GGCGCCGGCAGCTTCGCGCCGGCGGCTGGAAGCGGGCGGGCTGCACGGGCGGCTCGAGTGCGGGGACCCCAGCCCCTCGCCCTCTGAGCGCCGCCCCTGCCACCTGCTGCCAAGTCACCGG)、 sequence number 57 (CCCCGGCCGCGCCGGGCGGCGGGGCTCGGGATTCGGGAGACCGCGCGGCGCGAAGCCACGCGTCAGCCCCCACTGTCCCGGCGCCCTCGCCCCAGGCCTCGGGCTCTTCCTCCGCACCTCGTA)、 sequence number 58 (ACGCGGGGACTGGAAAGGGCGCCTGGGTGGGAAGAGGCGCTGGCGGGTGATCGTCCCCACCGGGCCAGTCCCCGGGATCTGCTGCCGCCCCTCTCCGAAATTCACAGCCAGAGCGGGCGCAC), SEQ ID NO: 65 (CTCTTTGGCAAGTGGTTTGTGCATCAGGAGAAACTTTCCACCTGCGAGCCGAACCGGCGCCGAGTGCGTGTGTTTCTGCCTTTTTTTGTTGTCGTTGCCTCCACCCCTCCCCATTCTTCTCT), and SEQ ID NO: 70 (GGCAACCCTGACTCGGACCGCTCGGGAGAGCCCCAGGAGAGGCCAGCGCCGCGCAGCAGCCGCCCCGCTGCGCCCACCTCCCCGGCTGCTCCCGGAGGGCTCACAAAGGCGGTGGCCGCCCG).

[0055] Table 2: A subset of polynucleotides selected from Table 1 that have CpG methylation sites useful in embodiments of the present invention. The 16 CGID biomarkers discussed herein can be found in Table 2 included in this application. These 16 candidate DNA methylation biomarkers are It was hypermethylated in the range of CIN1 and controls, with the largest effect size (Cohen's D > 1.3) between CIN3 and controls, and the largest Spearman correlation r > 0.4 with CIN phase progression. [Table 9] [Table 10]

[0056] In one embodiment of the present invention, a combination of DNA methylation biomarkers for screening and early detection of cervical cancer is disclosed, which detects cervical cancer by measuring the DNA methylation level of a CGID in DNA derived from a cervical specimen, and comprises the CGID derived using the APDMA method for deriving a "methylation predictor of cervical cancer" using a linear regression equation and a receiver operating characteristic (ROC) assay, wherein the CGID is selected from the group consisting of SEQ ID NO:3, SEQ ID NO:4, SEQ ID NO:7, SEQ ID NO:17, SEQ ID NO:19, SEQ ID NO:31, SEQ ID NO:34, SEQ ID NO:39, SEQ ID NO:42, SEQ ID NO:43, SEQ ID NO:49, SEQ ID NO:56, SEQ ID NO:57, SEQ ID NO:58, SEQ ID NO:65, SEQ ID NO:70, and combinations thereof.

[0057] Table 3: A subset of polynucleotides selected from Table 2 that have CpG methylation sites useful in embodiments of the present invention. Two CGID biomarkers, cgl3944175 set forth in SEQ ID NO: 3 (CCGCCGCGGGTTCCCAGGGCTGGTGGTAGTTGCCGTCCCACACGTACGTGGCGGGGTCCTCGTCAGCGAAGACCTCGCGGAACATGTCGACCATGTAGAGGTCCTCGGCGCGGTTGCCATCC) and cg03419058 set forth in SEQ ID NO: 31 (GGCGGTGCGAGCTCCCCGCCTGCGGGACGCACGGAGACCGCGGTCAGCGCGCCGCCTGGCCGGCCCAGCGCGCCCAGCCCGCGCCCAGCCCCGTCCACTCCCGTCCAGCCCCGCCGCCCGGC), discussed herein, are found in Table 3 included within this application. The subset of Table 3 represents the minimum number of CGID biomarkers that distinguish CIN3 precancerous lesions from controls identified using penalized regression reducing the number of CGIDs to five, followed by multivariable linear regression with these five CGIDs as independent variables and CIN3 status as the dependent variable. A linear regression equation consisting of the weighted methylation levels of these two sites was highly significant for predicting CIN3 (p<5x10 -15).

[0058] [Table 11]

[0059] In one embodiment of the present invention, a combination of DNA methylation biomarkers for screening and early detection of cervical cancer is disclosed, which detects cervical cancer by measuring the DNA methylation levels of CGIDs in DNA derived from cervical specimens, and comprises said CGIDs derived using the APDMA method for deriving "cervical cancer methylation predictors" using a linear regression equation and a receiver operating characteristic (ROC) assay, wherein the CGIDs are set forth in SEQ ID NO: 3 and SEQ ID NO: 31.

[0060] In one embodiment, the present invention provides kits and processes for detecting cervical cancer, comprising means and reagents for detecting DNA methylation measurements of a panel of polygenic DNA methylation biomarkers of cervical cancer.

[0061] In one embodiment, the present invention provides a kit for detecting cervical cancer, comprising means and reagents for measuring DNA methylation of the CGID biomarkers of Table 1 and combinations thereof.

[0062] In one embodiment, the present invention provides a kit for detecting cervical cancer, comprising means and reagents for measuring DNA methylation of CGIDs and deriving DNA methylation predictors of cervical cancer, and instructions for use, wherein the CGIDs are set forth in SEQ ID NO: 1 to SEQ ID NO: 79 and combinations thereof.

[0063] In one embodiment, the present invention provides a kit comprising a panel of CGIDs in the form of a chip for detecting cervical cancer, wherein the panel of CGIDs are set forth in SEQ ID NO: 1 to SEQ ID NO: 79 and combinations thereof.

[0064] In one embodiment, the present invention provides a kit that uses the CGID biomarkers disclosed herein.

[0065] In one embodiment, the present invention provides a kit using a DNA pyrosequencing methylation assay to predict cervical cancer by measuring DNA methylation of CGIDs, wherein the CGIDs are set forth in SEQ ID NO: 1 to SEQ ID NO: 79 and combinations thereof.

[0066] In one embodiment, the present invention provides a kit using a DNA pyrosequencing methylation assay to predict cervical cancer using the above-mentioned CGID biomarkers, for example, using the primers disclosed below and standard conditions for the pyrosequencing reaction recommended by the manufacturer (Pyromark, Qiagen): For cg03419058: The forward (biotinylated) primer set forth in SEQ ID NO: 80 has the polynucleotide sequence GGTTTTTGGGTAGGAAGGATAGTAG. The reverse primer is set forth in SEQ ID NO: 81, which has the polynucleotide sequence AAACAAATCTAACCCCTAAAAAAAC. A pyrosequencing primer set forth in SEQ ID NO: 82 having the polynucleotide sequence CAAACTAAACACACTAAACC. For cgl3944175: The forward primer is set forth in SEQ ID NO: 83 and has the polynucleotide sequence GGGTTTTTAGGGTTGGTGGTA. The reverse (biotinylated) primer set forth in SEQ ID NO: 84 has the polynucleotide sequence TCCTCATAATAATAAATAACAACC. A pyrosequencing primer set forth in SEQ ID NO: 85 having the polynucleotide sequence TATGTATGTGGTGGGGTT.

[0067] In one embodiment, the present invention provides a kit using a DNA pyrosequencing methylation assay to predict cervical cancer by measuring DNA methylation of a combination of CGIDs, wherein the forward, biotinylated primer is set forth in SEQ ID NO: 80, the reverse primer is set forth in SEQ ID NO: 81, and the pyrosequencing primer is set forth in SEQ ID NO: 82.

[0068] In one embodiment, the present invention provides a kit using a DNA pyrosequencing methylation assay to predict cervical cancer by measuring DNA methylation of a combination of CGIDs, wherein the forward, biotinylated primer is set forth in SEQ ID NO: 83, the reverse primer is set forth in SEQ ID NO: 84, and the pyrosequencing primer is set forth in SEQ ID NO: 85.

[0069] In one embodiment, the present invention provides a kit for predicting cervical cancer in DNA from cervical specimens using the above-described CGID biomarkers, employing a polygenic multiplexed amplicon bisulfite sequencing DNA methylation assay (PMSA). For example, the kit uses the primers and standard conditions disclosed below, including bisulfite conversion, and includes sequential amplification with target-specific primers (PCR1) followed by barcoding primers (PCR2), multiplexed sequencing on a single next-generation Miseq sequencer (Illumina), demultiplexing using Illumina software, data extraction and methylation quantification using standard methylation analysis methods, such as the Methylkit, followed by calculation of a weighted DNA methylation score and cancer prediction. The first PCR is performed as follows: For CGID cg03419058: The forward primer is set forth in SEQ ID NO: 80 and has the polynucleotide sequence: 5' GGTTTTTGGGTAGGAAGGATAGTAG 3'. The reverse primer is set forth in SEQ ID NO: 81 and has the polynucleotide sequence: 5' AAACAAATCTAACCCCTAAAAAAAC 3'. For CGID cg13944175: The forward primer is set forth in SEQ ID NO: 83 and has the polynucleotide sequence: 5' GGGTTTTTAGGGTTGGTGGTA 3'. The reverse primer is set forth in SEQ ID NO: 84 and has the polynucleotide sequence 5' TCCTCATAATAATAAATAACAACC 3'. To barcode (index) the samples, the present invention uses the following primers: A second PCR reaction was used. The forward primer is set forth in SEQ ID NO: 86 and has the polynucleotide sequence 5'AATgATACggCgACCACCgAgATCTACACTCTTTCCCTACACgAC 3'. The barcoding (reverse) primer is set forth in SEQ ID NO: 87, having the polynucleotide sequence: 5'CAAgCAgAAgACggCATACgAgATAGTCATCGgTgACTggAgTTCAgACgTg 3' (where the red base is an index, and 1200 variations of this index have been used).

[0070] In one embodiment, the present invention provides a kit that uses a multiplex target amplification bisulfite sequencing methylation assay on a next-generation sequencer to detect cervical cancer by measuring the DNA methylation levels of a combination of CGIDs, wherein the CGIDs are set forth in SEQ ID NO: 1 to SEQ ID NO: 79 and combinations thereof.

[0071] In another embodiment, the present invention provides a kit using a multiplex target amplification bisulfite sequencing methylation assay on a next-generation sequencer to detect cervical cancer by measuring the DNA methylation levels of a combination of CGIDs, wherein the CGIDs are set forth in SEQ ID NO: 3 and 5' as forward primer a primer set forth in SEQ ID NO: 88 having the polynucleotide sequence of 3' ACACTCTTTCCCTACACGACGCTCTTCCGATCTNNNNNGGGTTTTTAGGGTTGGTGGTA, and a 5' primer set forth in SEQ ID NO: 88 as a reverse primer The primer has the polynucleotide sequence set forth in SEQ ID NO: 89: GTGACTGGAGTTCAGACGTGTGCTCTTCCGATCTTCCTCATAATAATAAATAA CAACC 3'.

[0072] In another embodiment, the present invention provides a kit using a multiplex target amplification bisulfite sequencing methylation assay on a next-generation sequencer to detect cervical cancer by measuring DNA methylation levels of a combination of CGIDs, wherein the CGIDs are set forth in SEQ ID NO: 31, and a primer set forth in SEQ ID NO: 90 having a polynucleotide sequence of 5' ACACTCTTTCCCTACACGACGCTCTTCCGATCTNNNNNGGTAGGTTTTTGGGTAGGAAGGATAGTAG 3' as a forward primer, and 5' The primer has the polynucleotide sequence set forth in SEQ ID NO: 91, which is GTGACTGGAGTTCAGACGTGTGCTCTTCCGATCTAAACAAATCTAACCCCTAAAAAAAC 3'.

[0073] In one embodiment, the present invention provides for the use of a receiver operating characteristic (ROC) assay to detect cancer by defining a threshold between cervical cancer and normal cervix using weighted DNA methylation measurements of the CGID biomarkers in Table 1 and combinations thereof, or a subset of these CGIDs and combinations thereof, such as, for example, in Table 2. Samples above the threshold are classified as cancer.

[0074] In one embodiment, the present invention provides for the use of a hierarchical clustering analysis assay for predicting cancer for use in obtaining early detection of cancer positives by using measurements of methylation of the CGID biomarkers listed in Table 1 and combinations thereof.

[0075] In one embodiment, the present invention provides a kit that uses a mass spectrometry-based (Epityper®) or PCR-based methylation assay of DNA extracted from a sample to detect cancer by measuring the DNA methylation levels of a combination of CGIDs described in a panel of DNA methylation biomarkers for cervical cancer screening and early detection, wherein the panel is derived by APDMA methods having sequences selected from the group consisting of SEQ ID NO: 1 to SEQ ID NO: 79 and combinations thereof. The panel includes a selected CGID, and optionally the panel is used in combination with other biomarkers as an early predictor of cervical cancer.

[0076] In one embodiment, the present invention provides the use of a multivariate linear regression equation or neural network analysis to calculate a methylation score predictive of cervical cancer by using measurements of a combination of DNA methylation CGIDs described in a panel of DNA methylation biomarkers for the screening and early detection of cervical cancer, wherein the panel comprises CGIDs derived by the APDMA method having sequences selected from the group consisting of SEQ ID NO: 1 to SEQ ID NO: 79 and combinations thereof, and optionally the panel is used in combination with other biomarkers as an early predictor of cervical cancer.

[0077] In one embodiment, the present invention provides the use of a multivariate linear regression equation or neural network analysis to calculate a methylation score predictive of cervical cancer by measuring a combination of DNA methylation CGIDs described in "DNA methylation biomarker combinations for cervical cancer screening and early detection," wherein the combination comprises CGIDs derived using the APDMA method to detect cervical cancer by measuring the DNA methylation levels of said CGIDs in DNA from a cervical specimen and deriving a "methylation predictor of cervical cancer" using a linear regression equation and a receiver operating characteristic (ROC) assay, wherein the CGIDs are selected from the group consisting of those set forth in SEQ ID NO:3, SEQ ID NO:4, SEQ ID NO:7, SEQ ID NO:17, SEQ ID NO:19, SEQ ID NO:31, SEQ ID NO:34, SEQ ID NO:39, SEQ ID NO:42, SEQ ID NO:43, SEQ ID NO:49, SEQ ID NO:56, SEQ ID NO:57, SEQ ID NO:58, SEQ ID NO:65, SEQ ID NO:70, and combinations thereof.

[0078] In an alternative embodiment, the present invention provides the use of a multivariate linear regression equation or neural network analysis to calculate a methylation score that predicts cervical cancer by using measurements of a combination of DNA methylation CGIDs described in a DNA methylation biomarker combination, wherein the CGIDs are set forth in SEQ ID NO: 3 and SEQ ID NO: 31.

[0079] In one embodiment, the present invention provides the use of a receiver operating characteristic (ROC) assay to define a "methylation score" threshold that distinguishes cervical cancer from non-cancerous cervical tissue by using measurements of DNA methylation combinations described in a panel of DNA methylation biomarkers for the screening and early detection of cervical cancer, wherein the panel comprises CGIDs derived by the APDMA method having sequences selected from the group consisting of SEQ ID NO: 1 to SEQ ID NO: 79 and combinations thereof, and optionally the panel is used in combination with other biomarkers as early predictors of cervical cancer.

[0080] In one embodiment, the present invention provides the use of a receiver operating characteristic (ROC) assay to define a "methylation score" threshold that distinguishes cervical cancer from non-cancerous cervical tissue by using measurement of a DNA methylation combination described in "DNA methylation biomarker combinations for screening and early detection of cervical cancer," wherein the combination comprises CGIDs derived using the APDMA method for detecting cervical cancer by measuring the DNA methylation levels of said CGIDs in DNA from cervical specimens and deriving a "methylation predictor of cervical cancer" using a linear regression equation and a receiver operating characteristic (ROC) assay, wherein the CGIDs are selected from the group consisting of those set forth in SEQ ID NO:3, SEQ ID NO:4, SEQ ID NO:7, SEQ ID NO:17, SEQ ID NO:19, SEQ ID NO:31, SEQ ID NO:34, SEQ ID NO:39, SEQ ID NO:42, SEQ ID NO:43, SEQ ID NO:49, SEQ ID NO:56, SEQ ID NO:57, SEQ ID NO:58, SEQ ID NO:65, SEQ ID NO:70, and combinations thereof.

[0081] In an alternative embodiment, the present invention provides for the use of a receiver operating characteristic (ROC) assay to define a "methylation score" threshold that distinguishes cervical cancer from non-cancerous cervical tissue using combined DNA methylation measurements, wherein the CGIDs are set forth in SEQ ID NO: 3 and SEQ ID NO: 31.

[0082] In one embodiment, the present invention provides a computer-implemented method for obtaining candidate DNA methylation biomarkers for early detection of cervical cancer diagnosis, the method comprising: providing genome-wide DNA methylation data of a number of independent genomic CG positions, CGIDs of the human genome; processing the genome-wide DNA methylation data by normalization to derive normalized DNA methylation beta values; using the normalized DNA methylation beta values ​​to calculate Spearman correlation between advanced stages of pre-cancerous and non-transformed cervical cells; and obtaining candidate CGIDs by analysis of progressive DNA methylation changes (APDMA) to obtain candidate DNA methylation biomarkers for early detection of cervical cancer diagnosis. [Example]

[0083] The following examples are given as illustrations of the present invention and therefore should not be construed as limiting the scope of the present invention.

[0084] Example 1: Analysis of Progressive DNA Methylation Changes Method (APDMA) to identify and obtain CG locations (CGIDs) whose methylation levels are early predictors of cervical cancer. This invention addresses one of the outstanding challenges in cervical cancer screening: finding robust biomarkers that provide highly accurate and sensitive assessments of risk that can guide early intervention and treatment. A common approach is to use case-control logistic regression of genome-wide DNA methylation data to identify sites that are more or less methylated in either cancer cells or controls. However, due to dilution at the low frequency of cancer cells in the specimen, many statistically significant DNA methylation changes in cancers detected by these methods are heterogeneous, and many evolve late in cancer progression, making them of very limited value for early detection. Furthermore, mixtures of normal and cancer cells may erase quantitative differences in methylation profiles rather than categorical differences. As is well understood, DNA methylation is a binary property; a given cell is either methylated or unmethylated at a particular CG position in the genome.

[0085] In this example, the invention relates to methylated CGIDs, selected as a fundamental feature of cervical cancer, which are nearly uniformly methylated throughout cervical cancer specimens but never methylated in normal tissues, and which are classified as cervical cancers despite being classified as such. They appear very early in the precancerous stage in the context of normal cells and gradually increase in frequency from CIN1 to CIN3 stages as the cancer progresses. Categorically distinct methylated CGIDs in normal and cancer tissues have been found to be detectable even when cancer cells are detected at low frequencies in specimens by deep sequencing of bisulfite-converted DNA, which provides single-DNA molecule resolution. The frequency of molecules with methylated CGIDs represents the proportion of cancer cells in a sample. Measuring the methylation of such CGIDs by other methods can also determine the incidence of cancer cells in a specimen and serve as a DNA methylation biomarker for risk and prediction of cervical cancer in a sample.

[0086] It is clinically known that a proportion of CIN precancerous lesions develop into cervical cancer, and therefore they offer a particularly unique window for detecting early DNA methylation changes in cancer. Early prediction of who will develop cervical cancer is a clinically important issue. This is of paramount importance. The present invention provides a method for obtaining such early detection DNA methylation biomarkers, characterized by the following technical features: first, methylated CGIDs categorically characteristic of early cancer cells are not uniformly methylated in normal cervical tissue; second, these CGIDs are rarely methylated in early precancerous specimens; third, the frequency of these important methylated CGIDs should increase as the precancerous stage progresses from CIN1 to CIN3, predicting an increased risk of cervical cancer in women with CIN3 lesions; and fourth, because methylation of these CGIDs is a key hallmark of cervical cancer, these CGIDs must be uniformly abundant in cervical cancer specimens. In this example, certain CGIDs whose methylation increases with the progression of CIN stages from CIN1 to CIN3 were found to be ubiquitously methylated in cervical cancer specimens but, as described herein, not uniformly methylated in normal tissue. Thus, the presently disclosed method of the present invention provides a panel of candidate CGID biomarkers for the early detection of cervical cancer in women, particularly those with precancerous lesions.

[0087] As summarized in Figure 1 , the following steps of the progressive DNA methylation change (APDMA) method were performed to delineate CGID biomarkers whose methylation status detects early-stage cervical cancer.

[0088] cervical specimen This study used cervical specimens collected from women referred to McGill University Hospital for colposcopic examination due to abnormal cervical cancer screening results or for initial treatment of cervical lesions (19). Briefly, 643 women aged 16–70 years were enrolled between June 2015 and April 2016. Specimens were tested for the presence of oncogenic HPV DNA using the Roche cobas® 4800 HPV test, which detects HPV1 and HPV18 separately, and 12 other high-risk types (HPV 31, 33, 35, 39, 45, 51, 52, 56, 58, 59, 66, and 68) in pooled results. Pap smears were classified according to the Bethesda classification as follows: NILM: negative for intraepithelial lesion or malignancy; ASC-US: atypical squamous cells—significance unknown; ASC-H: atypical squamous cells—cannot exclude HSIL; LSIL: low squamous intraepithelial lesion; HSIL: high squamous intraepithelial lesion; AGC: atypical glandular cells; and carcinoma (20). Cervical abnormalities were biopsied, and histological results were graded by a senior pathologist at McGill as normal, CIN1, CIN2, CIN3, or invasive cancer. This study received ethical approval from the Institutional Review Boards of McGill University and the Jewish General Hospital. Study participants provided written informed consent.

[0089] The sample set consisted of 186 randomly selected physician-collected specimens from women: 50 had CIN1, 40 had CIN2, and 42 had CIN3, while 54 had normal biopsies.

[0090] DNA extraction and genome-wide methylation analysis DNA was extracted from the original exfoliated cervical cell specimen and suspended in liquid-based cytology PreservCyt solution (PreservCyt, Hologic Inc., Mississauga, IL). DNA extracted using a Qiagen DNA extraction kit was subjected to bisulfite treatment and hybridization to IlluminaEpic arrays using standard procedures described by the manufacturer at the Genome Quebec Innovation Center in Montreal. Epic arrays provide excellent coverage of the human promoter and enhancer repertoire, representing all known regions regulating transcription. Provides ledge (21).

[0091] Normalization of all samples and derivation of normalized DNA methylation values ​​(beta) Following the Illumina Infinium HD Technology User Guide and as recommended by the McGill Genome Quebec Innovation Center, samples were randomized with respect to their position on the slide and array, and all samples were hybridized and scanned simultaneously to mitigate batch effects. Illumina array hybridization and scanning were performed by the McGill Genome Quebec Innovation Center according to the manufacturer's guidelines. Illumina arrays were analyzed using the ChAMP Bioconductor package in R by Morris et al., 2014 (25). IDAT files were used as input for the champ.load function using the minfi quality control and normalization options. Raw data were filtered for probes with a detection value of P > 0.01 in at least one sample. In the current method, probes on the X or Y chromosome, probes containing SNPs identified by Marzouka et al., 2015 (24), and probes aligning to multiple locations identified by Marzouka et al., 2015 (24) were excluded to mitigate sex effects. Batch effects were analyzed for unnormalized data using the function champ.svd. Five of the first six principal components were associated with group and batch (slide). Intra-array normalization to adjust the data for bias introduced by the Infinium Type 2 probe design was performed using the function champ.norm (norm = " BMI ” ) was performed using beta mixture quantile normalization (BMIQ) (25). BMIQ was then normalized using the champ.runcombat function, after which batch effects were corrected.

[0092] Discovery of CGID, in which methylation frequency correlates with progression of CIN Next, in the current method, we used the beta values ​​of the batch-corrected normalized data to calculate Spearman correlations between CIN stages (stage code 0 for non-transformed healthy control cervical cells and stage codes 1–3 for CIN stages ranging from CIN1 to CIN3) using the Spearman corr function in R and correcting for multiple testing using Benjamini-Hochberg's "fdr" method (adjusted P-value (Q) of <0.05). The methylation levels of 7715 CGIDs were significantly correlated with the progression of precancerous CIN stages from 1 to 3 (q >0.05) (see Figure 2). As precancerous lesions progressed from normal to CIN1–CIN3 stages, most sites became hypermethylated, while a small proportion became hypomethylated (see Figure 2).

[0093] Shortlist of candidate CGIDs To identify CGIDs corresponding to the assumptions of the APDMA method, 79 progressively methylated CGIDs were shortlisted (see Table 1 above), which had background methylation in normal cells (<10% methylation) and an average increase of 10% or a decrease of >10% during the transition from CIN1 to CIN3. Next, we tested whether these CGIDs uniformly identified cervical cancers in publicly available Illumina 450K genome-wide DNA methylation data from 270 cervical cancer specimens (see GSE68339). Based on the DNA methylation of the tested CGIDs, we generated a heatmap of these 79 CGIDs whose methylation frequency increased during the progression of cervical precancerous stages, as obtained by the currently disclosed APDMA method. The heatmap revealed that these 79 CGIDs exhibit categorically distinct DNA methylation profiles between cervical cancer and normal cervix. Clearly, the majority of sites were completely unmethylated in normal tissues and highly methylated in cancer tissues, whereas a small number of sites were methylated in normal tissues and unmethylated in cervical cancer (see Figure 3). Thus, the method of the present invention allows even low-frequency methylation to be clearly seen in a background of completely unmethylated molecules. These hypermethylated CGIDs are associated with these as preferred biomarkers because they are easily detectable.

[0094] Example 2: Discovery of a polygenic DNA methylation biomarker set for early detection of cervical cancer. The present disclosure further shortlists 16 CGIDs from the list obtained and disclosed in Example 1 and Table 1, which are hypermethylated in the range of CIN3 to CIN1 and controls, and have the largest effect size (Cohen's D>1.3) between CIN3 and controls, and the largest Spearman correlation r>0.4 in CIN phase progression (see Table 2 above).

[0095] Next, to obtain the minimum number of CGIDs required to distinguish CIN3 precancerous lesions from controls, the method performed a penalty regression, reducing the number of CGIDs to 5. Next, the method performed a multivariate linear regression with these five CGIDs as independent variables and CIN3 status as the dependent variable. The remaining two CGIDs were significant (see Table 3 above). The linear regression equation composed of the weighted methylation levels of these two sites was highly significant for predicting CIN3 (p<5x10 -15 ).

[0096] Example 3: Utility of bigenic DNA methylation markers for detecting cervical cancer. Next, this disclosure first validated the bigenic DNA methylation markers (cg03419058; cgl3944175) on a publicly available database of 450K DNA methylation markers for cervical cancer (see GSE68339). A bivariate linear regression model with cervical cancer as the dependent variable and the methylation levels of the two CGIDs (cg03419058; cgl3944175) as independent variables was observed to be highly significant (p<2.2x10-16, F=8703, R=0.9873). The receiver operating characteristics (ROCs) of the methylation scores (calculated using the linear regression equations disclosed in Figure 4A) were compared by calculating the area under the curve (AUC) (see Figure 4B). The sensitivity and specificity of the bigenic methylation score for distinguishing cervical cancer from normal cervical tissue were observed to be 1 (see Figure 4C).

[0097] Therefore, the above DNA methylation biomarkers and calculated methylation scores may be useful for screening and early detection of cervical cancer not only in women at risk but also in the general healthy population of women using cervical specimens collected during routine gynecological Pap smears.

[0098] Example 4: Utility of bigenic DNA methylation biomarkers to determine cervical cancer methylation scores in individual specimens from healthy controls, CIN1 to CIN3, and cervical cancer patients. Methylation scores (predictive of cervical cancer) were calculated for each individual specimen: control, CIN1 to CIN3 (from the McGill cohort described in Example 1 above), and cervical cancer (see GSE68339) using the equation shown in Figure 4A (see Figure 5A) (for mean values ​​for the various groups, see Figure 5B). The results show an increase in methylation scores in advanced precancerous lesions, as expected from clinical observations of increased risk of cervical cancer with advancing CIN stage. The methylation score can be used to screen women with CIN lesions for cervical cancer risk.

[0099] Example 5: Spearman correlation between methylation score and progression from precancerous cervical cancer to cervical cancer. Spearman correlation analysis was performed between the methylation scores of cervical specimens from healthy conditions, precancerous stages CIN1 to CIN3, and cervical cancer (control, n = 54 CIN1, n=50; CIN2, n=40; ​​CIN3, n=42; cervical cancer, n=270). The results showed a highly significant correlation (p<2.2x10) between the bigenic marker methylation score and progression from pre-cancer to cancer. -16 and r = 0.88) (see Figure 6).

[0100] Example 6: Validation of a methylation biomarker (cgl3944175) for detecting cervical cancer. Because data for only one CGID biomarker was available in the TCGA cervical cancer data, the present disclosure calculated cervical cancer methylation scores using a linear regression equation using DNA methylation data for only the CGID, cgl3944175. Spearman correlations were calculated between cancer stage and methylation score (see statistics in Figure 7A and correlation chart in Figure 7B). In this disclosure, CIN1 to CIN3 are from the McGill cohort, as previously described in Example 1 of the application, and score assignments are based on an assigned scale: control: 0, CIN1-3: 1-3, respectively, and cervical cancer: 4.

[0101] Example 7: Utility of bigenic methylation biomarkers for detecting cervical cancer in precancerous cervical specimens. We used bigenic methylation biomarkers to predict which CIN1 to CIN3 samples would progress to cervical cancer. A methylation score was calculated for each specimen based on the methylation values ​​of two CG sites obtained from Epic array data. Using a cancer threshold calculated from a comparison of cervical cancer and healthy cervical specimens (see Figure 3), predictions were made for each sample (see Figure 8A). As expected, the percentage of specimens predicted to become cancerous increased from a few percent for CIN1 specimens to 60% for CIN3 specimens (see Figure 8B).

[0102] Although the present invention has been described in relation to its preferred embodiments, it should be understood that many other possible modifications and variations can be made without departing from the spirit and scope of the invention. [Industrial Applicability]

[0103] These new DNA methylation biomarkers can be developed into diagnostic kits for the early and accurate diagnosis of human cervical cancer. They are direct indicators of cellular changes during the initiation and development of cervical cancer. They are nearly uniformly methylated throughout cervical cancer specimens but unmethylated in normal tissues, demonstrating a fundamental hallmark of cervical cancer with a gradual increase in frequency from CIN1 to CIN3 precancerous stages. These biomarkers not only complement pathology for the accurate early detection of cervical cancer in CIN lesions, but also serve as early detection and risk prediction biomarkers in asymptomatic women. These biomarkers provide a useful perspective on existing epigenetic DNA methylation markers, which play a key role in gene regulation, and can be used as diagnostic tools in the form of CGID. These biomarkers could provide a fast, inexpensive, accurate, stable, and high-throughput diagnostic kit for accurate, early, and yet unfeasible diagnosis of human cervical cancer at precancerous stages that are currently inaccessible. [Prior art documents] [Non-patent literature]

[0104] [Non-Patent Document 1] El-Zein M, Richardson L, Franco EL. Cervical cancer screening of HPV bacterial populations: Cytology, molecular testing, both or none. J. Clin. Virol. 2015;76:S62-S68. doi: 10.1016 / j.jcv.2015.11.020. [Non-patent document 2] Boers A, Wang R, van Leeuwen RW, et al. Discovery of new methylation markers to improve screening for cervical intraepithelial neoplasia grade 2 / 3. Clin. Epigenetics 2016;8(29). doi: 10.1186 / sl3148-016-0196-3.

Table 3

Fashion 4

Wood 5

Outdoor Configuration 6

Direct Environment 7

Outdoor Track 8

Outdoor Tools9

Outdoor Tools 10

Outdoor Content11

Outdoor Tools 12

Outdoor Content13

Outdoor Tools 14

Outdoor Tools 15

Non-Patent Document 16

Non-Patent Document 17

Non-Patent Document 18

Non-Patent Document 19

Non-Patent Document 20

Non-Patent Document 21

Non-Patent Document 22

Claims

1. A kit for detecting cervical cancer, comprising means, reagents and primers for a DNA methylation assay for predicting cervical cancer by measuring DNA methylation of at least CGID set forth in SEQ ID NO: 3, wherein the primers are selected from the following: (a) a biotinylated forward primer consisting of nucleotide sequence SEQ ID NO: 83, a reverse primer consisting of nucleotide sequence SEQ ID NO: 84, and a pyrosequencing primer consisting of nucleotide sequence SEQ ID NO: 85 for measuring DNA methylation of the CGID set forth in SEQ ID NO: 3; or (b) a forward primer consisting of nucleotide sequence number 88 and a reverse primer consisting of nucleotide sequence number 89 for measuring DNA methylation of the CGID set forth in SEQ ID NO: 3;

2. The kit described in claim 1, further comprising reagents and primers for measuring DNA methylation of CGID described in sequence number 31, wherein the primers are selected from the following: (c) a biotinylated forward primer consisting of nucleotide sequence SEQ ID NO: 80, a reverse primer consisting of nucleotide sequence SEQ ID NO: 81, and a pyrosequencing primer consisting of SEQ ID NO: 82 for measuring DNA methylation of the CGID set forth in SEQ ID NO: 31; or (d) a forward primer consisting of nucleotide sequence number 90 and a reverse primer consisting of nucleotide sequence number 91 for measuring DNA methylation of the CGID described in sequence number 31;

3. 3. The kit of claim 1 or claim 2 for predicting cervical cancer by DNA pyrosequencing methylation assay.

4. 3. The kit of claim 1 or claim 2 for detecting cervical cancer by multiplex target amplification bisulfite sequencing methylation assay.

5. A kit as described in claim 1 or claim 2 for detecting cervical cancer by methylation assay using mass spectrometry (Epityper (registered trademark)) or PCR method of DNA extracted from a sample.

6. A kit as described in claim 1 or claim 2, further comprising reagents and primers for measuring DNA methylation of CGIDs described in SEQ ID NO: 4, SEQ ID NO: 7, SEQ ID NO: 17, SEQ ID NO: 19, SEQ ID NO: 34, SEQ ID NO: 39, SEQ ID NO: 42, SEQ ID NO: 43, SEQ ID NO: 49, SEQ ID NO: 56, SEQ ID NO: 57, SEQ ID NO: 58, SEQ ID NO: 65, SEQ ID NO: 70, and combinations thereof.

Images