DETECTION OF HYPERMETHYLATED GENES FOR DIAGNOSING GASTRIC CANCER

Abstract

The invention relates to a method for diagnosing or identifying gastric cancer in a subject. The inventors indeed identified three DNA methylation biomarkers that, alone, or preferably in combination, can help diagnosing or following-up gastric cancer patients very specifically, discriminating with other type of cancers. Further, it can be used for determining, and/or adapting a suitable therapeutic regimen for a subject diagnosed for gastric cancer. The present invention also relates to kits comprising primers or probes to detect, diagnose, or identify hypermethylated genes.

Claims

1. An in vitro method comprising determining the level or amount of methylation of the MSC-AS1 gene and of at least one gene selected from the group consisting of: the ZNF790-AS1 gene and the KCNA3 gene, in a biological sample of said subject.

2. The method of claim 1, wherein if said genes are hypermethylated as compared with reference values, then said subject is diagnosed or identified as suffering from a gastric cancer.

3. The method of claim 1, wherein said biological sample is a blood sample of said subject.

4. The method of claim 1, wherein said methylation is determined by Next Generation Sequencing (NGS) or by qPCR, preferably by digital PCR (dPCR).

5. The method of claim 1, wherein said level or amount of methylation is determined in the nucleotide region of SEQ ID NO:4 in the MSC-AS1 gene and in the nucleotide region of SEQ ID NO:5 in the ZNF790-AS1 gene or in the nucleotide region of SEQ ID NO:6 in the KCNA3 gene.

6. The method of claim 1, wherein said level or amount of methylation of the MSC-AS1 gene and of the ZNF790-AS1 gene or of the KCNA3 gene is determined by dPCR by using the primers of SEQ ID NO:7-10, and SEQ ID NO:11-14 or SEQ ID NO:15-16 respectively, and using the probes of SEQ ID NO:17-18 and of SEQ ID NO:19-20 or of SEQ ID NO:21 respectively.

7. The method according to claim 1, comprising determining simultaneously or sequentially the level or amount of methylation of the MSC-AS1 gene, the ZNF790-AS1 gene and the KCNA3 gene in a biological sample of said subject

8. An in vitro method comprising: a) determining the level or amount of methylation of the MSC-AS1 gene and of at least one gene selected from the group consisting of the ZNF790-AS1 gene and the KCNA3 gene, in a biological sample of said subject, at a first time point, b) determining the level or amount of methylation of said genes selected previously in the step a), in a biological sample of said subject, at a second time point, and c) comparing the level or amount of methylation determined in step a) to the level or amount determined in step b) or to reference values.

9. The method of claim 8, wherein the sample in step a) is obtained prior to the treatment for gastric cancer and the sample in step b) is obtained after said subject has been treated for gastric cancer.

10. The method according to claim 8, wherein the level or amount of methylation is determined simultaneously or sequentially in the MSC-AS1 gene, the ZNF790-AS1 gene and the KCNA3 gene in a biological sample of said subject.

11. The method of claim 8, wherein said reference values are obtained in a healthy subject.

12. The method of claim 8, for monitoring the progress of gastric cancer in a subject being diagnosed for gastric cancer or for determining or adapting a therapeutic regimen suitable for a subject suffering from gastric cancer.

13. The method of claim 8, for predicting a clinical outcome of a subject afflicted with gastric cancer.

14. A kit comprising primers targeting specifically the nucleotide region of SEQ ID NO:4 in the MSC-AS1 gene and either the nucleotide region of SEQ ID NO:5 in the ZNF790-AS1 gene or the nucleotide region of SEQ ID NO:6 in the KCNA3 gene.

15. (canceled)

16. The kit of claim 14, said kit further comprising probes targeting specifically the nucleotide region of SEQ ID NO:4 in the MSC-AS1 gene and either the nucleotide region of SEQ ID NO:5 in the ZNF790-AS1 gene or the nucleotide region of SEQ ID NO:6 in the KCNA3 gene.

17. The kit of claim 14, comprising the primers of SEQ ID NO:7-10 and either SEQ ID NO:11-14 or SEQ ID NO:15-16 and their corresponding probes of SEQ ID NO:17-21.

18. A method, for: a) diagnosing or identifying gastric cancer in a subject, b) predicting a clinical outcome in a subject afflicted with gastric cancer, c) determining the therapeutic regimen of a subject with gastric cancer, and/or d) monitoring the progress of gastric cancer in a subject being diagnosed for gastric cancer, said method comprising using the kit of claim 14.

Description

FIGURE LEGENDS

[0123] FIG. 1 discloses the DNA methylation level of the selected biomarkers, ZNF790-AS1, MSC-AS1, and KCNA3 in tumor and non-tumor adjacent tissues from GC patients by Methyl-seq (n=10). Paired non-parametric Wilcoxon test was used for the analysis of the hypermethylation difference between normal and adjacent tissue DNA.

[0124] FIG. 2 discloses the validation of DNA methylation of selected biomarkers, ZNF790-AS1, MSC-AS1, and KCNA3 by ddPCR (n=20). Paired non-parametric Wilcoxon test was used for the analysis of the hypermethylation difference between tumor and adjacent normal tissue DNA.

[0125] FIG. 3 discloses the validation of DNA methylation of MSC-AS1 biomarker in GC patients by ddPCR. The difference of DNA methylation in GC patient plasma (n=22) and healthy individuals plasma (n=10) was tested by ddPCR. Mann-whitney test was used for the analysis of the significance.

[0126] FIG. 4 discloses the survival probability of patients according to the detected quantity of methylated DNA. The GC patients were separated into three tertiles depending on the quantity of copy of methylated sequences detected per milliliter of plasma. Log-rank, tft: logrank trend test.

EXAMPLES

[0127] Materials and Methods:

[0128] Patients and Healthy Control Individuals

[0129] Matched tumor and adjacent non-tumor tissue biopsies from 20 gastric cancer patients were included in this study. Plasma samples from 33 healthy individuals were purchased from Biological Specialty Corporation (Colmar, USA) and BioPredict. Inc (Oradell, USA). Plasma from 55 advanced gastric cancer patients were collected in EDTA tubes. This study was approved by the local ethics committee and informed written consent was obtained from all the patients.

[0130] Tumor Sample Preparation, Storage, DNA Extraction, and Quantification

[0131] Tumor and adjacent non-tumor tissue biopsies were flash frozen in liquid nitrogen immediately after resection until further analysis. Each tumor was reviewed by a pathologist and the tumor cell content was assessed by hematoxylin-eosin-safran staining. DNA was extracted with the QIAampDNAMini Kit (Qiagen) according to the manufacturer's instructions. DNA concentration was measured by Qubit 2.0 fluorometer (Invitrogen, Life Technologies) with the use of the dsDNA BR Assay (Invitrogen). Extracted DNA samples were stored at −20° C. before testing.

[0132] Plasma Sample Preparation, Storage, DNA Extraction, and Quantification

[0133] Plasma samples of healthy individuals were received in dry ice, aliquoted and immediately frozen at −80° C. Before extraction, plasma samples were centrifuged at 3000 g for 10 min and then extracted with the use of the QIAmp Circulating Nucleic Acid Kit (Qiagen) or the ccfDNA Plasma kit (Promega) by RSC Maxwell instrument according to the manufacturer's instructions. Plasma form gastric cancer patients were extracted with the use of the ccfDNA Plasma kit (Promega) by RSC Maxwell instrument or the QIAmp Circulating Nucleic Acid Kit (Qiagen). The quantity of DNA was measured by Qubit 2.0 fluorometer (Invitrogen, Life Technologies) with the use of the dsDNA HS Assay (Invitrogen). Extracted DNA samples were stored at −20° C. before testing.

[0134] Buffy Coat Preparation, DNA Extraction, and Quantification

[0135] Buffy coat was prepared from whole blood of patients that do not suffer from gastric cancer or from healthy individuals and then extracted with the use of the QIAmp Circulating Nucleic Acid Kit (Qiagen) according to the manufacturer's instructions. The quantity of DNA was measured by Qubit 2.0 fluorometer (Invitrogen, Life Technologies) with the use of the dsDNA BR Assay (Invitrogen). Extracted DNA samples were stored at −20° C. before testing.

[0136] Genome-Wide DNA Methylome Profiling in Gastric Cancer Patients by Methyl-Seq

[0137] DNA methylome libraries were prepared with the use of SeqCap Epi CpGiant Enrichment kit (Roche) according to the manufacturer' instructions. Briefly, 500 ng to 1 μg of tumor or non-tumor tissue DNA was fragmented with the use of Bioruptor (Diagenode s.a, Belgium). NGS adaptors were ligated to the fragmented DNA followed by the bisulfite conversion treatment with the use of EZ DNA Methylation-Lightning Kit (Zymo research). After several steps of purification and amplification, bisulfite-converted samples were hybridized with SeqCap Epi probe pool and then captured by SeqCap Pure Capture Beads. Post capture amplification was applied before passing the samples to the Nextseq™ 500 sequencer (Illumina).

[0138] Analysis and Identification of DNA Methylation Biomarkers Based on Methyl-Seq Data

[0139] Following a quality control, sequence data was aligned to bisulfite converted reference genome and analyzed for differentially methylated CpG sites (DMC) and differentially methylated regions (DMR) between tumor and non-tumor DNA with the use of Methylene: an R script. DMRs are the regions containing at least 5 DMCs in more than 80% of the samples. A list of DMRs was generated including different information: the level of methylation in non-tumor tissue DNA, the level of methylation in tumor tissue DNA, the difference of methylation level between tumor and non-tumor tissue DNA, q value, gene symbol corresponding to the DMRs identified etc. Based on this list of DMR generated, different parameters were then used for identifying DNA methylation biomarkers for gastric cancer patients. First, all the DMRs with a q value less than 0.05 were selected. Second, DMRs with the difference of methylation level more than 40% between tumor and non-tumor DNA were chosen. Third, DMRs satisfied with the previous 2 criteria were then selected by the condition that the methylation level in non-tumor DNA was less than 5%. Following this, DMRs in the intergenic region were removed from the candidate biomarkers list and DMRs annotated with the genes were verified in the public database to remove those which have been described before.

[0140] Tissue DNA, Buffy Coat DNA and Plasma ccfDNA Bisulfite Conversion

[0141] Tissue DNA, buffy coat DNA and plasma ccfDNA were modified by bisulfite using the EZ DNA Methylation-Gold Kit (Zymo Research). In brief, bisulfite reaction was carried out in a thermocycler at 98° C. for 12 min and 64° C. for 2 h35 min. The cleanup of bisulfite-converted DNA followed the recommendations of the manufacturer and converted DNA was eluted in M-Elution Buffer and stored at −20° C.

[0142] Detection of Methylation Changes of Selected Biomarkers by Droplet-Based Digital PCR

[0143] The hypermethylation of selected biomarkers in tumor DNA was validated by ddPCR. Duplex format was used to analyze hypermethylation with albumin for normalizing the DNA amount. Primers and probes were listed in the Table 1 and 2 above. In particular, the primers pair of SEQ ID NO: 9-10 and probes of SEQ ID NO: 18 for MSC-AS1, primers pair of SEQ ID NO: 11-12 and probes of SEQ ID NO: 19 for ZNF790-AS1 and primers pair of SEQ ID NO: 15-16 and probes of SEQ ID NO: 21 for KCNA3 were used.

[0144] DNA methylation of targeted sequences was analyzed by ddPCR using either the Raindrop ddPCR system (Raindance technologies, today Bio-Rad) or the QX-200 platform (BIO-RAD Technologies). In brief, when using the first system, 12.5 μL Kapa probe Fast qPCR master mix (Kapa Biosystems) was mixed with the assay solution containing: 0.75 μL 40 mM dNTP Mix (New England BioLabs), 0.5 μL 25 mM MgCl2, 1 μL 25× Droplet Stabilizer (RainDance Technologies), 1.25 μL 20× Assay Mix containing 8 μM of forward and reverse primers, 4 μM of 6-FAM and 12 μM of VIC Taqman® labeled-probes, and target modified DNA template to a final reaction volume of 25 μL. When possible, a minimum of 10 ng of modified DNA was used in each reaction.

[0145] When using the second system, the mixture of PCR reagents (BIO-RAD Technologies) was prepared following manufacturer's recommendations. A triplex panel was developed to detect the KCNA3 and MSC-AS1 methylated target sequences as well as the unmethylated Albumin sequence as a reference gene. A duplex assay detecting ZNF790-AS1 methylated target sequence as well as the unmethylated Albumin sequence as a reference gene was also developed. In the triplex panel, final concentrations in reaction mix of forward and reverse primers were 400 nM for albumin and KCNA3 and 600 nM for MSC-AS1. 600 nM of VIC Taqman® labeled-probes were used to detect albumin unmethylated sequences, 400 nM of FAM Taqman® labeled-probes were used to detect MSC-AS1 methylated sequences and 200 of both FAM and VIC Taqman® labeled-probes were used to detect KCNA3 methylated sequences (see Method 1 for primer and probe sequences). The duplex detecting the target sequence of ZNF 790-AS1 was also optimized for the QX-200 platform. In this assay, final concentrations in reaction mix of forward and reverse primers were 400 nM for albumin and 600 nM for ZNF790-AS1. Moreover, 600 nM of VIC Taqman® labeled-probes were used to detect albumin unmethylated sequences, 400 nM of FAM Taqman® labeled-probes were used to detect ZNF790-AS1 methylated sequences. The PCR step for ddPCR was performed on a BIO-RAD C1000 or 51000 using the following program: 10 min at 95° C. (using a 2.5° C./second ramp rate), followed by 45 cycles of: 94° C., 30 s and 57.3° C. (triplex assay) or 58.4° C. (duplex assay), 1 min (using a 2.5° C./second ramp rate), with an ultimate step of 10 min at 98° C. After completion, the emulsions were either stored at 4° C. or processed immediately to measure the end-point fluorescence signal from each droplet. Data was analyzed using the Quantasoft BIO-RAD software as described by the manufacturer. Populations were clustered according to fluorescence levels, allowing to precisely count both tumor and normal amplifiable DNA molecules.

[0146] For the different assays, limit of blank (LOB) was calculated as described previously [Taly, V., et al. 2013] using commercial DNA extracted from whole Blood (Promega) for both duplex and triplex developed assays. It is defined by the frequency of positive droplets measured in normal control DNA samples with no hyper-methylated DNA present (n=10 to 23 for each assay depending on samples availability). The calculated LOB was subtracted from each sample for calculating their methylation level.

[0147] The sample analysis was performed following the procedure described earlier [Taly, V., et al. 2013]. Samples were considered positive when the number of observed droplets was higher than LOB value. The methylation level of each sample was calculated as the ratio of the number of droplets containing methylated sequences over the number of droplets containing albumin sequences.

[0148] Two DNA controls were used for ensuring the proper realization of the modification treatment (Positive control: universal hypermethylated DNA and negative control: normal human genomic DNA).

[0149] Measurement of the Methylation Level of Selected Biomarkers in Plasma Circulating Cell Free DNA (ccfDNA) from Healthy Individuals and Gastric Cancer Patients

[0150] DNA methylation of selected biomarkers in plasma from healthy individuals or gastric cancer patients was measured by ddPCR with the use of same reaction conditions as described before. Duplex format was used to analyze hypermethylation with albumin for normalizing the DNA amount. The same primers and probes were used as listed in the Table 1 and 2 above. In particular, the primers pair of SEQ ID NO: 9-10 and probes of SEQ ID NO: 18 for MSC-AS1, primers pair of SEQ ID NO: 11-12 and probes of SEQ ID NO: 19 for ZNF790-AS1 and primers pair of SEQ ID NO: 15-16 and probes of SEQ ID NO: 21 for KCNA3 were used.

[0151] Calculation of Detection Sensitivity and Specificity

[0152] The sum of the average and standard deviation of methylation level in non-tumor tissue DNA was used as the threshold for calculating sensitivity and specificity of each selected biomarker. Sensitivity is the percentage of the patients showing higher methylation level in tumor tissues than the threshold. Specificity is the percentage of the patients showing lower methylation level in non-tumor tissues than the threshold.

[0153] Statistical Analysis

[0154] Statistical analyses were performed using Prism Software (GraphPad Software Inc.) and R software 3.6.3 Studio. A P value 0.05 was considered as significant. For the analysis of the hypermethylation difference between normal and adjacent tissues, paired non-parametric Wilcoxon test was used. For the analysis of the hypermethylation difference between healthy and gastric cancer plasma ccfDNA, Mann-whitney test was used. The demographic characteristics of patients were compared by the Chi-square or Fisher's exact test. Continuous data were analyzed with the independent-samples t-test. Survival rates were calculated using the Kaplan-Meier method. Progression-free survival (PFS) was measured from the date of blood sampling used for ctDNA determination in the study to the first documented radiological progression (RECIST V1.1), or death, whichever occurred first. Patients lost to follow-up were censored at last follow-up visit. Association between outcomes and baseline characteristics was assessed with Cox proportional hazard models and hazard ratio (HR) with 95% CI were provided.

[0155] Results

[0156] Identification of DNA Methylation Biomarkers Based on Methyl-Seq Data

[0157] A total of 20 samples (paired tumor and non-tumor tissue DNA) from 10 gastric cancer (GC) patients were analyzed for methylome profiles by Methyl-seq following SeqCap Epi CpGiant Enrichment libraries preparation. 187,166 DMR was identified in gastric tumor tissue DNA compared with non-tumor tissue DNA. Among them, 15,543 DMRs were with a q value less than 0.05. Following this, a more stringent threshold: difference of methylation level is more than 40% between tumor DNA and non-tumor DNA and the methylation level is less than 5% of methylation in non-tumor DNA was applied for the identification of potential DNA methylation biomarkers for gastric cancer. Finally, 3 potential biomarkers: KCNA3, ZNF790-AS1, and MSC-AS1 were identified.

[0158] Hypermethylation of the Selected Biomarkers in Tumor DNA from GC Patients by Methyl-Seq

[0159] As shown in FIG. 1, compared with those in non-tumor tissue DNA, methylation level of the selected biomarkers: KCNA3, ZNF790-AS1, and MSC-AS1, was significantly increased in gastric tumor tissue DNA. The detection sensitivity and specificity of all these biomarkers by Methyl-seq were more than 80%:

TABLE-US-00003 TABLE 3 sensitivity and specificity of the biomarkers of the invention when assessed alone by MethylSeq Methyl-seq ZNF790-AS1 MSC-AS1 KCNA3 Sensitivity 100 90 80 Specificity 80 100 90

[0160] Validation of DNA Methylation Changes of Selected Biomarkers by ddPCR

[0161] The difference of DNA methylation of several potential biomarkers in tumor and non-tumor tissue from GC patients was validated by ddPCR (n=20) with primers pair of SEQ ID NO: 9-10 and probes of SEQ ID NO: 18 for MSC-AS1, primers pair of SEQ ID NO: 11-12 and probes of SEQ ID NO: 19 for ZNF790-AS1 and primers pair of SEQ ID NO: 15-16 and probes of SEQ ID NO: 21 for KCNA3. Compared to those in non-tumor tissue, DNA methylation in tumor tissue was significantly increased as already demonstrated by Methyl-seq (FIG. 2).

[0162] The detection sensitivity and specificity of the biomarkers MSC-A1, KCNA3 and ZNF-90-AS1 by ddPCR in gastric cancers were more than 80%:

TABLE-US-00004 TABLE 4 sensitivity and specificity of the biomarkers of the invention when assessed alone by ddPCR >Average ( Non KCNA3 ZNF790-AS1 MSC-AS1 tumor) + 1SD (12.0%) (11.82%) (11.29%) sensitivity 80 85 85 Specificity 90 90 90

[0163] Combination of Different Biomarkers to Reach a Higher Sensitivity and Specificity

[0164] The sensitivity and specificity of each individual biomarker by Methyl-seq and ddPCR are shown in Table 3 and Table 4 respectively. The highest sensitivity with the use of only one biomarker is 85%. To reach a higher detection sensitivity and specificity, different biomarkers can be combined (Table 5 and 6, results obtained by MethylSeq or ddPCR respectively). As targeted in the invention, the detection sensitivity by ddPCR can furthermore be improved to 100% by combining MSC-AS1 with ZNF790-AS1 or MSC-AS1 with KCNA3.

TABLE-US-00005 TABLE 5 Detection sensitivity and specificity of different combination of selected biomarkers when assessed by MethylSeq Specificity Sensitivity KCNA3 ZNF790-AS1 MSC-AS1 KCNA3 ZNF790-AS1 MSC-AS1 KCNA3 100  90 80 ZNF790-AS1 100 90 20 MSC-AS1

TABLE-US-00006 TABLE 6 Detection sensitivity and specificity of different combination of selected biomarkers when assessed by ddPCR Specificity Sensitivity KCNA3 ZNF790-AS1 MSC-AS1 KCNA3 ZNF790-AS1 MSC-AS1 KCNA3 85 90 90 ZNF790-AS1 90 90 90 MSC-AS1

[0165] Methylation Profiles of the Selected Biomarkers in Plasma ccfDNA and Buffy Coat Fraction from Healthy Individuals and Patients by ddPCR

[0166] The purpose of these potential biomarkers is to detect DNA hypermethylation in gastric cancer patients but not in healthy individuals. In order to validate this, the methylation level in plasma ccfDNA from healthy individuals was investigated by ddPCR. No significant level of methylation was observed in healthy plasma ccfDNA for all these potential biomarkers: KCNA3, ZNF790-AS1 and MSC-AS1 (FIG. 3 for MSC-AS1, results for other markers (Table 7 and Table 9 for duplex and triplex assays respectively)). Moreover, DNA extracted from Buffy Coat (from patients that do not suffer from gastric cancer, from patients with gastric cancer or healthy individuals) was also tested with the markers and no methylation of the tested markers was observed in these samples (Table 8 and 10 for duplex and triplex assays respectively). The number of the samples ran for each biomarker depended on the availability of DNA. Buffy coat DNA 1-6 were from colorectal cancer patients, buffy coat DNA 7-16 were from healthy individuals and other buffy coats were from Gastric cancer patients. As mentioned above for LOB calculation, commercial DNA extracted from whole Blood (Promega) was also used to validate the assay. Added to buffy coat DNA, these controls were performed to ensure that the markers were not positive in cells contained in the blood ensuring no false positive in case of blood cell hemolysis (for example due to pre-analytical sample handling).

[0167] However, when tested on DNA extracted from plasma of healthy subjects and gastric cancer patients, methylation was observed only for gastric patients (see below) both using duplex and triplex developed assays leading to a potential specificity of 100% for the detection of GC in blood samples.

TABLE-US-00007 TABLE 7 Methylation level of the selected biomarkers in healthy plasma circulating cell free (ccfDNA) measured by ddPCR in duplex assays. Methylation level of selected biomarkers was measured in healthy plasma ccfDNA (n = 8 to 14 depending on the availability) by ddPCR. Observed droplets Methylation of Sample ID Alb ZNF790-AS1 ZNF790-AS1 (%) Healthy plasma 1 136 0 0 Healthy plasma 2 522 0 0 Healthy plasma 3 416 0 0 Healthy plasma 4 391 0 0 Healthy plasma 5 810 0 0 Healthy plasma 6 280 0 0 Healthy plasma 7 327 0 0 Healthy plasma 8 199 0 0 Observed droplets Methylation of Sample ID Alb MSC-AS1 MSC-AS1 (%) Healthy plasma 9 319 0 0 Healthy plasma 10 862 0 0 Healthy plasma 11 136 0 0 Healthy plasma 12 95 0 0 Healthy plasma 13 141 0 0 Healthy plasma 14 100 0 0 Healthy plasma 15 140 0 0 Healthy plasma 16 92 0 0 Healthy plasma 17 349 0 0 Healthy plasma 18 146 0 0 Observed droplets Methylation of Sample ID Alb KCNA3 KCNA3 (%) Healthy plasma 19 124 0 0 Healthy plasma 20 157 0 0 Healthy plasma 21 377 0 0 Healthy plasma 22 304 0 0 Healthy plasma 23 271 0 0 Healthy plasma 24 195 0 0 Healthy plasma 25 155 0 0 Healthy plasma 26 314 0 0 Healthy plasma 27 123 0 0 Healthy plasma 28 143 0 0 Healthy plasma 29 273 0 0 Healthy plasma 30 307 0 0 Healthy plasma 31 273 0 0 Healthy plasma 32 276 0 0

TABLE-US-00008 TABLE 8 Methylation level of the selected biomarkers in healthy or CRC buffy coat DNA measured by ddPCR in duplex assays. Observed droplets Methylation of Sample ID Alb ZNF790-AS1 ZNF790-AS1 (%) Buffy coat1 89308 0 0.00 Buffy coat2 84986 0 0.00 Buffy coat3 133999 0 0.00 Buffy coat4 35100 0 0.00 Buffy coat5 82174 0 0.00 Buffy coat6 164001 0 0.00 Observed droplets Methylation of Sample ID Alb MSC-AS1 MSC-AS1 (%) Buffy coat7 44320 1 0.002 Buffy coat8 49069 0 0.00 Buffy coat9 25621 0 0.00 Buffy coat10 48497 0 0.00 Buffy coat11 47815 0 0.00 Buffy coat12 51857 3 0.006 Buffy coat13 60317 0 0.00 Buffy coat14 63723 0 0.00 Buffy coat15 56384 0 0.00 Buffy coat16 40207 0 0.00 Observed droplets Methylation of Sample ID Alb KCNA3 KCNA3 (%) Buffy coat1 93139 0 0.00 Buffy coat2 473631 0 0.00 Buffy coat3 89761 0 0.00 Buffy coat4 213596 0 0.00 Buffy coat5 148803 0 0.00 Buffy coat6 83607 0 0.00

TABLE-US-00009 TABLE 9 Methylation level of the selected biomarkers in healthy plasma circulating cell free (ccfDNA) measured by ddPCR in triplex assays detecting target methylated sequences of MSC-AS1 and KCNA3 (n = 8) by ddPCR. Methyl- Methyl- ation of Observed ation of Observed droplets MSC-AS1 droplets KCNA3 Sample ID Alb MSC-AS1 (%) KCNA3 (%) Healthy plasma 33 1512 0 0 0 0 Healthy plasma 34 1650 0 0 0 0 Healthy plasma 35 2054 0 0 0 0 Healthy plasma 36 1486 0 0 0 0 Healthy plasma 37 1798 0 0 0 0 Healthy plasma 38 1672 0 0 0 0 Healthy plasma 39 1313 0 0 0 0

TABLE-US-00010 TABLE 10 Methylation level of the selected biomarkers in buffy coat DNA from GC patients measured by ddPCR in triplex assays detecting target methylated sequences of MSC-AS1 and KCNA3 (n = 22). Methyl- Methyl- ation of Observed ation of Observed droplets MSC-AS1 droplets KCNA3 Sample ID Alb MSC-AS1 (%) KCNA3 (%) Buffy Coat 17 6327 0 0 0 0 Buffy Coat 18 6341 0 0 0 0 Buffy Coat 19 6616 0 0 0 0 Buffy Coat 20 6149 0 0 0 0 Buffy Coat 21 6977 0 0 0 0 Buffy Coat 22 2752 0 0 0 0 Buffy Coat 23 1908 0 0 0 0 Buffy Coat 24 4093 0 0 0 0 Buffy Coat 25 6274 0 0 0 0 Buffy Coat 26 4375 0 0 0 0 Buffy Coat 27 4332 0 0 0 0 Buffy Coat 28 5749 0 0 0 0 Buffy Coat 29 6740 3 0.04 0 0 Buffy Coat 30 5498 0 0 2 0.03 Buffy Coat 31 6726 0 0 0 0 Buffy Coat 32 6117 0 0 0 0 Buffy Coat 33 6771 0 0 0 0 Buffy Coat 34 6701 1 0.02 0 0 Buffy Coat 35 6283 0 0 2 0.03 Buffy Coat 36 6618 0 0 0 0 Buffy Coat 37 6821 0 0 0 0 Buffy Coat 38 6746 0 0 0 0

[0168] Monitoring Progression of Gastric Cancer.

[0169] The methylation status of circulating cell free DNA as determined by the use of the markers MSC-A1 and KCNA3 demonstrated to be correlated with progression free survival (FIG. 4). 55 patients in first, second or third line therapy were tested for the presence of methylated tumor DNA by the above described procedure. More than seventy-six percent (78%) were positive (n=43) for at least one marker. The methylation status of the GC patients was associated with higher progression free survival (see FIG. 4). When dividing patients according to the quantity of detected methylated markers (tertiles, low, intermediate and high methylated ctDNA concentration in copy/mL of plasma), significant differences of disease free survival were observed. Indeed, increased probability of progression was observed for the group with highest concentration of methylated ctDNA vs the group with intermediate methylated ctDNA concentration vs the group with low methylated ctDNA concentration (Log-rank trend test, p=0.0092). Pertinence of these markers was further validated using univariate analysis. When analysis of methylation status (both when considering samples as positive vs negative for methylation or when considering the quantity of methylated ctDNA divided in three tertiles) gave significant results for prediction of cancer progression, analysis of potential GC markers (such as gender, Her2 amplification or tumor differenciation grade) did not give significant results. Table 11 summarizes the obtained data for patients treated in first and second line chemotherapy (n=48). Importantly quantity of ctDNA was also associated with progression of GC when taken as a continuous variable (p-value=0.024).

TABLE-US-00011 TABLE 11 Univariate analysis between GC patient ctDNA methylation status or quantity of methylated sequences copies/mL of plasma determined by methylation-based ddPCR analysis targeting MSC-AS1 and KCNA3 (samples were considered positive when detected positive for at least one of the two markers). Analysis of other potential prognosis factors of GC such as amplification of HER2 (Amp_HER2), differentiation grade of the tumor (Grade_Diff) or gender is also included. Summary Characteristic N Statistics.sup.1 HR.sup.2 95% CI.sup.2 p-value ctDNA 48 Neg 11 (23%) — — Pos 37 (77%) 4.66 1.06, 20.4 0.041 Gender 49 F 15 (31%) — — M 34 (69%) 1.81 0.64, 5.06 0.3 Age 49 65 (58, 72) 1.01 0.98, 1.04 0.6 Grade_diff 38 High 8 (21%) — — Low 16 (42%) 0.74 0.24, 2.24 0.6 Moderate 14 (37%) 0.60 0.17, 2.06 0.4 Amp_HER2 39 8 (21%) No — — Yes 1.35 0.48, 3.82 0.6 ctDNA_copy_tertile 48 low 17 (35%) — — intermediate 16 (33%) 3.30 1.01, 10.8 0.048 high 15 (31%) 4.46 1.34, 14.8 0.015 .sup.1Statistics presented: n (%); median (IQR) .sup.2HR = Hazard Ratio, CI = Confidence Interval

BIBLIOGRAPHIC REFERENCES

[0170] Juhling, F., et al., metilene: fast and sensitive calling of differentially methylated regions from bisulfite sequencing data. Genome Res, 2016. 26(2): p. 256-62. [0171] Garrigou, S., et al., A Study of Hypermethylated Circulating Tumor DNA as a Universal Colorectal Cancer Biomarker. Clin Chem, 2016. 62(8): p. 1129-39. [0172] Milbury, C. A., et al., Determining lower limits of detection of digital PCR assays for cancer-related gene mutations. Biomol Detect Quantif, 2014. 1(1): p. 8-22. [0173] Taly, V., et al., Multiplex picodroplet digital PCR to detect KRAS mutations in circulating DNA from the plasma of colorectal cancer patients. Clin Chem, 2013. 59(12): p. 1722-31. [0174] Mikeska T, et al. DNA methylation biomarkers in cancer: progress towards clinical implementation. Expert Rev Mol Diagn. 2012 June; 12(5):473-87. [0175] Jain, Wojdacz & Su, Challenges for the application of DNA methylation biomarkers in molecular diagnostic testing for cancer. Expert Rev Mol Diagn. 2013 April; 13(3):283-94. [0176] Delpu Y et al, DNA methylation and cancer diagnosis. Int J Mol Sci. 2013 Jul. 18; 14(7):15029-58 [0177] Mandard A M, Dalibard F, Mandard J C, et al. Pathologic assessment of tumor regression after preoperative chemoradiotherapy of esophageal carcinoma. Clinicopathologic correlations. Cancer. 1994; 73(11):2680-268 [0178] Pinheiro L B et al, Evaluation of a Droplet Digital Polymerase Chain Reaction Format for DNA Copy Number Quantification. Anal. Chem. 2012, 84, 1003-1011