Discovering methylation markers and development of a sense-antisense and dual-MGB probe PCR assay in plasma for colorectal cancer early detection

Discovering methylation markers and development of a sense-antisense and dual-MGB probe PCR assay in plasma for colorectal cancer early detection | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Discovering methylation markers and development of a sense-antisense and dual-MGB probe PCR assay in plasma for colorectal cancer early detection Yanteng Zhao, Zhijie Wang, Qiuning Yu, Xin Liu, Xue Liu, Shuling Dong, and 4 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-4838443/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Background : Screening for colorectal cancer (CRC) using plasma cell-free DNA (cfDNA) methylation is more challenging than stool testing due to the low abundance of cfDNA. Therefore, the development of signal amplification assays based on appropriate markers is essential to increase sensitivity. Methods : A total of 17 existing 450K microarray datasets including tissue, healthy white blood cell (WBC) and plasma cfDNA data from public databases were used to identify differentially methylated CpGs (DMCs) common to CRC and adenoma. The methylation status of candidate DMCs was confirmed by Sanger sequencing with CRC and normal tissues. A sense-antisense and dual MGB probe (SADMP) assay was then developed. Subsequently, the biomarkers were validated in 712 plasma samples using the SADMP method. Results : A total of 2237 DMCs showed overlap between the cancer vs. normal and adenoma vs. normal groups. Of these, 75 were hypomethylated in 30 other non-CRC cancers. After LASSO regression, this number was reduced to eight. Two of these, NTMT1 and MAP3K14-AS1 , were identified as promising candidate markers following WBC validation and primer/probe design evaluation. The SADMP technology demonstrated the ability to amplify the detection signal to approximately twice the original level. Overall, the dual-target SADMP assay demonstrated a sensitivity of 84.8% for CRC (stage I: 75.0%), a sensitivity of 32.0% for advanced adenomas (AA), and a specificity of 91.5% in controls. Conclusions : The dual-target assay demonstrated high performance for CRC and AA detection in plasma-based tests, suggesting that it may serve as a promising noninvasive tool for CRC detection. Cancer Biology colorectal cancer methylation NTMT1 MAP3K14-AS1 Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Background The majority of early-stage advanced colorectal neoplasms, including colorectal cancer (CRC) and advanced adenoma (AA) are curable, particularly AA, which can be removed at the time of diagnosis by colonoscopy. Consequently, the early detection of CRC and AA represents a highly valuable endeavor. A number of tests based on cell-free DNA (cfDNA) methylation have been developed and have shown good performance for CRC detection. Epi proColon [1] is the first blood-based test utilized for the early detection of CRC and was developed based on methylated cfDNA of Septin9. The updated version of Epi proColon® 2.0 CE demonstrated an improved sensitivity of 74.8%-81% and specificity of 96.3%-99% in a prospective cohort study [2]. The Chinese version of Epi proColon simplified sample processing with a single reaction system in a larger volume (60 ul) instead of a 2/3 algorithm (20 ul for three runs), resulting a sensitivity of 73% and a specificity of 94.5% [3]. However, its sensitivity for detecting polyps or AA was relatively low in most studies, ranging from 8-40% [4-6]. Due to this relatively poor sensitivity, particularly for precancerous lesions, the US Preventive Services Task Force and the American Cancer Society currently do not include the test in their CRC screening guidelines. Moreover, plasma methylated Septin9 is not a CRC specific marker that it showed an ability to detect multiple cancer types, including hepatocellular carcinoma [7], gastric cancer [8],cervical cancer [9], and breast cancer [10]. Therefore, developing a sensitive and specific blood assay for CRC screening is meaningful. Due to the disparate mechanisms by which cancer signals enter the blood and stool, it is probable that the markers are not universal, and the accuracy of CRC markers in the blood is typically low in comparison to those in stool. Consequently, an ab initio search for blood methylation markers for early CRC and AA is imperative. Furthermore, the detection of the methylation signal of cell-free tumor DNA (ctDNA) in plasma is challenging due to the presence of ctDNA fragments released by multiple organs or tissues, as well as DNA damage caused by bisulfite conversion [11]. Therefore, the identification of suitable markers and the development of effective detection techniques are essential to ensure optimal performance [12]. Classical PCR-based methylation detection techniques are often developed based on single-strand bisulfite converted DNA (BS-DNA), leaving the information of the other strand unused [13]. Furthermore, only one Taqman MGB probe is usually designed to provide fluorescent signals. Theoretically, designing primers for both sense and antisense strand DNA simultaneously or using multiple MGB probes will improve the sensitivity of a marker to detect methylation signals by enhancing fluorescent signals. Sarah Ø. Jensen et al [14] made the first attempt to design a pair of primers for both sense and antisense strands, thereby enhancing the performance of three targets for CRC detection. Additionally, the dual-strand technique was also successfully applied to methylated HOXA9 in ovarian cancer to improve the detection sensitivity [15]. In this study, we utilized all publicly available methylation data for CRC and adenomas to identify methylation markers, with a particular focus on those that are hypermethylated in adenomas. Furthermore, in order to enhance the ability of candidate markers to detect low-abundance cfDNA methylation signals in plasma, we attempted to apply dual-strand and dual-MGB probe techniques simultaneously, which we called the sense-antisense and dual-MGB probe (SADMP) technique, to develop a novel CRC plasma test. Finally, the test performance was comprehensively assessed in our recruited training and validation cohorts. Methods Data preparation Thirteen methylation datasets, which were GSE77954, GSE101764, GSE131013, GSE199057, GSE164811, GSE193535, GSE129364, GSE139404, GSE107352, GSE75546, GSE77965, GSE68060 and EMTAB6450 from public databases were selected based on three criteria: 1) were generated by Illumina HumanMethylation 450k BeadChip and had the raw IDAT files, 2) the sample size was greater than 10, and 3) consisted of CRC or adenoma or normal adjacent tissue (NAT). All the IDAT files were then processed using the minfi tool [16] to obtain methylation β values. They were integrated as a single dataset (n= 1165), which we defined as Phase I discovery set for candidate marker identification. Level 3 methylation data for 31 cancer types were retrieved from The Cancer Genome Atlas (TCGA) database (https://portal.gdc.cancer.gov/). Data for 8258 primary tumor tissues corresponding to 31 cancer types and their 710 NATs were retained. The 31 cancer types included ACC (n=79), BLCA (n=412), BRCA (n=778), CESC (n=306), CHOL (n=36), CRC (n=379), DLBC (n=48), ESCA (n=183), GBM (n=137), HNSC (n=523), KICH (n=65), KIRC (n=312), KIRP (n=271), LGG (n=513), LIHC (n=374), LUAD (n=456), LUSC (n=364), MESO (n=87), OV (n=10), PAAD (n=183), PCPG (n=178), PRAD (n=495), SARC (n=257), SKCM (n=104), STAD (n=393), TGCT (n=133), THCA (n=503), THYM (n=124), UCEC (n=418), UCS (n=57), UVM (n=80). The TCGA dataset was defined as Phase II discovery dataset. DMCs were deemed eligible if they exhibited methylation levels =0.55 on CRC, and <0.15 on 710 NATs. The other three datasets, GSE48684 [17], GSE40279 [18], and GSE122126 [19], generated by the same platform of Illumina HumanMethylation BeadChip in Gene Expression Omnibus (GEO) were used as validation sets to verify the methylation status of candidate differentially methylated CpGs (DMCs). The GSE48684 cohort consisted of 105 qualified samples, of which 41 were NAT and 64 were CRC. The GSE40279 cohort comprised whole blood cell (WBC) samples collected from 656 healthy individuals. The GSE122126 cohort included three CRC and 12 normal plasma samples. Sample collection This case-control study enrolled 772 cases, including 60 tissue samples (30 CRC and 30 NATs) and 712 plasma samples from the First Affiliated Hospital of Zhengzhou University between April 2022 and June 2022. Tissue samples were obtained from the preserved formalin-fixed paraffin-embedded (FFPE) sections collected from surgical patients. The plasma cohort was randomly divided into a training set and a validation set in a 2:1 ratio. The training set comprised 474 participants, including 115 healthy blood donors, 123 non-digestive disease patients (NDD), 65 intestinal disease patients (ID), 14 polyps patients, 10 non-advanced adenoma patients (Non-AA), 16 AA, 125 CRC patients, and 6 patients with other cancers. In the validation set, there were 235 participants, including 57 healthy blood donors, 60 NDD, 30 intestinal ID, 6 polyps patients, 5 Non-AA, 9 AA, 66 CRC patients, and 2 patients with other cancers ( Supplemental Table 1 ). This study was approved by the ethics committee. Each participant signed an informed consent form. Patients with polyps, adenomas, CRC and other cancers were further confirmed by histopathological examinations. The included CRC patients were required to meet the following criteria: 1) did not receive any radiotherapy, chemotherapy or surgery, 2) ages older than 18. All CRC patients were classified as I, II, III, and IV stages according to the American Joint Cancer Committee (AJCC) staging system. Differential methylation analysis We performed differential methylation analysis using the rank-sum test on three groups in the discovery set: cancer vs. normal (NAT), adenoma vs. normal, and cancer vs. adenoma. DMCs were defined as significant if they meet two criteria: an FDR < 0.05 and a fold change ranking in the top 1% of all probes. The DMCs were then classified as hyper- or hypo- DMCs based on whether they exhibited high or low methylation levels in the tumor or adenoma samples compared to the normal samples. LASSO regression was implemented in the R package 'glmnet' to reduce the number of features. LASSO regression was repeated 100 times, and the frequencies of probes with non-zero coefficients in the regressions were counted. Tissue and plasma DNA extraction Genomic DNA from tissue samples and plasma cfDNA was isolated using the FFPE DNA rapid extraction kit (TianGen, Beijing) and the plasma/serum cfDNA extraction kit (Ammunition Life-tech, Wuhan), respectively, according to the protocols. Briefly, cfDNA extraction was divided into two steps. First, approximately 10 ml of blood was drawn from participants at room temperature and then centrifuged twice with 1300g and 14000g for 10 min at 4°C within 2 hours. Then, about 2 ml of plasma was retained for cfDNA purification. Purified cfDNA was eluted in 45 ul TE buffer and then treated with bisulfite using a DNA bisulfite modification kit (Ammunition Life-tech, Wuhan) in accordance with the instructions described in the aforementioned study [20], or stored at -80 °C until required. Sanger sequencing Sanger sequencing (ABI 3730XL, GENEWIZ, Suzhou) was performed for the target regions of NTMT1 and MAP3K14-AS1 after bisulfite treatment to confirm their methylation status on 30 CRC tissue samples and 30 NATs. The sequencing primers are shown in Supplemental Table 2 . Following bisulfite treatment, cytosine in methylated CpG sites remained cytosine, while unmethylated cytosine was converted to thymine. Sanger sequencing results thus allowed for the direct determination of the methylation status of target regions. Developing the SADMP and methylation-specific PCR (MSP) technique on plasma The location and sequences of MSP primers/ probes are displayed in Supplemental Table 3 . For NTMT1 , two pairs of methylation primers targeted the sense and antisense strands, and their corresponding MGB probes were designed. For MAP3K14-AS1 , the MGB-probe 1 and MGB-probe 2 were designed according to the antisense strand sequence and the reverse complementary sequence of the antisense strand. Gradient dilutions of synthetic plasmids or plasmid mixtures are employed as templates for the determination of test parameters. Specifically, standard curve experiments were carried out to assess the amplification efficiency of each pair of primers. The experiments consisted of five ten-fold dilutions of fully methylated plasmid DNA templates with five replicates at each dilution (1, 10, 100, 1000, and 10000 copies per run). Primer tolerance experiments were designed to assess the specific amplification ability of methylated primers against methylated templates. Eight concentration dilutions of fully methylated DNA templates (0, 1, 5, 10, 50, 100, 200, and 400 copies/μL) mixed with a high concentration of unmethylated plasmid (10 7 copies/μL) were prepared to evaluate the ability of methylated primers selectively amplifying methylated templates under the background of unmethylated templates. Each concentration experiment was repeated five times. The amplification system of MSP was shown in Supplemental Table S4 . The test employed a triplex PCR with a FAM channel for the internal reference gene ACTB , a ROX channel for NTMT1 (dual-strands), and VIC channel for MAP3K14-AS1 (dual-MGB probes). The PCR procedure was pre-denaturation at 95°C for 5 min (step 1), denaturation at 95°C for 15s (step 2), annealing at 60°C for 30s (step 3), repeating step 2 and 3 forty-five times in plasmid samples and fifity times in plasma samples on the 7500 Fast Real-Time PCR System (Applied Biosystems, USA), the cycle threshold (Ct) was assigned to 50 for the target without amplification. Statistical analysis The data processing and analysis in this study were conducted using R software (version 4.1.0). The ‘glm’ function was employed to fit the logistic regression model with a parameter of ‘family=binomial’. Receiver operating characteristic (ROC) curve analysis was performed using the ‘pROC’ package, and the area under the ROC curve (AUC) was subsequently calculated to assess the test classification performance. The optimal sensitivity and specificity of the test were estimated when Youden’s index reached its maximal value. Sensitivity and specificity were calculated using the following formulas: And the Youden index = sensitivity+specificity-1. The rank-sum test and Kruskal test were employed for the comparisons between two groups and the comparisons between multiple groups, respectively. The Chi-square test was utilized for the comparisons between categorical variables. The Other statistical methods used in this study were described in the corresponding results. Results Study design and participant characteristics The study is divided into four steps, as illustrated in Fig. 1 . In the initial step, candidate markers were identified in a set of 1165 samples from 13 datasets. Of these, 452 were normal, 168 were adenomas, and 545 were CRC. These markers were then narrowed down using data from 31 cancer types in the TCGA dataset, DMCs that were hypermethylated in cancer other than CRC were filtered out. The available DMCs were further validated in three independent datesets. In the second step, the methylation status of candidate CpGs was confirmed by Sanger sequencing using 30 CRC tissues and 30 NATs. In the third step, the SADMP technique and MSP system were established. This involved designing appropriate primers, optimizing the qPCR amplification system, and evaluating the technical parameters of the assay. In the last step, the developed assay was tested on training and validation sets of plasma. Its performance was assessed by estimating indicators such as sensitivity, specificity, and AUC. Landscape of the methylation patterns of the 13 discovery datasets The methylation levels of adenoma and cancer samples were found to be lower than normal, as shown in Fig. 2 A. The methylation density curves for the three groups displayed bimodal distributions with hyper- and hypomethylation peaks, respectively (Fig. 2 B). The hyper-methylation peaks of adenoma and tumor samples were lower than those of normal samples, indicating higher methylation levels in normal samples. We employed t-SEN to analyze and visualize the structure of the discovery set and identified significant differences between normal and cancer samples. However, adenomas exhibited overlap with both normal and cancer samples (Fig. 2 C). We then selected the top 1% of most variable probes to cluster the discovery set using the K-means algorithm. The results showed that both normal and cancer samples clustered together, while adenomas were separated into two subgroups with a remarkable difference. They showed both high (Methy-H) and low (Methy-L) methylation status and were closer to cancer and normal samples, respectively (Fig. 2 D). Further analysis revealed that tubular adenomas constituted the largest proportion of Methy-L adenomas (23/57 or 40.35%), while villous adenomas constituted the largest proportion of Methy-H adenomas (47/73 or 64.38%). To eliminate any potential bias between the datasets, we conducted Fisher’s test separately for the Methy-H and Methy-L adenomas, with the datasets serving as the stratification factor. However, we did not identify any significant differences (P > 0.05). Identification of candidate DMCs First, we conducted a comparative analysis of the DMCs between cancer, adenoma, and normal samples using the 13 discovery datasets. The three comparisons (cancer vs. normal, adenoma vs. normal, and cancer vs. adenoma) yielded 3000, 3051, and 1545 DMCs, respectively. Notably, these DMCs were predominantly hyper-DMCs (Fig. 3 A). Further investigations revealed that the majority of DMCs were distributed in upstream regions of genes (5'UTR, TSS1500, TSS500, and 1st Exon), except for cancer vs. adenoma where DMCs were mainly located in gene bodies (Fig. 3 B). We observed an extremely high proportion of overlapping DMCs between cancer vs. normal and adenoma vs. normal, which was significantly higher than cancer vs. adenoma (Fig. 3 C). To identify the most appropriate DMCs, we focused on those that were overlapped between cancer vs. normal and adenoma vs. normal (2237 probes) and evaluated their methylation levels on 31 cancer types of TCGA. This yielded 75 accessible probes (Fig. 3 D). LASSO regression was employed to reduce the number of DMCs, with eight presented in 100 replicates with non-zero coefficients (Fig. 3 E). The methylation levels of NTMT1 and MAP3K14-AS1 in validation sets The methylation levels of the eight probes were then analyzed in 656 WBC samples, and seven DMCs exhibited relatively low methylation levels (less than 0.1) with the exception of cg17892556 (Fig. 4 A). Therefore, the seven DMCs are recognized as the most promising markers. Two specific DNA methylation sites, cg14015706 and cg08247376, were selected for further analysis as they were suitable for primer/probe design. These sites are located in the first exon of NTMT1 and 200 bp upstream of the MAP3K14-AS1 transcription start site, respectively. The GSE48684 dataset revealed significantly higher methylation levels for both probes in cancer samples than in normal samples (Fig. 4 B). Furthermore, the two probes demonstrated hypermethylation in CRC plasma and hypomethylation in healthy plasma (Fig. 4 C), indicating a high consistency of the methylation status between tissue and plasma. The SADMP technique The Sanger sequencing of the target regions revealed that the CpGs sites among the amplicons were more frequently methylated in CRC than NATs, which provided the foundation for the development of SADMP technique. For NTMT1 , two pairs of methylation primers targeted the sense and antisense strands, and their corresponding MGB probes were designed (Fig. 5 A). For MAP3K14-AS1 , the MGB-probe1 and MGB-porbe2 were designed according to the antisense strand sequence and the reverse complementary sequence of antisense strand (Fig. 5 F). The estimated amplification efficiencies of MTNT1 and MAP3K14-AS1 were 98.14% and 110.08%, respectively (Fig. 5 B and 5 G). The sense and anti-sense strands were transformed into completely different DNA sequences after bisulfite treatment and both could be used as PCR templates. It can be postulated that the copy numbers detected by any single pair of primers or single MGB-probe assay should be half of the diluted copy numbers (including sense and anti-sense strands), and that the SADMP assay should be equal to the diluted copy numbers. For NTMT1 , the detected copy numbers by sense and antisense strand assays were slightly less than half of the theoretical copy numbers with slopes of 0.39 and 0.31, respectively. The detected copy numbers by dual-strand assay were approximately 0.73 times of theoretical copy numbers, which was approximately two-fold compared to any single-strand assay (Fig. 5 C), as expected. Meanwhile, Ct value of dual-strand assay was almost one cycle earlier than that of any single-strand assay (average △Ct = 1.21, Fig. 5 D), which was consistent with the theoretical △Ct value (should be one). The MAP3K14-AS1 SADMP showed similar results, with the number of copies detected was doubled compared to the single probe assay (Fig. 5 H), and the Ct values were shifted forward by 0.99 cycles (Fig. 5 I). Additionally, in the serially diluted experiment, no amplification cures for both targets were observed when a high proportion of unmethylated DNA was used as templates (the methylated DNA copies were 0) (Fig. 5 E and 5 G). These results suggested that the developed assays were explicitly targeted for methylated DNAs even at the high background of unmethylated DNAs. We also found that the NTMT1 SADMP assay was able to robustly detect methylated DNAs at a concentration of 10 copies/µL (Fig. 5 E), while it was 5 copies/µL for the MAP3K14-AS1 SADMP ( Fig. 5 G). Validation and evaluation of biomarkers using MSP in plasma samples In the plasma sample set, the Ct values of ACTB exhibited a slight decrease in trend from the healthy control groups to the cancer group (p < 0.0001) (Fig. 6 A). Ct values of NTMT1 and MAP3K14-AS1 were much lower in CRC samples and AA samples than other control groups (Fig. 6 B and 6 C). The Ct values could discriminate the CRC and AA samples from healthy samples very well. Furthermore, the Ct values-based result determination exhibited favorable outcomes in a previous study [ 21 , 22 ]. Consequently, the Ct values were utilized as the result analysis indicators. In order to determine the algorithm and cutoff, ROC curve analysis was conducted in the training set. In the CRC vs control groups, the AUC values for NTMT1 and MAP3K14-AS1 were 0.882 (95% CI: 0.838–0.925) and 0.785 (95% CI: 0.729–0.840), respectively (Fig. 6 D). In the CRC + AA vs control groups, the AUC values for NTMT1 and MAP3K14-AS1 were 0.852 (95% CI: 0.807–0.898) and 0.761 (95% CI: 0.707–0.815), respectively (Fig. 6 E). As AA was precancerous lesions, it was also benefit patients from detecting it. So we use the CRC + AA vs control groups of training set for further analysis. The performance of the combination of the two biomarker were evaluated by two strategies (Table 1 ). Strategy 1 involved the construction of a logistic regression model, with an estimated AUC value was 0.881 (95% CI: 0.840–0.922), the optimal sensitivity for CRC and specificity of 86.4% and 91.0%, respectively (Table 1 ). Strategy 2 was the 1/2 algorithm, which determined a positive measurement when the Ct of any single marker was less than its corresponding threshold. The cutoff was determined as Ct value corresponding to the maximum Youden index in the training dataset using CRC + AA vs control data. The cutoff values were 48.2 and 48.4 for NTMT1 and MAP3K14-AS1 , respectively. At these cutoff, the two target sensitivities were 80.0% and 59.2%, with specificities of 92.8% and 97.0%, respectively (Fig. 6 F). The sensitivity and specificity of the combination of the two biomarkers were 86.4% and 91.0% respectively (Fig. 6 F). Notably, strategy 2 yielded an identical sensitivity and specificity to strategy 1. Given that strategy 2 is considerably more straightforward for physicians to interpret test results in clinical practice, we have adopted the 1/2 algorithm as the combination algorithm for the dual-target test. Table 1 Diagnostic performance of single biomarker and marker combination in training set Biomarker & combination AUC (95% CI) Sensitivity (CRC) Sensitivity (CRC + AA) Specificity Cutoff Combination method NTMT1 0.852 (0.807–0.898) 80.0% 74.5% 92.8% 48.2 / MAP3K14-AS1 0.761 (0.707–0.815) 59.2% 54.6% 97.0% 48.4 / NTMT1 & MAP3K14-AS1 0.881 (0.840–0.922) 86.4% 80.8% 91.0% 0.1491 Logistic regression NTMT1 & MAP3K14-AS1 / 86.4% 80.8% 91.0% 48.2, 48.4 1/2 algorithm After the algorithm and cutoff was set, the validation set and all sample set were analyzed. The validation set had a 81.8% sensitivity and 92.5% specificity (Fig. 6 G) and the all sample set had a 84.8% sensitivity and 91.5% specificity (Fig. 6 H). The dual-target test had a higher sensitivity than that of any single biomarker and a slightly lower specificity in the three data sets (Fig. 6 F-H). It indicated that the combination of the two biomarkers were better than single biomarker. Performance in subgroups of plasma samples The performance of the dual-target test was evaluated in subgroups of all plasma samples. The test demonstrated a detection rate of 26.7% in non-AA samples and a detection rate of 32.0% in AA samples (Fig. 7 A). It exhibited sensitivities of 75.0%, 81.2% and 90.3% in stages I, II and III-IV of CRC samples, respectively (Fig. 7 B). The test exhibited specificities of 95.9%, 92.3%, 84.2% and 90.9% in healthy, NDD, ID and polyps samples, respectively (Fig. 7 C). The positive predictive value (PPV), negative predictive value (NPV) and accuracy for the dual-target test were 79.4%, 94.0% and 89.6%, respectively (Fig. 7 D). With regard to the detection rates in adenoma and different stages of CRC, the dual-target test demonstrated superior performance compared to that of a single biomarker (Fig. 7 A-B). With respect to specificities in subgroups, the dual-target test exhibited a marginal decline in value compared to that of a single biomarker (Fig. 7 C). However, with regard to NPV and accuracy, the dual-target test demonstrated superior performance compared to that of a single biomarker (Fig. 7 D). With regard to PPV, the dual-target test exhibited a slight decline in performance relative to that of a single biomarker (Fig. 7 D). Discussion It is commonly accepted that patients diagnosed with CRC at an early stage can be treated more effectively and have a better prognosis. Several stool DNA-based tests have been developed that demonstrate excellent performance in detecting CRCs at their early stages [ 23 – 26 ]. Blood sampling is more acceptable than stool sampling, but blood-based tests are less reported and usually exhibit lower sensitivities than stool-DNA tests, ranging from 47–87% [ 27 ]. This study presented a systematic pipeline for the discovery of methylation markers from scratch, test development, and evaluation in training and validation plasma sets. The SADMP technique enhanced the ability to detect methylation signals. Following a comprehensive evaluation, the test obtained an overall sensitivity and specificity of 84.8% and 91.5%, respectively, for the detection of CRC at a volume of 2 ml plasma. Adenoma and CRC displayed lower methylation levels overall, with the exception of regulatory regions (Fig. 2 B and 3 B), a finding that has been reported in previous study [ 17 ]. Previous studies have focused on the CpG Island Methylator Phenotype (CIMP) found in CRC. This study found that tubular adenomas are common in the methy-L subclass, while villous adenomas are more often in methy-H subclass. Previous studies have indicated that CIMP is rarely found in tubular adenomas, but frequently in tubulovillous and villous adenomas [ 28 ], which is in accordance with this study. The large proportion of overlapping DMCs between cancer vs normal and adenoma vs normal indicated that many CpGs have undergone aberrant methylation events at the precancerous stage, which provides robust evidence for discovering the methylation markers for CRC early detection. It should be noted that the CRC markers are not necessarily applicable to adenomas, as a significant proportion of the DMCs in the Cancer vs Normal group were not present in the Adenoma vs Normal group, as illustrated in Fig. 3 C. This emphasises the importance of this study's inclusion of adenoma samples in the marker discovery set and the selection of DMCs in which cancers and adenomas overlap in a way that many other studies of CRC markers have not done. Blood-based tests are more susceptible to interfering diseases and may result in a high false-positive rate. Therefore, specificity is a critical indicator. In the marker discovery step, 31 cancer types in the TCGA database and WBC were used to control the low methylation levels of candidate markers in other tissues and blood background, effectively attenuating the interference of unintended cfDNAs. In the assay development phase, we designed highly selective MSP primers that did not show normal amplification curves even when unmethylated DNAs were used as templates at 10 7 copies (Fig. 5 E and 5 J). These aforementioned measures guarantee a high specificity in plasma samples. Two combination algorithms were utilized to assess the performance of the dual-target test in the training set, and both algorithms indicated that the combined markers had better AUC values and higher sensitivities than any single marker (Fig. 6 D and 6 E). However, the dual-target test showed a decreased specificity compared to both single markers, from 92.8% and 97.0–91.0% (Fig. 6 H), which was also observed in other studies [ 29 , 30 ]. When healthy individuals were selected as controls, the specificity improved to 95.9% (Fig. 7 C), which is comparable to the SEPT9 test [ 2 ]. These data demonstrate the excellent performance of the dual-target test in detecting CRC. A number of studies [ 31 – 35 ] have demonstrated that the methylation levels of NTMT1 (whose antisense chain counterpart is C9orf50 ) and MAP3K14-AS1 can be employed for the screening and diagnosis of CRC. In particular, the study conducted by Sarah Ø Jensen et al [ 34 ]. indicated that the C9orf50 methylation assay exhibited a plasma sensitivity of 76% and specificity of 91% for CRC. Ludovic Barault and colleagues [ 35 ] demonstrated that the MAP3K14-AS1 methylation assay in plasma exhibited a sensitivity of 69.8% and a specificity of 100% for CRC. The dual-strand technique has been proven to enhance the performance of markers in previous studies [ 14 , 15 ]. This technique was also observed to be effective in our study. By detecting the methylation signals of NTMT1 sense- and antisense-strand simultaneously, the Ct value of the dual-strand assay was able to shift forward by one compared to the single-strand assay (Fig. 5 D). In contrast to previous studies, the current study also included two MGB probes located downstream of forward and reverse primers of MAP3K14-AS1. During PCR strand extension, the polymerase enzymes cleaved the 5-primer sequence of probes and released two fluorescent groups. The dual-MGB probe technique theoretically doubled the fluorescent signals when both probes shared the same channel, leading to an earlier Ct value similar to that of the dual-strand technique (Fig. 5 I). Serial dilution experiments confirmed the superiority of dual-MGB probes over one MGB probe (Fig. 5 H-J). These results suggest that applying the SADMP technique can be a feasible strategy to enhance the detection sensitivity of candidate markers. Early diagnosis or screening techniques are essential to improve patient survival time, particularly when curable treatments are available. Studies have shown that the 5-year survival rate of early detected CRC is almost 90%, while it was only 20% for advanced CRC [ 36 ]. The dual-target test showed a sensitivity of 75.0% and 81.2% for stage I and stage II CRC detection, notably, the dual-target test obtained a positive detection rate of 32.00% (8/25) for AA (Fig. 7 B and 7 C), implying its ability to detect early CRC and precancerous lesions. The current study has some limitations that may hamper the interpretation of these results. 1) Participants in this study were enrolled from a single center, which may bias these results. 2) the SADMP techniques may not applicable for all candidate markers. The dual-strand technique may be attempted when both sense and antisense strands are suitable for designing MSP primers, while the multiple MGB probe technique is limited by the amplicon length, which is usually less than 100 bp. Conclusion In this study, we employed several public databases of adenomas and CRC for marker discovering, and ultimately identified two promising markers, NTMT1 and MAP3K14-AS1 . We then constructed the SADMP technology based on these two markers, which enhanced the sensitivity of the detection. The dual-target assay has a high sensitivity for AA and early stage CRC, and its clinical application value merits further investigation. Abbreviations AA Advanced adenoma CRC Colorectal cancer cfDNA Cell-free DNA SADMP Sense-antisense and dual-MGB probe NAT Normal adjacent tissue TCGA The Cancer Genome Atlas GEO Gene Expression Omnibus DMC Differentially methylated CpGs WBC Whole blood cell (WBC) FFPE Formalin-fixed paraffin-embedded NDD Non-digestive disease patients ID Intestinal disease Non-AA Non-advanced adenoma MSP Methylation-specific PCR ROC Receiver operating characteristic AUC Area under the ROC curve PPV Positive predictive value NPV Negative predictive value Declarations Ethics approval and consent to participate This study was approved by the ethics committee of the First Affiliated Hospital of Zhengzhou University (approval number: 2022-KY-0631-002). Each participant signed an informed consent form. Clinical Trial Registration This study is a sub-project of the Clinical Study of Pan-cancer DNA Methylation Test in plasma (Clinical Trials ID: NCT05685524) Availability of data and materials The datasets (GSE77954, GSE101764, GSE131013, GSE164811, GSE193535, GSE129364, GSE139404, GSE107352, GSE75546, GSE77965, GSE199057, GSE68060, GSE48684, GSE40279 and GSE122126) supporting the conclusions of this article are available in the Gene Expression Omnibus database (https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi). The E-MTAB-6450 data is available in EMBL biostudies (https://www.ebi.ac.uk/biostudies/arrayexpress). The source code and intermediate data can be accessed from git-hub (https://github.com/amsinfor/CRC-methylation). Competing interests The authors declare that they have no competing interests. Funding This study was supported by science and technology project of Henan Province (232102310028). Authors' contributions QK Y and YT Z conceptualized and designed the study. YT Z，ZJ W and DH Z developed the SADMP technique. ZJ W QN Y, and X L provided clinical samples and patient information. X L and SL D conducted cfDNA extraction and bisulfite converted. YT Z, ZJ W and T Z performed data analysis and prepared the manuscript. XP L validated the study results. All the authors reviewed the manuscript. References Powrózek T, Krawczyk P, Kucharczyk T, Milanowski J. Septin 9 promoter region methylation in free circulating dna—potential role in noninvasive diagnosis of lung cancer: preliminary report. Med Oncol 2014;31: 917. doi: 10.1007/s12032-014-0917-4 Lamb YN, Dhillon S. Epi procolon® 2.0 ce: a blood-based screening test for colorectal cancer. Mol Diagn Ther 2017;21: 225-32. doi: 10.1007/s40291-017-0259-y Sun J, Fei F, Zhang M, Li Y, Zhang X, Zhu S, et al. The role of msept9 in screening, diagnosis, and recurrence monitoring of colorectal cancer. Bmc Cancer 2019; 19: 450. doi: 10.1186/s12885-019-5663-8 Jung G, Hernández-Illán E, Moreira L, Balaguer F, Goel A. Epigenetics of colorectal cancer: biomarker and therapeutic potential. Nat Rev Gastroenterol Hepatol 2020;17(2):111-130. doi: 10.1038/s41575-019-0230-y. Church TR, Wandell M, Lofton-Day C, Mongin SJ, Burger M, Payne SR, et al. Prospective evaluation of methylated SEPT9 in plasma for detection of asymptomatic colorectal cancer. Gut 2014; 63(2):317-25. doi: 10.1136/gutjnl-2012-304149. Nassar FJ, Msheik ZS, Nasr RR, Temraz SN. Methylated circulating tumor DNA as a biomarker for colorectal cancer diagnosis, prognosis, and prediction. Clin Epigenetics 2021;13(1):111. doi: 10.1186/s13148-021-01095-5. Li B, Huang H, Huang R, Zhang W, Zhou G, Wu Z, et al. Sept9 gene methylation as a noninvasive marker for hepatocellular carcinoma. Dis Markers 2020; 2020: 6289063. doi: 10.1155/2020/6289063 Cao CQ, Chang L, Wu Q. Circulating methylated septin 9 and ring finger protein 180 for noninvasive diagnosis of early gastric cancer. Transl Cancer Res 2020; 9: 7012-21. doi: 10.21037/tcr-20-1330 Jiao X, Zhang S, Jiao J, Zhang T, Qu W, Muloye GM, et al. Promoter methylation of sept9 as a potential biomarker for early detection of cervical cancer and its overexpression predicts radioresistance. Clin Epigenetics 2019;11: 120. doi: 10.1186/s13148-019-0719-9 Matsui S, Kagara N, Mishima C, Naoi Y, Shimoda M, Shimomura A, et al. Methylation of the sept9_v2 promoter as a novel marker for the detection of circulating tumor dna in breast cancer patients. Oncol Rep 2016; 36: 2225-35. doi: 10.3892/or.2016.5004 Song P, Wu LR, Yan YH, Zhang JX, Chu T, Kwong LN,et al. Limitations and opportunities of technologies for the analysis of cell-free DNA in cancer diagnostics. Nat Biomed Eng 2022;6(3):232-245. doi: 10.1038/s41551-021-00837-3. Babayan A, Pantel K. Advances in liquid biopsy approaches for early detection and monitoring of cancer. Genome Med 2018; 10: 21. doi: 10.1186/s13073-018-0533-6 Li LC, Dahiya R. Methprimer: designing primers for methylation pcrs. Bioinformatics 2002;18: 1427-31. doi: 10.1093/bioinformatics/18.11.1427 Jensen SØ, Øgaard N, Nielsen HJ, Bramsen JB, Andersen CL. Enhanced performance of dna methylation markers by simultaneous measurement of sense and antisense dna strands after cytosine conversion. Clin Chem 2020; 66: 925-33. doi: 10.1093/clinchem/hvaa100 Faaborg L, Fredslund AR, Waldstrøm M, Høgdall E, Høgdall C, Adimi P, et al. Analysis of hoxa9 methylated ctdna in ovarian cancer using sense-antisense measurement. Clin Chim Acta 2021; 522: 152-7. doi: 10.1016/j.cca.2021.08.020 Aryee MJ, Jaffe AE, Corrada-Bravo H, Ladd-Acosta C, Feinberg AP, Hansen KD, et al. Minfi: a flexible and comprehensive bioconductor package for the analysis of infinium dna methylation microarrays. Bioinformatics 2014; 30: 1363-9. doi: 10.1093/bioinformatics/btu049 Luo Y, Wong CJ, Kaz AM, Dzieciatkowski S, Carter KT, Morris SM, et al. Differences in dna methylation signatures reveal multiple pathways of progression from adenoma to colorectal cancer. Gastroenterology 2014; 147: 418-29. doi: 10.1053/j.gastro.2014.04.039 Hannum G, Guinney J, Zhao L, Zhang L, Hughes G, Sadda S, et al. Genome-wide methylation profiles reveal quantitative views of human aging rates. Mol Cell 2013; 49: 359-67. doi: 10.1016/j.molcel.2012.10.016 Moss J, Magenheim J, Neiman D, Zemmour H, Loyfer N, Korach A, et al. Comprehensive human cell-type methylation atlas reveals origins of circulating cell-free dna in health and disease. Nat Commun 2018; 9: 5068. doi: 10.1038/s41467-018-07466-6 Li R, Qu B, Wan K, Lu C, Li T, Zhou F, et al. Identification of two methylated fragments of an sdc2 cpg island using a sliding window technique for early detection of colorectal cancer. Febs Open Bio . (2021) 11: 1941-52. doi: 10.1002/2211-5463.13180 Zhang L, Dong L, Lu C, Huang W, Yang C, Wang Q, et al. Methylation of SDC2/TFPI2 and Its Diagnostic Value in Colorectal Tumorous Lesions. Front Mol Biosci. 2021;8:706754. doi: 10.3389/fmolb.2021.706754. Bian Y, Gao Y, Lu C, Tian B, Xin L, Lin H, et al. Genome-wide methylation profiling identified methylated KCNA3 and OTOP2 as promising diagnostic markers for esophageal squamous cell carcinoma. Chin Med J (Engl) 2023; doi: 10.1097/CM9.0000000000002832. Wang Z, Shang J, Zhang G, Kong L, Zhang F, Guo Y, et al. Evaluating the clinical performance of a dual-target stool dna test for colorectal cancer detection. J Mol Diagn 2022; 24: 131-43. doi: 10.1016/j.jmoldx.2021.10.012 Imperiale TF, Ransohoff DF, Itzkowitz SH, Levin TR, Lavin P, Lidgard GP, et al. Multitarget stool dna testing for colorectal-cancer screening. N Engl J Med 2014;370: 1287-97. doi: 10.1056/NEJMoa1311194 Zhao G, Liu X, Liu Y, Li H, Ma Y, Li S, et al. Aberrant dna methylation of sept9 and sdc2 in stool specimens as an integrated biomarker for colorectal cancer early detection. Front Genet 2020; 11: 643. doi: 10.3389/fgene.2020.00643 Oh TJ, Oh HI, Seo YY, Jeong D, Kim C, Kang HW, et al. Feasibility of quantifying sdc2methylation in stool dna for early detection of colorectal cancer. Clin Epigenetics 2017; 9: 126. doi: 10.1186/s13148-017-0426-3 Nassar FJ, Msheik ZS, Nasr RR, Temraz SN. Methylated circulating tumor dna as a biomarker for colorectal cancer diagnosis, prognosis, and prediction. Clin Epigenetics 2021; 13: 111. doi: 10.1186/s13148-021-01095-5 Kakar S, Deng G, Cun L, Sahai V, Kim YS. Cpg island methylation is frequently present in tubulovillous and villous adenomas and correlates with size, site, and villous component. Hum Pathol 2008; 39: 30-6. doi: 10.1016/j.humpath.2007.06.002 Zhao G, Li H, Yang Z, Wang Z, Xu M, Xiong S, et al. Multiplex methylated dna testing in plasma with high sensitivity and specificity for colorectal cancer screening. Cancer Med 2019; 8: 5619-28. doi: 10.1002/cam4.2475 Bagheri H, Mosallaei M, Bagherpour B, Khosravi S, Salehi AR, Salehi R. Tfpi2 and ndrg4 gene promoter methylation analysis in peripheral blood mononuclear cells are novel epigenetic noninvasive biomarkers for colorectal cancer diagnosis. J Gene Med 2020;22: e3189. doi: 10.1002/jgm.3189 Cao Y, Zhao G, Yuan M, Liu X, Ma Y, Cao Y, et al. KCNQ5 and C9orf50 Methylation in Stool DNA for Early Detection of Colorectal Cancer. Front Oncol 2021; 29;10:621295. doi: 10.3389/fonc.2020.621295. Zhang Y, Wu Q, Xu L, Wang H, Liu X, Li S, et al. Sensitive detection of colorectal cancer in peripheral blood by a novel methylation assay. Clin Epigenetics 2021;13(1):90. doi: 10.1186/s13148-021-01076-8. Jensen SØ, Øgaard N, Ørntoft MW, Rasmussen MH, Bramsen JB, Kristensen H, et al. Novel DNA methylation biomarkers show high sensitivity and specificity for blood-based detection of colorectal cancer-a clinical biomarker discovery and validation study. Clin Epigenetics 2019;11(1):158. doi: 10.1186/s13148-019-0757-3. Barault L, Amatu A, Siravegna G, Ponzetti A, Moran S, Cassingena A, et al. Discovery of methylated circulating DNA biomarkers for comprehensive non-invasive monitoring of treatment response in metastatic colorectal cancer. Gut 2018; 67(11):1995-2005. doi: 10.1136/gutjnl-2016-313372. Huang H, Cao W, Long Z, Kuang L, Li X, Feng Y, et al. DNA methylation-based patterns for early diagnostic prediction and prognostic evaluation in colorectal cancer patients with high tumor mutation burden. Front Oncol 2023;12:1030335. doi: 10.3389/fonc.2022.1030335. Ladabaum U, Dominitz JA, Kahi C, Schoen RE. Strategies for colorectal cancer screening. Gastroenterology 2020; 158: 418-32. doi: 10.1053/j.gastro.2019.06.043 Additional Declarations The authors declare no competing interests. Supplementary Files SupplementalTablesRSpreprint.xlsx Supplementary materials Supplemental Table 1. The clinical features of training and validation cohorts used in this study Supplemental Table 2. The information of primers for Sanger sequencing in this study Supplemental Table 3. The information of primers and MGB probes used in this study Supplemental Table 4. The amplification system of the MSP Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-4838443","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":334978130,"identity":"c0246662-e11f-4ad5-bb3c-4adfd72d7ebc","order_by":0,"name":"Yanteng Zhao","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA8ElEQVRIiWNgGAWjYJCCAx8qbHjYmJkPPvhgYCNHhAZmxoMzzqTJ8LO3JRvOKEgzJkYL82HetsM2kj1n1IR5PhxOJKhBfkb+gcM8Zw7zGNzIYWO2MWBOYGA/fHQDPi2MM5IZDs6pSAdqyT32OMeALY+BJy3tBl5nSSQzHHhzxhqoJS/dOMeAp5hBgscMrxY2kBbeNmaQw8ykLQwkEhsIaeEBajnI2+bMA/S+mTSDgQFhLRI8jw1AgcwDDuQegwRjNkJ+kW9PfPwBGJX24Kj88ee/HD/74WN4tTAIJKD7Dq9yEOA/QFDJKBgFo2AUjHQAAFbQTKR1tf7EAAAAAElFTkSuQmCC","orcid":"","institution":"Department of Transfusion, The First Affiliated Hospital of Zhengzhou University, Zhengzhou 450052, China","correspondingAuthor":true,"prefix":"","firstName":"Yanteng","middleName":"","lastName":"Zhao","suffix":""},{"id":334979910,"identity":"2e623535-227a-4c00-978d-73343039ef6a","order_by":1,"name":"Zhijie Wang","email":"","orcid":"","institution":"Department of Gastroenterology, Affiliated Hangzhou First People's Hospital, West lake University School of Medicine, Hangzhou 310003, China.","correspondingAuthor":false,"prefix":"","firstName":"Zhijie","middleName":"","lastName":"Wang","suffix":""},{"id":334979911,"identity":"217be15f-8e0c-4ec8-8458-6d9fa648e12a","order_by":2,"name":"Qiuning Yu","email":"","orcid":"","institution":"Otorhinolaryngology Hospital, The First Affiliated Hospital of Zhengzhou University, Zhengzhou 450052, China","correspondingAuthor":false,"prefix":"","firstName":"Qiuning","middleName":"","lastName":"Yu","suffix":""},{"id":334979912,"identity":"862297b8-d9b3-4b4f-96c0-21f0a4d5b790","order_by":3,"name":"Xin Liu","email":"","orcid":"","institution":"Department of Transfusion, The First Affiliated Hospital of Zhengzhou University, Zhengzhou 450052, China","correspondingAuthor":false,"prefix":"","firstName":"Xin","middleName":"","lastName":"Liu","suffix":""},{"id":334979913,"identity":"e3d1f38c-3836-4339-b8e3-bf47da67a6e3","order_by":4,"name":"Xue Liu","email":"","orcid":"","institution":"Department of Transfusion, The First Affiliated Hospital of Zhengzhou University, Zhengzhou 450052, China","correspondingAuthor":false,"prefix":"","firstName":"Xue","middleName":"","lastName":"Liu","suffix":""},{"id":334979914,"identity":"75119747-8743-45a6-8f45-89f7fc506c96","order_by":5,"name":"Shuling Dong","email":"","orcid":"","institution":"Department of Transfusion, The First Affiliated Hospital of Zhengzhou University, Zhengzhou 450052, China","correspondingAuthor":false,"prefix":"","firstName":"Shuling","middleName":"","lastName":"Dong","suffix":""},{"id":334979915,"identity":"3cae2939-2005-4da1-a8b9-5d8ab2b92e0e","order_by":6,"name":"Xianping Lv","email":"","orcid":"","institution":"Department of Transfusion, The First Affiliated Hospital of Zhengzhou University, Zhengzhou 450052, China","correspondingAuthor":false,"prefix":"","firstName":"Xianping","middleName":"","lastName":"Lv","suffix":""},{"id":334979916,"identity":"965d69ac-3085-44b6-8b29-bb7bef11d3a9","order_by":7,"name":"Tiao Zhang","email":"","orcid":"","institution":"Chongqing International Travel Healthcare Center (Chongqing Customs Port Outpatient), Chongqing 401147, China","correspondingAuthor":false,"prefix":"","firstName":"Tiao","middleName":"","lastName":"Zhang","suffix":""},{"id":334979917,"identity":"10dd0a17-7354-4c48-8d11-0a4f2b6ecf15","order_by":8,"name":"Dihan Zhou","email":"","orcid":"","institution":"Wuhan Ammunition Life-tech Company, Ltd., Wuhan 430200, China","correspondingAuthor":false,"prefix":"","firstName":"Dihan","middleName":"","lastName":"Zhou","suffix":""},{"id":334979918,"identity":"5ba8fd79-6b56-4ea4-8474-738f85eb143c","order_by":9,"name":"Qiankun Yang","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAwklEQVRIiWNgGAWjYFACxgYGhgMMDPzMzAcfkKZFsp0t2YAEm4BaDM7zmAkQpdjgeHLj54Izh/OMDzOYMTDU2EQT1nLmYbP0jBuHi80OM6Q9YDiWlttAUMuNxAZpng+HE7cdZjhuwNhwmCgtzb9BWjY3M7ZJEKulTZrnxuHEDczMbMRpkTzzsM2a50x64ozDbMwGCcT4he94+uPbPMesE/v7z3988KHGhrAWhQMJSLwEHKpQgHwDUcpGwSgYBaNgRAMAh9pG7WnubbcAAAAASUVORK5CYII=","orcid":"","institution":"Department of Transfusion, The First Affiliated Hospital of Zhengzhou University, Zhengzhou 450052, China","correspondingAuthor":true,"prefix":"","firstName":"Qiankun","middleName":"","lastName":"Yang","suffix":""}],"badges":[],"createdAt":"2024-08-01 01:06:20","currentVersionCode":1,"declarations":{"humanSubjects":true,"vertebrateSubjects":false,"conflictsOfInterestStatement":false,"humanSubjectEthicalGuidelines":true,"humanSubjectConsent":true,"humanSubjectClinicalTrial":true,"humanSubjectCaseReport":false,"vertebrateSubjectEthicalGuidelines":false},"doi":"10.21203/rs.3.rs-4838443/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-4838443/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":61750792,"identity":"070cb9c3-cdb1-4ebb-bd9b-56c628c2dbd0","added_by":"auto","created_at":"2024-08-05 07:41:30","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":80582,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eThe flowchart of this study.\u003c/strong\u003e The four steps were illustrated by different colored blocks.\u003c/p\u003e","description":"","filename":"Binder11.png","url":"https://assets-eu.researchsquare.com/files/rs-4838443/v1/aaf49cda2bcc124d0e6b4fb7.png"},{"id":61751550,"identity":"70a8fa3e-460f-4e04-844c-85d718a282f5","added_by":"auto","created_at":"2024-08-05 07:49:30","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":1431806,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eLandscape of the methylation patterns of the phase I discovery set.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eA\u003c/strong\u003e: boxplot showing the overall methylation levels of normal, adenoma and cancer samples. The average β value of all probes for each sample was calculated as the sample overall methylation level. P-values were estimated by Kruskal test. \u003cstrong\u003eB\u003c/strong\u003e: Density curves of probe methylation levels in normal, adenoma and cancer samples. \u003cstrong\u003eC\u003c/strong\u003e: t-SNE visualizing normal, adenoma and cancer samples in the discovery set. \u003cstrong\u003eD\u003c/strong\u003e: Heatmap showing the most variable probes between normal, adenoma and cancer samples.\u003c/p\u003e","description":"","filename":"Binder12.png","url":"https://assets-eu.researchsquare.com/files/rs-4838443/v1/3907575e075bc1b234032eea.png"},{"id":61750829,"identity":"38728773-31ad-434f-a75a-42e1b6649a7a","added_by":"auto","created_at":"2024-08-05 07:41:31","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":233212,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eIdentification of candidate markers. A\u003c/strong\u003e: Percentage of hyper-DMC and hypo-DMC between the three comparisons. \u003cstrong\u003eB\u003c/strong\u003e: Percentage of DMC at different genomic regions between the three comparisons. \u003cstrong\u003eC\u003c/strong\u003e: Venn diagram showing DMCs between the three comparisons. \u003cstrong\u003eD\u003c/strong\u003e: Methylation values of 75 probes meeting the criteria on TCGA 31 cancer types. The “other” refers to 30 non-CRC cancer samples. “Normal” refers to 710 NATs. \u003cstrong\u003eE\u003c/strong\u003e: Frequency of probes with non-zero coefficient in 100 LASSO regressions.\u003c/p\u003e","description":"","filename":"Binder13.png","url":"https://assets-eu.researchsquare.com/files/rs-4838443/v1/494105e150c8c3423d4f2f25.png"},{"id":61750796,"identity":"4b33d74d-947a-4cd9-86ce-d9aa44d6ae01","added_by":"auto","created_at":"2024-08-05 07:41:30","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":69204,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eMethylation profiles of the two candidate probes in different datasets. A:\u003c/strong\u003eThe methylation profiles of cg14015706 and cg08247376 in 656 healthy WBC in GSE40279. \u003cstrong\u003eB:\u003c/strong\u003e The methylation β values of cg14015706 and cg08247376 between normal and cancer samples in GSE48684 dataset. \u003cstrong\u003eC:\u003c/strong\u003e The cfDNA methylation profiles of cg14015706 and cg08247376 in normal and cancer plasma samples from GSE122126. Numbers in the heatmap indicated the methylation β values.\u003c/p\u003e","description":"","filename":"Binder14.png","url":"https://assets-eu.researchsquare.com/files/rs-4838443/v1/1fcaeebe768f661e1f180eb7.png"},{"id":61750798,"identity":"b9e0d6ba-1af2-41fe-a3b4-6c3e83f043fe","added_by":"auto","created_at":"2024-08-05 07:41:31","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":162497,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eThe development and validation of SADMP.\u003c/strong\u003e \u003cstrong\u003eA:\u003c/strong\u003e SADMP strategy for NTMT1. Two pairs of primers were designed according to the sense and antisense BS-strand sequences. CBS indicated the complementary sequence of sense and antisense BS-strands. \u003cstrong\u003eB\u003c/strong\u003e: The standard amplification curve of NTMT1. \u003cstrong\u003eC: \u003c/strong\u003eThe agreement between detected copies by SADMP and single-strand assays and diluted template DNA copies (theoretical copies) of NTMT1. The solid lines were fitted by a simple linear model. \u003cstrong\u003eD \u003c/strong\u003eand\u003cstrong\u003e E\u003c/strong\u003e: The detected Ct values of SADMP NTMT1 and single-strand NTMT1 assays for different diluted concentrations. \u003cstrong\u003eF\u003c/strong\u003e: SADMP strategy for NTMT1. Two MGB probes were designed. Probe 1 is located downstream of the forward primer, targeting the BS-strand template. Probe 2 is located downstream of the reverse primer, targeting the complementary sequence of BS-strand. \u003cstrong\u003eG\u003c/strong\u003e: The standard amplification curve of MAP3K14-AS1. \u003cstrong\u003eH: \u003c/strong\u003eThe agreement between detected copies by SADMP and single-probe assays and diluted template DNA copies (theoretical copies) of MAP3K14-AS1. The solid lines were fitted by a simple linear model. \u003cstrong\u003eI \u003c/strong\u003eand\u003cstrong\u003e J\u003c/strong\u003e: The detected Ct values of SADMP MAP3K14-AS1 and single-probe MAP3K14-AS1 assays for different diluted concentrations.\u003c/p\u003e","description":"","filename":"Binder15.png","url":"https://assets-eu.researchsquare.com/files/rs-4838443/v1/2cc557aec35fef6a2db51c93.png"},{"id":61750794,"identity":"c8525481-c539-45be-b8e8-1e59425f837b","added_by":"auto","created_at":"2024-08-05 07:41:30","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":114881,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003edual-target assessment and validation in plasma samples\u003c/strong\u003e.\u003cstrong\u003e A-C\u003c/strong\u003e：Ct values of ACTB (A), NTMT1 (B) and MAP3K14-AS1 (C) in various clinical groups. NDD: Non-digestive disease, ID: Intestinal disease, AA: advanced adenoma.\u003cstrong\u003e \u003c/strong\u003eThe eight groups were compared by Kruskal-Wallis test. P\u0026lt;0.05 is considered significant. **** is p\u0026lt;0.0001. \u003cstrong\u003eD-E\u003c/strong\u003e: ROC curves of two biomarkers in CRC vs Control groups (D) and CRC+AA vs Control groups (E) in training set. \u003cstrong\u003eF-G\u003c/strong\u003e: Sensitivity and specificity of individual biomarker and dual-target in training set (F), validation set (G) and total set (H) using 1/2 algorithm. Error bars represent 95% CI.\u003c/p\u003e","description":"","filename":"Binder16.png","url":"https://assets-eu.researchsquare.com/files/rs-4838443/v1/f2d9d3e00cd5aa5f1d1f2d24.png"},{"id":61750795,"identity":"d3920063-fe56-4d11-9239-35e8e0367b72","added_by":"auto","created_at":"2024-08-05 07:41:30","extension":"png","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":68069,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eSensitivity, specificity in various clinical groups and other performance indicators of single biomarker and dual-target in total plasma sample set\u003c/strong\u003e.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eA-B\u003c/strong\u003e: Sensitivity in Non-AA and AA groups (A), stages subgroups of CRC (B). \u003cstrong\u003eC\u003c/strong\u003e: Specificity in healthy, NDD, ID and polyps groups. \u003cstrong\u003eD\u003c/strong\u003e: PPV, NPV and accuracy of individual biomarkers and dual-target in CRC vs control groups. Control groups include healthy, NDD, ID, polyps, Non-AA and other cancers. Error bars represent 95% CI. NDD: Non-digestive disease, ID: Intestinal disease, AA: advanced adenoma.\u003c/p\u003e","description":"","filename":"Binder17.png","url":"https://assets-eu.researchsquare.com/files/rs-4838443/v1/c5e6928c4d6f4352559153a1.png"},{"id":61752354,"identity":"a87cee88-27ff-4ba9-86bb-403b2e3fdf60","added_by":"auto","created_at":"2024-08-05 07:57:32","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":2965914,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-4838443/v1/b9d7b5b3-fd76-4a9e-8732-857c4b014878.pdf"},{"id":61750793,"identity":"2cfd0b94-ea76-4d1a-aeec-3ef4debece07","added_by":"auto","created_at":"2024-08-05 07:41:30","extension":"xlsx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":16363,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eSupplementary materials\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eSupplemental Table 1. \u003c/strong\u003eThe clinical features of training and validation cohorts used in this study\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eSupplemental Table 2. \u003c/strong\u003e\u0026nbsp;The information of primers for Sanger sequencing in this study\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eSupplemental Table 3. \u003c/strong\u003e\u0026nbsp;The information of primers and MGB probes used in this study\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eSupplemental Table 4. \u003c/strong\u003e\u0026nbsp;The amplification system of the MSP\u003c/p\u003e","description":"","filename":"SupplementalTablesRSpreprint.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-4838443/v1/da0539a969caa6714d312cf7.xlsx"}],"financialInterests":"The authors declare no competing interests.","formattedTitle":"\u003cp\u003e\u003cstrong\u003eDiscovering methylation markers and development of a sense-antisense and dual-MGB probe PCR assay in plasma for colorectal cancer early detection\u003c/strong\u003e\u003c/p\u003e","fulltext":[{"header":"Background","content":"\u003cp\u003eThe majority of early-stage advanced colorectal neoplasms, including colorectal cancer (CRC) and advanced adenoma (AA) are curable, particularly AA, which can be removed at the time of diagnosis by colonoscopy. \u0026nbsp;Consequently, the early detection of CRC and AA represents a highly valuable endeavor. A number of tests based on cell-free DNA (cfDNA) methylation have been developed and have shown good performance for CRC detection. Epi proColon\u0026nbsp;[1]\u0026nbsp;is the first blood-based test utilized for the early detection of CRC and was developed based on methylated cfDNA of Septin9. The updated version of Epi proColon® 2.0 CE demonstrated an improved sensitivity of 74.8%-81% and specificity of 96.3%-99% in a prospective cohort study\u0026nbsp;[2]. The Chinese version of Epi proColon simplified sample processing with a single reaction system in a larger volume (60 ul) instead of a 2/3 algorithm (20 ul for three runs), resulting a sensitivity of 73% and a specificity of 94.5% [3]. \u0026nbsp;However, its sensitivity for detecting polyps or AA was relatively low in most studies, ranging from 8-40% [4-6]. Due to this relatively poor sensitivity, particularly for precancerous lesions, the US Preventive Services Task Force and the American Cancer Society \u0026nbsp;currently do not include the test in their CRC screening guidelines. Moreover, plasma methylated Septin9 is not a CRC specific marker that it showed an ability to detect multiple cancer types, including hepatocellular carcinoma [7], gastric cancer [8],cervical cancer [9], and breast cancer [10]. Therefore, developing a sensitive and specific blood assay for CRC screening is meaningful.\u003c/p\u003e\n\u003cp\u003e\u0026nbsp;Due to the disparate mechanisms by which cancer signals enter the blood and stool, it is probable that the markers are not universal, and the accuracy of CRC markers in the blood is typically low in comparison to those in stool. Consequently, an ab initio search for blood methylation markers for early CRC and AA is imperative. Furthermore, the detection of\u0026nbsp;the methylation signal of cell-free tumor DNA (ctDNA) in plasma is challenging due to the presence of ctDNA fragments released by multiple organs or tissues, as well as DNA damage caused by bisulfite conversion [11].\u0026nbsp;Therefore, the identification of \u0026nbsp;suitable markers and the development of effective detection techniques are essential to ensure optimal performance\u0026nbsp;[12].\u0026nbsp;\u0026nbsp;Classical PCR-based methylation detection techniques are often developed based on single-strand bisulfite converted DNA (BS-DNA), leaving the information of the other strand unused [13]. Furthermore, only one Taqman MGB probe is usually designed to provide fluorescent signals. Theoretically, designing primers for both sense and antisense strand DNA simultaneously or using multiple MGB probes will improve the sensitivity of a marker to detect methylation signals by enhancing fluorescent signals. Sarah Ø. Jensen et al\u0026nbsp;[14]\u0026nbsp;\u0026nbsp;made the first attempt to design a pair of primers for both sense and antisense strands, thereby enhancing the performance of three targets for CRC detection. Additionally, the dual-strand technique was also successfully applied to methylated \u003cem\u003eHOXA9\u003c/em\u003e in ovarian cancer to improve the detection sensitivity [15].\u003c/p\u003e\n\u003cp\u003eIn this study, we utilized all publicly available methylation data for CRC and adenomas to identify methylation markers, with a particular focus on those that are hypermethylated in adenomas. Furthermore, in order to enhance the ability of candidate markers to detect low-abundance cfDNA methylation signals in plasma, we attempted to apply dual-strand and dual-MGB probe techniques simultaneously, which we called the sense-antisense and dual-MGB probe (SADMP) technique, to develop a novel CRC plasma test. Finally, the test performance was comprehensively assessed in our recruited training and validation cohorts.\u003c/p\u003e"},{"header":"Methods","content":"\u003ch3\u003e\u003cem\u003eData preparation\u003c/em\u003e\u003c/h3\u003e\n\u003cp\u003eThirteen \u0026nbsp;methylation datasets, which were GSE77954, GSE101764, GSE131013, \u0026nbsp;GSE199057, GSE164811, GSE193535, GSE129364, GSE139404, GSE107352, GSE75546, GSE77965, GSE68060 and EMTAB6450 from public databases were selected based on three criteria: 1) were generated by Illumina HumanMethylation 450k BeadChip and had the raw IDAT files, 2) the sample size was greater than 10, and 3) consisted of CRC or adenoma or normal adjacent tissue (NAT). All the IDAT files were then processed using the minfi tool [16] to obtain methylation \u0026beta; values. They were integrated as a single dataset (n= 1165), which we defined as Phase I discovery set for candidate marker identification.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eLevel 3 methylation data for 31 cancer types were retrieved from The Cancer Genome Atlas (TCGA) database (https://portal.gdc.cancer.gov/). Data for 8258 primary tumor tissues corresponding to 31 cancer types and their 710 NATs were retained. The 31 cancer types included ACC (n=79), BLCA (n=412), BRCA (n=778), CESC (n=306), CHOL (n=36), CRC (n=379), DLBC (n=48), ESCA (n=183), GBM (n=137), HNSC (n=523), KICH (n=65), KIRC (n=312), KIRP (n=271), LGG (n=513), LIHC (n=374), LUAD (n=456), LUSC (n=364), MESO (n=87), OV (n=10), PAAD (n=183), PCPG (n=178), PRAD (n=495), SARC (n=257), SKCM (n=104), STAD (n=393), TGCT (n=133), THCA (n=503), THYM (n=124), UCEC (n=418), UCS (n=57), UVM (n=80). The TCGA dataset was defined as Phase II discovery dataset. DMCs were deemed eligible if they exhibited methylation levels \u0026lt;0.2 on other cancer types (non-CRC), \u0026gt;=0.55 on CRC, and \u0026lt;0.15 on 710 NATs.\u003c/p\u003e\n\u003cp\u003eThe other three datasets, GSE48684 [17],\u003csup\u003e\u0026nbsp;\u003c/sup\u003eGSE40279 [18], and GSE122126 [19], generated by the same platform of Illumina HumanMethylation BeadChip in Gene Expression Omnibus (GEO) were used as validation sets to verify the methylation status of candidate differentially methylated CpGs (DMCs). The GSE48684 cohort consisted of 105 qualified samples, of which 41 were NAT and 64 were CRC. The GSE40279 cohort comprised whole blood cell (WBC) samples collected from 656 healthy individuals. The GSE122126 cohort included three CRC and 12 normal plasma samples.\u0026nbsp;\u003c/p\u003e\n\u003ch3\u003e\u003cem\u003e\u0026nbsp;Sample collection\u003c/em\u003e\u003c/h3\u003e\n\u003cp\u003eThis case-control study enrolled 772 cases, including 60 tissue samples (30 CRC and 30 NATs) and 712 plasma samples from the First Affiliated Hospital of Zhengzhou University between April 2022 and June 2022. Tissue samples were obtained from the preserved formalin-fixed paraffin-embedded (FFPE) sections collected from surgical patients. The plasma cohort was randomly divided into a training set and \u0026nbsp;a validation set in a 2:1 ratio. The training set comprised 474 participants, including 115 healthy blood donors, 123 non-digestive disease patients (NDD), 65 intestinal disease patients (ID), 14 polyps patients, 10 non-advanced adenoma patients (Non-AA), 16 AA, 125 CRC patients, and 6 patients with other cancers. In the validation set, there were 235 participants, including 57 healthy blood donors, 60 NDD, 30 intestinal ID, 6 polyps patients, 5 Non-AA, 9 AA, 66 CRC patients, and 2 patients with other cancers (\u003cstrong\u003eSupplemental\u003c/strong\u003e \u003cstrong\u003eTable 1\u003c/strong\u003e). This study was approved by the ethics committee. Each participant signed an informed consent form. Patients with polyps, adenomas, CRC and other cancers were further confirmed by histopathological examinations. The included CRC patients were required to meet the following criteria: 1) did not receive any radiotherapy, chemotherapy or surgery, 2) ages older than 18. All CRC patients were classified as I, II, III, and IV stages according to the American Joint Cancer Committee (AJCC) staging system.\u0026nbsp;\u003c/p\u003e\n\u003ch3\u003e\u003cem\u003eDifferential methylation analysis\u003c/em\u003e\u003c/h3\u003e\n\u003cp\u003e\u0026nbsp;We performed differential methylation analysis using the rank-sum test on three groups in the discovery set: cancer vs. normal (NAT), adenoma vs. normal, and cancer vs. adenoma. DMCs were defined as significant if they meet two criteria: an \u0026nbsp;FDR \u0026lt; 0.05 and a fold change ranking in the \u0026nbsp;top 1% of all probes. The DMCs were then classified as hyper- or hypo- DMCs based on whether they exhibited high or low methylation levels in the tumor or adenoma samples compared to the normal samples. LASSO regression was implemented in the R package \u0026apos;glmnet\u0026apos; to reduce the number of features. \u0026nbsp;LASSO regression was repeated 100 times, and the frequencies of probes with non-zero coefficients in the regressions were counted.\u003c/p\u003e\n\u003ch3\u003e\u003cem\u003eTissue and plasma DNA extraction\u003c/em\u003e\u003c/h3\u003e\n\u003cp\u003eGenomic DNA from tissue samples and plasma cfDNA was isolated using the FFPE DNA rapid extraction kit (TianGen, Beijing) and the plasma/serum cfDNA extraction kit (Ammunition Life-tech, Wuhan), respectively, according to the protocols. Briefly, cfDNA extraction was divided into two steps. First, approximately 10 ml of blood was drawn from participants at room temperature and then centrifuged twice with 1300g and 14000g for 10 min at 4\u0026deg;C within 2 hours. Then, about 2 ml of plasma was retained for cfDNA purification. Purified cfDNA was eluted in 45 ul TE buffer and then treated with bisulfite using a DNA bisulfite modification kit (Ammunition Life-tech, Wuhan) in accordance with the instructions described in the aforementioned study [20], or stored at -80 \u0026deg;C until required.\u003c/p\u003e\n\u003ch3\u003e\u003cem\u003eSanger sequencing\u0026nbsp;\u003c/em\u003e\u003c/h3\u003e\n\u003cp\u003eSanger sequencing (ABI 3730XL, GENEWIZ, Suzhou) was performed for the target regions of \u003cem\u003eNTMT1\u0026nbsp;\u003c/em\u003eand\u003cem\u003e\u0026nbsp;MAP3K14-AS1\u003c/em\u003e after bisulfite treatment to confirm their methylation status on 30 CRC tissue samples and 30 NATs. The sequencing primers are shown in \u003cstrong\u003eSupplemental Table 2\u003c/strong\u003e. Following bisulfite treatment, cytosine in methylated CpG sites remained cytosine, while unmethylated cytosine was converted to thymine. \u0026nbsp; Sanger sequencing results thus allowed for the direct determination of the methylation status of target regions. \u0026nbsp; \u0026nbsp;\u003c/p\u003e\n\u003ch3\u003e\u003cem\u003eDeveloping the SADMP \u0026nbsp;and methylation-specific PCR (MSP) technique on plasma\u003c/em\u003e\u003c/h3\u003e\n\u003cp\u003eThe location and sequences of MSP primers/ probes are displayed in \u003cstrong\u003eSupplemental\u003c/strong\u003e \u003cstrong\u003eTable 3\u003c/strong\u003e. \u0026nbsp;For \u003cem\u003eNTMT1\u003c/em\u003e, two pairs of methylation primers targeted the sense and antisense strands, and their corresponding MGB probes were designed. For \u003cem\u003eMAP3K14-AS1\u003c/em\u003e, the MGB-probe 1 and MGB-probe 2 were designed according to the antisense strand sequence and the reverse complementary sequence of the \u0026nbsp;antisense strand.\u003c/p\u003e\n\u003cp\u003eGradient dilutions of synthetic plasmids or plasmid mixtures are employed as templates for the determination of test parameters. Specifically, standard curve experiments were carried out to assess the amplification efficiency of each pair of primers. The experiments consisted of five ten-fold dilutions of fully methylated plasmid DNA templates with five replicates at each dilution (1, 10, 100, 1000, and 10000 copies per run). Primer tolerance experiments were designed to assess the specific amplification ability of methylated primers against methylated templates. Eight concentration dilutions of fully methylated DNA templates (0, 1, 5, 10, 50, 100, 200, and 400 copies/\u0026mu;L) mixed with a high concentration of unmethylated plasmid (10\u003csup\u003e7\u003c/sup\u003e copies/\u0026mu;L) were prepared to evaluate the ability of methylated primers selectively amplifying methylated templates under the background of unmethylated templates. Each concentration experiment was repeated five times.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThe amplification system of MSP was shown in \u003cstrong\u003eSupplemental Table S4\u003c/strong\u003e. The test employed a triplex PCR with a FAM channel for the internal reference gene \u003cem\u003eACTB\u003c/em\u003e, a ROX channel for \u003cem\u003eNTMT1\u003c/em\u003e (dual-strands), and VIC channel for \u003cem\u003eMAP3K14-AS1\u003c/em\u003e (dual-MGB probes). The PCR procedure was pre-denaturation at 95\u0026deg;C for 5 min (step 1), denaturation at 95\u0026deg;C for 15s (step 2), annealing at 60\u0026deg;C for 30s (step 3), repeating step 2 and 3 forty-five times in plasmid samples and fifity times in plasma samples on the 7500 Fast Real-Time PCR System (Applied Biosystems, USA), the cycle threshold (Ct) was assigned to 50 for the target without amplification.\u003c/p\u003e\n\u003ch3\u003e\u003cem\u003eStatistical analysis\u0026nbsp;\u003c/em\u003e\u003c/h3\u003e\n\u003cp\u003eThe data processing and analysis in this study were conducted using \u0026nbsp;R software (version 4.1.0). The \u0026lsquo;glm\u0026rsquo; function was employed to fit the logistic regression model with a parameter of \u0026lsquo;family=binomial\u0026rsquo;. Receiver operating characteristic (ROC) curve analysis was performed using the \u0026lsquo;pROC\u0026rsquo; package, and the area under the ROC curve (AUC) was subsequently calculated to assess the test classification performance. The optimal sensitivity and specificity of the test were estimated when Youden\u0026rsquo;s index reached its maximal value. Sensitivity and specificity were calculated using the following formulas:\u003c/p\u003e\n\u003cp\u003e\u003cimg src=\"data:image/png;base64,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\"\u003e\u003c/p\u003e\n\u003cp\u003eAnd the Youden index = sensitivity+specificity-1.\u003c/p\u003e\n\u003cp\u003eThe rank-sum test and Kruskal test were employed for the comparisons between two groups and the comparisons between multiple groups, respectively. The Chi-square test was utilized for the comparisons between categorical variables. The Other statistical methods used in this study were described in the corresponding results.\u003c/p\u003e"},{"header":"Results","content":"\u003cdiv id=\"Sec10\" class=\"Section2\"\u003e\n \u003ch2\u003eStudy design and participant characteristics\u003c/h2\u003e\n \u003cp\u003eThe study is divided into four steps, as illustrated in Fig. \u003cspan class=\"InternalRef\"\u003e1\u003c/span\u003e. In the initial step, candidate markers were identified in a set of 1165 samples from 13 datasets. Of these, 452 were normal, 168 were adenomas, and 545 were CRC. These markers were then narrowed down using data from 31 cancer types in the TCGA dataset, DMCs that were hypermethylated in cancer other than CRC were filtered out. The available DMCs were further validated in three independent datesets. In the second step, the methylation status of candidate CpGs was confirmed by Sanger sequencing using 30 CRC tissues and 30 NATs. In the third step, the SADMP technique and MSP system were established. This involved designing appropriate primers, optimizing the qPCR amplification system, and evaluating the technical parameters of the assay. In the last step, the developed assay was tested on training and validation sets of plasma. Its performance was assessed by estimating indicators such as sensitivity, specificity, and AUC.\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec11\" class=\"Section2\"\u003e\n \u003ch2\u003eLandscape of the methylation patterns of the 13 discovery datasets\u003c/h2\u003e\n \u003cp\u003eThe methylation levels of adenoma and cancer samples were found to be lower than normal, as shown in Fig. \u003cspan class=\"InternalRef\"\u003e2\u003c/span\u003eA. The methylation density curves for the three groups displayed bimodal distributions with hyper- and hypomethylation peaks, respectively (Fig. \u003cspan class=\"InternalRef\"\u003e2\u003c/span\u003eB). The hyper-methylation peaks of adenoma and tumor samples were lower than those of normal samples, indicating higher methylation levels in normal samples.\u003c/p\u003e\n \u003cp\u003eWe employed t-SEN to analyze and visualize the structure of the discovery set and identified significant differences between normal and cancer samples. However, adenomas exhibited overlap with both normal and cancer samples (Fig. \u003cspan class=\"InternalRef\"\u003e2\u003c/span\u003eC). We then selected the top 1% of most variable probes to cluster the discovery set using the K-means algorithm. The results showed that both normal and cancer samples clustered together, while adenomas were separated into two subgroups with a remarkable difference. They showed both high (Methy-H) and low (Methy-L) methylation status and were closer to cancer and normal samples, respectively (Fig. \u003cspan class=\"InternalRef\"\u003e2\u003c/span\u003eD).\u003c/p\u003e\n \u003cp\u003eFurther analysis revealed that tubular adenomas constituted the largest proportion of Methy-L adenomas (23/57 or 40.35%), while villous adenomas constituted the largest proportion of Methy-H adenomas (47/73 or 64.38%). To eliminate any potential bias between the datasets, we conducted Fisher\u0026rsquo;s test separately for the Methy-H and Methy-L adenomas, with the datasets serving as the stratification factor. However, we did not identify any significant differences (P\u0026thinsp;\u0026gt;\u0026thinsp;0.05).\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec12\" class=\"Section2\"\u003e\n \u003ch2\u003eIdentification of candidate DMCs\u003c/h2\u003e\n \u003cp\u003eFirst, we conducted a comparative analysis of the DMCs between cancer, adenoma, and normal samples using the 13 discovery datasets. The three comparisons (cancer vs. normal, adenoma vs. normal, and cancer vs. adenoma) yielded 3000, 3051, and 1545 DMCs, respectively. Notably, these DMCs were predominantly hyper-DMCs (Fig. \u003cspan class=\"InternalRef\"\u003e3\u003c/span\u003eA). Further investigations revealed that the majority of DMCs were distributed in upstream regions of genes (5\u0026apos;UTR, TSS1500, TSS500, and 1st Exon), except for cancer vs. adenoma where DMCs were mainly located in gene bodies (Fig. \u003cspan class=\"InternalRef\"\u003e3\u003c/span\u003eB). We observed an extremely high proportion of overlapping DMCs between cancer vs. normal and adenoma vs. normal, which was significantly higher than cancer vs. adenoma (Fig. \u003cspan class=\"InternalRef\"\u003e3\u003c/span\u003eC). To identify the most appropriate DMCs, we focused on those that were overlapped between cancer vs. normal and adenoma vs. normal (2237 probes) and evaluated their methylation levels on 31 cancer types of TCGA. This yielded 75 accessible probes (Fig. \u003cspan class=\"InternalRef\"\u003e3\u003c/span\u003eD). LASSO regression was employed to reduce the number of DMCs, with eight presented in 100 replicates with non-zero coefficients (Fig. \u003cspan class=\"InternalRef\"\u003e3\u003c/span\u003eE).\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec13\" class=\"Section2\"\u003e\n \u003ch2\u003eThe methylation levels of NTMT1 and MAP3K14-AS1 in validation sets\u003c/h2\u003e\n \u003cp\u003eThe methylation levels of the eight probes were then analyzed in 656 WBC samples, and seven DMCs exhibited relatively low methylation levels (less than 0.1) with the exception of cg17892556 (Fig. \u003cspan class=\"InternalRef\"\u003e4\u003c/span\u003eA). Therefore, the seven DMCs are recognized as the most promising markers.\u003c/p\u003e\n \u003cp\u003eTwo specific DNA methylation sites, cg14015706 and cg08247376, were selected for further analysis as they were suitable for primer/probe design. These sites are located in the first exon of \u003cem\u003eNTMT1\u003c/em\u003e and 200 bp upstream of the \u003cem\u003eMAP3K14-AS1\u003c/em\u003e transcription start site, respectively. The GSE48684 dataset revealed significantly higher methylation levels for both probes in cancer samples than in normal samples (Fig. \u003cspan class=\"InternalRef\"\u003e4\u003c/span\u003eB). Furthermore, the two probes demonstrated hypermethylation in CRC plasma and hypomethylation in healthy plasma (Fig. \u003cspan class=\"InternalRef\"\u003e4\u003c/span\u003eC), indicating a high consistency of the methylation status between tissue and plasma.\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec14\" class=\"Section2\"\u003e\n \u003ch2\u003eThe SADMP technique\u003c/h2\u003e\n \u003cp\u003eThe Sanger sequencing of the target regions revealed that the CpGs sites among the amplicons were more frequently methylated in CRC than NATs, which provided the foundation for the development of SADMP technique. For \u003cem\u003eNTMT1\u003c/em\u003e, two pairs of methylation primers targeted the sense and antisense strands, and their corresponding MGB probes were designed (Fig. \u003cspan class=\"InternalRef\"\u003e5\u003c/span\u003eA). For \u003cem\u003eMAP3K14-AS1\u003c/em\u003e, the MGB-probe1 and MGB-porbe2 were designed according to the antisense strand sequence and the reverse complementary sequence of antisense strand (Fig. \u003cspan class=\"InternalRef\"\u003e5\u003c/span\u003eF). The estimated amplification efficiencies of \u003cem\u003eMTNT1\u003c/em\u003e and \u003cem\u003eMAP3K14-AS1\u003c/em\u003e were 98.14% and 110.08%, respectively (Fig. \u003cspan class=\"InternalRef\"\u003e5\u003c/span\u003eB and \u003cspan class=\"InternalRef\"\u003e5\u003c/span\u003eG). The sense and anti-sense strands were transformed into completely different DNA sequences after bisulfite treatment and both could be used as PCR templates. It can be postulated that the copy numbers detected by any single pair of primers or single MGB-probe assay should be half of the diluted copy numbers (including sense and anti-sense strands), and that the SADMP assay should be equal to the diluted copy numbers. For \u003cem\u003eNTMT1\u003c/em\u003e, the detected copy numbers by sense and antisense strand assays were slightly less than half of the theoretical copy numbers with slopes of 0.39 and 0.31, respectively. The detected copy numbers by dual-strand assay were approximately 0.73 times of theoretical copy numbers, which was approximately two-fold compared to any single-strand assay (Fig. \u003cspan class=\"InternalRef\"\u003e5\u003c/span\u003eC), as expected. Meanwhile, Ct value of dual-strand assay was almost one cycle earlier than that of any single-strand assay (average △Ct\u0026thinsp;=\u0026thinsp;1.21, Fig. \u003cspan class=\"InternalRef\"\u003e5\u003c/span\u003eD), which was consistent with the theoretical △Ct value (should be one). The MAP3K14-AS1 SADMP showed similar results, with the number of copies detected was doubled compared to the single probe assay (Fig. \u003cspan class=\"InternalRef\"\u003e5\u003c/span\u003eH), and the Ct values were shifted forward by 0.99 cycles (Fig. \u003cspan class=\"InternalRef\"\u003e5\u003c/span\u003eI).\u003c/p\u003e\n \u003cp\u003eAdditionally, in the serially diluted experiment, no amplification cures for both targets were observed when a high proportion of unmethylated DNA was used as templates (the methylated DNA copies were 0) (Fig. \u003cspan class=\"InternalRef\"\u003e5\u003c/span\u003eE and \u003cspan class=\"InternalRef\"\u003e5\u003c/span\u003eG). These results suggested that the developed assays were explicitly targeted for methylated DNAs even at the high background of unmethylated DNAs. We also found that the \u003cem\u003eNTMT1\u003c/em\u003e SADMP assay was able to robustly detect methylated DNAs at a concentration of 10 copies/\u0026micro;L (Fig. \u003cspan class=\"InternalRef\"\u003e5\u003c/span\u003eE), while it was 5 copies/\u0026micro;L for the MAP3K14-AS1 SADMP ( Fig. \u003cspan class=\"InternalRef\"\u003e5\u003c/span\u003eG).\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec15\" class=\"Section2\"\u003e\n \u003ch2\u003eValidation and evaluation of biomarkers using MSP in plasma samples\u003c/h2\u003e\n \u003cp\u003eIn the plasma sample set, the Ct values of ACTB exhibited a slight decrease in trend from the healthy control groups to the cancer group (p\u0026thinsp;\u0026lt;\u0026thinsp;0.0001) (Fig. \u003cspan class=\"InternalRef\"\u003e6\u003c/span\u003eA). Ct values of \u003cem\u003eNTMT1\u003c/em\u003e and \u003cem\u003eMAP3K14-AS1\u003c/em\u003e were much lower in CRC samples and AA samples than other control groups (Fig. \u003cspan class=\"InternalRef\"\u003e6\u003c/span\u003eB and \u003cspan class=\"InternalRef\"\u003e6\u003c/span\u003eC).\u003c/p\u003e\n \u003cp\u003eThe Ct values could discriminate the CRC and AA samples from healthy samples very well. Furthermore, the Ct values-based result determination exhibited favorable outcomes in a previous study [\u003cspan class=\"CitationRef\"\u003e21\u003c/span\u003e, \u003cspan class=\"CitationRef\"\u003e22\u003c/span\u003e]. Consequently, the Ct values were utilized as the result analysis indicators. In order to determine the algorithm and cutoff, ROC curve analysis was conducted in the training set. In the CRC vs control groups, the AUC values for \u003cem\u003eNTMT1\u003c/em\u003e and \u003cem\u003eMAP3K14-AS1\u003c/em\u003e were 0.882 (95% CI: 0.838\u0026ndash;0.925) and 0.785 (95% CI: 0.729\u0026ndash;0.840), respectively (Fig. \u003cspan class=\"InternalRef\"\u003e6\u003c/span\u003eD). In the CRC\u0026thinsp;+\u0026thinsp;AA vs control groups, the AUC values for \u003cem\u003eNTMT1\u003c/em\u003e and \u003cem\u003eMAP3K14-AS1\u003c/em\u003e were 0.852 (95% CI: 0.807\u0026ndash;0.898) and 0.761 (95% CI: 0.707\u0026ndash;0.815), respectively (Fig. \u003cspan class=\"InternalRef\"\u003e6\u003c/span\u003eE).\u003c/p\u003e\n \u003cp\u003eAs AA was precancerous lesions, it was also benefit patients from detecting it. So we use the CRC\u0026thinsp;+\u0026thinsp;AA vs control groups of training set for further analysis. The performance of the combination of the two biomarker were evaluated by two strategies (Table \u003cspan class=\"InternalRef\"\u003e1\u003c/span\u003e). Strategy 1 involved the construction of a logistic regression model, with an estimated AUC value was 0.881 (95% CI: 0.840\u0026ndash;0.922), the optimal sensitivity for CRC and specificity of 86.4% and 91.0%, respectively (Table \u003cspan class=\"InternalRef\"\u003e1\u003c/span\u003e). Strategy 2 was the 1/2 algorithm, which determined a positive measurement when the Ct of any single marker was less than its corresponding threshold. The cutoff was determined as Ct value corresponding to the maximum Youden index in the training dataset using CRC\u0026thinsp;+\u0026thinsp;AA vs control data. The cutoff values were 48.2 and 48.4 for \u003cem\u003eNTMT1\u003c/em\u003e and \u003cem\u003eMAP3K14-AS1\u003c/em\u003e, respectively. At these cutoff, the two target sensitivities were 80.0% and 59.2%, with specificities of 92.8% and 97.0%, respectively (Fig. \u003cspan class=\"InternalRef\"\u003e6\u003c/span\u003eF). The sensitivity and specificity of the combination of the two biomarkers were 86.4% and 91.0% respectively (Fig. \u003cspan class=\"InternalRef\"\u003e6\u003c/span\u003eF). Notably, strategy 2 yielded an identical sensitivity and specificity to strategy 1. Given that strategy 2 is considerably more straightforward for physicians to interpret test results in clinical practice, we have adopted the 1/2 algorithm as the combination algorithm for the dual-target test.\u003c/p\u003e\n \u003cdiv class=\"gridtable\"\u003e\u0026nbsp;\u003ctable id=\"Tab1\" border=\"1\"\u003e\n \u003ccaption language=\"En\"\u003e\n \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e\n \u003cdiv class=\"CaptionContent\"\u003e\n \u003cp\u003eDiagnostic performance of single biomarker and marker combination in training set\u003c/p\u003e\n \u003c/div\u003e\n \u003c/caption\u003e\n \u003ccolgroup cols=\"7\"\u003e\u003c/colgroup\u003e\n \u003cthead\u003e\n \u003ctr\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eBiomarker \u0026amp; combination\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eAUC (95% CI)\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eSensitivity\u003c/p\u003e\n \u003cp\u003e(CRC)\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eSensitivity\u003c/p\u003e\n \u003cp\u003e(CRC\u0026thinsp;+\u0026thinsp;AA)\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eSpecificity\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eCutoff\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eCombination method\u003c/p\u003e\n \u003c/th\u003e\n \u003c/tr\u003e\n \u003c/thead\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cem\u003eNTMT1\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.852 (0.807\u0026ndash;0.898)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e80.0%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e74.5%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e92.8%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e48.2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e/\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cem\u003eMAP3K14-AS1\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.761 (0.707\u0026ndash;0.815)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e59.2%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e54.6%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e97.0%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e48.4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e/\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cem\u003eNTMT1\u003c/em\u003e \u0026amp; \u003cem\u003eMAP3K14-AS1\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.881 (0.840\u0026ndash;0.922)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e86.4%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e80.8%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e91.0%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.1491\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eLogistic regression\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cem\u003eNTMT1\u003c/em\u003e \u0026amp; \u003cem\u003eMAP3K14-AS1\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e/\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e86.4%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e80.8%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e91.0%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e48.2, 48.4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e1/2 algorithm\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n \u003c/table\u003e\n \u003c/div\u003e\n \u003cp\u003eAfter the algorithm and cutoff was set, the validation set and all sample set were analyzed. The validation set had a 81.8% sensitivity and 92.5% specificity (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e6\u003c/span\u003eG) and the all sample set had a 84.8% sensitivity and 91.5% specificity (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e6\u003c/span\u003eH). The dual-target test had a higher sensitivity than that of any single biomarker and a slightly lower specificity in the three data sets (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e6\u003c/span\u003eF-H). It indicated that the combination of the two biomarkers were better than single biomarker.\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec16\" class=\"Section2\"\u003e\n \u003ch2\u003ePerformance in subgroups of plasma samples\u003c/h2\u003e\n \u003cp\u003eThe performance of the dual-target test was evaluated in subgroups of all plasma samples. The test demonstrated a detection rate of 26.7% in non-AA samples and a detection rate of 32.0% in AA samples (Fig. \u003cspan class=\"InternalRef\"\u003e7\u003c/span\u003eA). It exhibited sensitivities of 75.0%, 81.2% and 90.3% in stages I, II and III-IV of CRC samples, respectively (Fig. \u003cspan class=\"InternalRef\"\u003e7\u003c/span\u003eB). The test exhibited specificities of 95.9%, 92.3%, 84.2% and 90.9% in healthy, NDD, ID and polyps samples, respectively (Fig. \u003cspan class=\"InternalRef\"\u003e7\u003c/span\u003eC). The positive predictive value (PPV), negative predictive value (NPV) and accuracy for the dual-target test were 79.4%, 94.0% and 89.6%, respectively (Fig. \u003cspan class=\"InternalRef\"\u003e7\u003c/span\u003eD).\u003c/p\u003e\n \u003cp\u003eWith regard to the detection rates in adenoma and different stages of CRC, the dual-target test demonstrated superior performance compared to that of a single biomarker (Fig. \u003cspan class=\"InternalRef\"\u003e7\u003c/span\u003eA-B). With respect to specificities in subgroups, the dual-target test exhibited a marginal decline in value compared to that of a single biomarker (Fig. \u003cspan class=\"InternalRef\"\u003e7\u003c/span\u003eC). However, with regard to NPV and accuracy, the dual-target test demonstrated superior performance compared to that of a single biomarker (Fig. \u003cspan class=\"InternalRef\"\u003e7\u003c/span\u003eD). With regard to PPV, the dual-target test exhibited a slight decline in performance relative to that of a single biomarker (Fig. \u003cspan class=\"InternalRef\"\u003e7\u003c/span\u003eD).\u003c/p\u003e\n\u003c/div\u003e"},{"header":"Discussion","content":"\u003cp\u003eIt is commonly accepted that patients diagnosed with CRC at an early stage can be treated more effectively and have a better prognosis. Several stool DNA-based tests have been developed that demonstrate excellent performance in detecting CRCs at their early stages [\u003cspan additionalcitationids=\"CR24 CR25\" citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e]. Blood sampling is more acceptable than stool sampling, but blood-based tests are less reported and usually exhibit lower sensitivities than stool-DNA tests, ranging from 47\u0026ndash;87% [\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e]. This study presented a systematic pipeline for the discovery of methylation markers from scratch, test development, and evaluation in training and validation plasma sets. The SADMP technique enhanced the ability to detect methylation signals. Following a comprehensive evaluation, the test obtained an overall sensitivity and specificity of 84.8% and 91.5%, respectively, for the detection of CRC at a volume of 2 ml plasma.\u003c/p\u003e \u003cp\u003eAdenoma and CRC displayed lower methylation levels overall, with the exception of regulatory regions (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eB and \u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eB), a finding that has been reported in previous study [\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e]. Previous studies have focused on the CpG Island Methylator Phenotype (CIMP) found in CRC. This study found that tubular adenomas are common in the methy-L subclass, while villous adenomas are more often in methy-H subclass. Previous studies have indicated that CIMP is rarely found in tubular adenomas, but frequently in tubulovillous and villous adenomas [\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e], which is in accordance with this study. The large proportion of overlapping DMCs between cancer vs normal and adenoma vs normal indicated that many CpGs have undergone aberrant methylation events at the precancerous stage, which provides robust evidence for discovering the methylation markers for CRC early detection. It should be noted that the CRC markers are not necessarily applicable to adenomas, as a significant proportion of the DMCs in the Cancer vs Normal group were not present in the Adenoma vs Normal group, as illustrated in Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eC. This emphasises the importance of this study's inclusion of adenoma samples in the marker discovery set and the selection of DMCs in which cancers and adenomas overlap in a way that many other studies of CRC markers have not done.\u003c/p\u003e \u003cp\u003eBlood-based tests are more susceptible to interfering diseases and may result in a high false-positive rate. Therefore, specificity is a critical indicator. In the marker discovery step, 31 cancer types in the TCGA database and WBC were used to control the low methylation levels of candidate markers in other tissues and blood background, effectively attenuating the interference of unintended cfDNAs. In the assay development phase, we designed highly selective MSP primers that did not show normal amplification curves even when unmethylated DNAs were used as templates at 10\u003csup\u003e7\u003c/sup\u003e copies (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eE and \u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eJ). These aforementioned measures guarantee a high specificity in plasma samples. Two combination algorithms were utilized to assess the performance of the dual-target test in the training set, and both algorithms indicated that the combined markers had better AUC values and higher sensitivities than any single marker (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eD and \u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eE). However, the dual-target test showed a decreased specificity compared to both single markers, from 92.8% and 97.0\u0026ndash;91.0% (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eH), which was also observed in other studies [\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e, \u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e]. When healthy individuals were selected as controls, the specificity improved to 95.9% (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003eC), which is comparable to the SEPT9 test [\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e]. These data demonstrate the excellent performance of the dual-target test in detecting CRC.\u003c/p\u003e \u003cp\u003eA number of studies [\u003cspan additionalcitationids=\"CR32 CR33 CR34\" citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e] have demonstrated that the methylation levels of \u003cem\u003eNTMT1\u003c/em\u003e (whose antisense chain counterpart is \u003cem\u003eC9orf50\u003c/em\u003e) and \u003cem\u003eMAP3K14-AS1\u003c/em\u003e can be employed for the screening and diagnosis of CRC. In particular, the study conducted by Sarah \u0026Oslash; Jensen et al [\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e]. indicated that the C9orf50 methylation assay exhibited a plasma sensitivity of 76% and specificity of 91% for CRC. Ludovic Barault and colleagues [\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e] demonstrated that the MAP3K14-AS1 methylation assay in plasma exhibited a sensitivity of 69.8% and a specificity of 100% for CRC. The dual-strand technique has been proven to enhance the performance of markers in previous studies [\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e, \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e]. This technique was also observed to be effective in our study. By detecting the methylation signals of \u003cem\u003eNTMT1\u003c/em\u003e sense- and antisense-strand simultaneously, the Ct value of the dual-strand assay was able to shift forward by one compared to the single-strand assay (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eD). In contrast to previous studies, the current study also included two MGB probes located downstream of forward and reverse primers of \u003cem\u003eMAP3K14-AS1.\u003c/em\u003e During PCR strand extension, the polymerase enzymes cleaved the 5-primer sequence of probes and released two fluorescent groups. The dual-MGB probe technique theoretically doubled the fluorescent signals when both probes shared the same channel, leading to an earlier Ct value similar to that of the dual-strand technique (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eI). Serial dilution experiments confirmed the superiority of dual-MGB probes over one MGB probe (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eH-J). These results suggest that applying the SADMP technique can be a feasible strategy to enhance the detection sensitivity of candidate markers.\u003c/p\u003e \u003cp\u003eEarly diagnosis or screening techniques are essential to improve patient survival time, particularly when curable treatments are available. Studies have shown that the 5-year survival rate of early detected CRC is almost 90%, while it was only 20% for advanced CRC [\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e]. The dual-target test showed a sensitivity of 75.0% and 81.2% for stage I and stage II CRC detection, notably, the dual-target test obtained a positive detection rate of 32.00% (8/25) for AA (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003eB and \u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003eC), implying its ability to detect early CRC and precancerous lesions.\u003c/p\u003e \u003cp\u003eThe current study has some limitations that may hamper the interpretation of these results. 1) Participants in this study were enrolled from a single center, which may bias these results. 2) the SADMP techniques may not applicable for all candidate markers. The dual-strand technique may be attempted when both sense and antisense strands are suitable for designing MSP primers, while the multiple MGB probe technique is limited by the amplicon length, which is usually less than 100 bp.\u003c/p\u003e"},{"header":"Conclusion","content":"\u003cp\u003eIn this study, we employed several public databases of adenomas and CRC for marker discovering, and ultimately identified two promising markers, \u003cem\u003eNTMT1\u003c/em\u003e and \u003cem\u003eMAP3K14-AS1\u003c/em\u003e. We then constructed the SADMP technology based on these two markers, which enhanced the sensitivity of the detection. The dual-target assay has a high sensitivity for AA and early stage CRC, and its clinical application value merits further investigation.\u003c/p\u003e"},{"header":"Abbreviations","content":"\u003cdiv class=\"DefinitionList\"\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eAA\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eAdvanced adenoma\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eCRC\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eColorectal cancer\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003ecfDNA\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eCell-free DNA\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eSADMP\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eSense-antisense and dual-MGB probe\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eNAT\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eNormal adjacent tissue\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eTCGA\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eThe Cancer Genome Atlas\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eGEO\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eGene Expression Omnibus\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eDMC\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eDifferentially methylated CpGs\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eWBC\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eWhole blood cell (WBC)\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eFFPE\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eFormalin-fixed paraffin-embedded\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eNDD\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eNon-digestive disease patients\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eID\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eIntestinal disease\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eNon-AA\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eNon-advanced adenoma\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eMSP\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eMethylation-specific PCR\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eROC\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eReceiver operating characteristic\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eAUC\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eArea under the ROC curve\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003ePPV\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003ePositive predictive value\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eNPV\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eNegative predictive value\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003c/div\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eEthics approval and consent to participate\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis study was approved by the ethics committee of the First Affiliated Hospital of Zhengzhou University (approval number: 2022-KY-0631-002). Each participant signed an informed consent form.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eClinical Trial Registration\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis study is a sub-project of the Clinical Study of Pan-cancer DNA Methylation Test in plasma (Clinical Trials ID: NCT05685524)\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAvailability of data and materials\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe datasets (GSE77954, GSE101764, GSE131013, GSE164811, GSE193535, GSE129364, GSE139404, GSE107352, GSE75546, GSE77965, GSE199057, GSE68060, GSE48684, GSE40279 and GSE122126) supporting the conclusions of this article are available in the Gene Expression Omnibus database (https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi). The E-MTAB-6450 data is available in EMBL biostudies (https://www.ebi.ac.uk/biostudies/arrayexpress). The source code and intermediate data can be accessed from git-hub (https://github.com/amsinfor/CRC-methylation). \u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCompeting interests\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors declare that they have no competing interests.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis study was supported by science and technology project of Henan Province (232102310028).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthors' contributions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eQK Y and YT Z conceptualized and designed the study. YT Z，ZJ W and DH Z developed the SADMP technique. ZJ W QN Y, and X L provided clinical samples and patient information. X L and SL D conducted cfDNA extraction and bisulfite converted. YT Z, ZJ W and T Z performed data analysis and prepared the manuscript. XP L validated the study results. All the authors reviewed the manuscript.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n \u003cli\u003ePowr\u0026oacute;zek T, Krawczyk P, Kucharczyk T, Milanowski J. Septin 9 promoter region methylation in free circulating dna\u0026mdash;potential role in noninvasive diagnosis of lung cancer: preliminary report. \u003cem\u003eMed Oncol\u0026nbsp;\u003c/em\u003e2014;31: 917. doi: 10.1007/s12032-014-0917-4\u003c/li\u003e\n \u003cli\u003eLamb YN, Dhillon S. Epi procolon\u0026reg; 2.0 ce: a blood-based screening test for colorectal cancer. \u003cem\u003eMol Diagn Ther\u0026nbsp;\u003c/em\u003e2017;21: 225-32. doi: 10.1007/s40291-017-0259-y\u003c/li\u003e\n \u003cli\u003eSun J, Fei F, Zhang M, Li Y, Zhang X, Zhu S, et al. The role of msept9 in screening, diagnosis, and recurrence monitoring of colorectal cancer. \u003cem\u003eBmc Cancer\u0026nbsp;\u003c/em\u003e2019;\u003cem\u003e\u0026nbsp;\u003c/em\u003e19: 450. doi: 10.1186/s12885-019-5663-8\u003c/li\u003e\n \u003cli\u003eJung G, Hern\u0026aacute;ndez-Ill\u0026aacute;n E, Moreira L, Balaguer F, Goel A. Epigenetics of colorectal cancer: biomarker and therapeutic potential. Nat Rev Gastroenterol Hepatol 2020;17(2):111-130. doi: 10.1038/s41575-019-0230-y.\u003c/li\u003e\n \u003cli\u003eChurch TR, Wandell M, Lofton-Day C, Mongin SJ, Burger M, Payne SR, et al. Prospective evaluation of methylated SEPT9 in plasma for detection of asymptomatic colorectal cancer. Gut 2014; 63(2):317-25. doi: 10.1136/gutjnl-2012-304149.\u003c/li\u003e\n \u003cli\u003eNassar FJ, Msheik ZS, Nasr RR, Temraz SN. Methylated circulating tumor DNA as a biomarker for colorectal cancer diagnosis, prognosis, and prediction. Clin Epigenetics 2021;13(1):111. doi: 10.1186/s13148-021-01095-5.\u003c/li\u003e\n \u003cli\u003eLi B, Huang H, Huang R, Zhang W, Zhou G, Wu Z, et al. Sept9 gene methylation as a noninvasive marker for hepatocellular carcinoma. \u003cem\u003eDis Markers\u0026nbsp;\u003c/em\u003e2020; 2020: 6289063. doi: 10.1155/2020/6289063\u003c/li\u003e\n \u003cli\u003eCao CQ, Chang L, Wu Q. Circulating methylated septin 9 and ring finger protein 180 for noninvasive diagnosis of early gastric cancer. \u003cem\u003eTransl Cancer Res\u0026nbsp;\u003c/em\u003e2020; 9: 7012-21. doi: 10.21037/tcr-20-1330\u003c/li\u003e\n \u003cli\u003eJiao X, Zhang S, Jiao J, Zhang T, Qu W, Muloye GM, et al. Promoter methylation of sept9 as a potential biomarker for early detection of cervical cancer and its overexpression predicts radioresistance. \u003cem\u003eClin Epigenetics\u003c/em\u003e 2019;11: 120. doi: 10.1186/s13148-019-0719-9\u003c/li\u003e\n \u003cli\u003eMatsui S, Kagara N, Mishima C, Naoi Y, Shimoda M, Shimomura A, et al. Methylation of the sept9_v2 promoter as a novel marker for the detection of circulating tumor dna in breast cancer patients. \u003cem\u003eOncol Rep\u0026nbsp;\u003c/em\u003e2016; 36: 2225-35. doi: 10.3892/or.2016.5004\u003c/li\u003e\n \u003cli\u003eSong P, Wu LR, Yan YH, Zhang JX, Chu T, Kwong LN,et al. Limitations and opportunities of technologies for the analysis of cell-free DNA in cancer diagnostics. Nat Biomed Eng 2022;6(3):232-245. doi: 10.1038/s41551-021-00837-3.\u003c/li\u003e\n \u003cli\u003eBabayan A, Pantel K. Advances in liquid biopsy approaches for early detection and monitoring of cancer. \u003cem\u003eGenome Med\u0026nbsp;\u003c/em\u003e2018;\u003cem\u003e\u0026nbsp;\u003c/em\u003e10: 21. doi: 10.1186/s13073-018-0533-6\u003c/li\u003e\n \u003cli\u003eLi LC, Dahiya R. Methprimer: designing primers for methylation pcrs. \u003cem\u003eBioinformatics\u0026nbsp;\u003c/em\u003e2002;18: 1427-31. doi: 10.1093/bioinformatics/18.11.1427\u003c/li\u003e\n \u003cli\u003eJensen S\u0026Oslash;, \u0026Oslash;gaard N, Nielsen HJ, Bramsen JB, Andersen CL. Enhanced performance of dna methylation markers by simultaneous measurement of sense and antisense dna strands after cytosine conversion. \u003cem\u003eClin Chem\u0026nbsp;\u003c/em\u003e2020; 66: 925-33. doi: 10.1093/clinchem/hvaa100\u003c/li\u003e\n \u003cli\u003eFaaborg L, Fredslund AR, Waldstr\u0026oslash;m M, H\u0026oslash;gdall E, H\u0026oslash;gdall C, Adimi P, et al. Analysis of hoxa9 methylated ctdna in ovarian cancer using sense-antisense measurement. \u003cem\u003eClin Chim Acta\u003c/em\u003e 2021;\u003cem\u003e\u0026nbsp;\u003c/em\u003e522: 152-7. doi: 10.1016/j.cca.2021.08.020\u003c/li\u003e\n \u003cli\u003eAryee MJ, Jaffe AE, Corrada-Bravo H, Ladd-Acosta C, Feinberg AP, Hansen KD, et al. Minfi: a flexible and comprehensive bioconductor package for the analysis of infinium dna methylation microarrays. \u003cem\u003eBioinformatics\u0026nbsp;\u003c/em\u003e2014; 30: 1363-9. doi: 10.1093/bioinformatics/btu049\u003c/li\u003e\n \u003cli\u003eLuo Y, Wong CJ, Kaz AM, Dzieciatkowski S, Carter KT, Morris SM, et al. Differences in dna methylation signatures reveal multiple pathways of progression from adenoma to colorectal cancer. \u003cem\u003eGastroenterology\u0026nbsp;\u003c/em\u003e2014; 147: 418-29. doi: 10.1053/j.gastro.2014.04.039\u003c/li\u003e\n \u003cli\u003eHannum G, Guinney J, Zhao L, Zhang L, Hughes G, Sadda S, et al. Genome-wide methylation profiles reveal quantitative views of human aging rates. \u003cem\u003eMol Cell\u0026nbsp;\u003c/em\u003e2013; 49: 359-67. doi: 10.1016/j.molcel.2012.10.016\u003c/li\u003e\n \u003cli\u003eMoss J, Magenheim J, Neiman D, Zemmour H, Loyfer N, Korach A, et al. Comprehensive human cell-type methylation atlas reveals origins of circulating cell-free dna in health and disease. \u003cem\u003eNat Commun\u0026nbsp;\u003c/em\u003e2018; 9: 5068. doi: 10.1038/s41467-018-07466-6\u003c/li\u003e\n \u003cli\u003eLi R, Qu B, Wan K, Lu C, Li T, Zhou F, et al. Identification of two methylated fragments of an sdc2 cpg island using a sliding window technique for early detection of colorectal cancer. \u003cem\u003eFebs Open Bio\u003c/em\u003e. (2021) 11: 1941-52. doi: 10.1002/2211-5463.13180\u003c/li\u003e\n \u003cli\u003eZhang L, Dong L, Lu C, Huang W, Yang C, Wang Q, et al. Methylation of SDC2/TFPI2 and Its Diagnostic Value in Colorectal Tumorous Lesions. Front Mol Biosci. 2021;8:706754. doi: 10.3389/fmolb.2021.706754.\u003c/li\u003e\n \u003cli\u003eBian Y, Gao Y, Lu C, Tian B, Xin L, Lin H, et al. Genome-wide methylation profiling identified methylated KCNA3 and OTOP2 as promising diagnostic markers for esophageal squamous cell carcinoma. Chin Med J (Engl) 2023; doi: 10.1097/CM9.0000000000002832.\u003c/li\u003e\n \u003cli\u003eWang Z, Shang J, Zhang G, Kong L, Zhang F, Guo Y, et al. Evaluating the clinical performance of a dual-target stool dna test for colorectal cancer detection. \u003cem\u003eJ Mol Diagn\u0026nbsp;\u003c/em\u003e2022;\u003cem\u003e\u0026nbsp;\u003c/em\u003e 24: 131-43. doi: 10.1016/j.jmoldx.2021.10.012\u003c/li\u003e\n \u003cli\u003eImperiale TF, Ransohoff DF, Itzkowitz SH, Levin TR, Lavin P, Lidgard GP, et al. Multitarget stool dna testing for colorectal-cancer screening. \u003cem\u003eN Engl J Med\u0026nbsp;\u003c/em\u003e2014;370: 1287-97. doi: 10.1056/NEJMoa1311194\u003c/li\u003e\n \u003cli\u003eZhao G, Liu X, Liu Y, Li H, Ma Y, Li S, et al. Aberrant dna methylation of sept9 and sdc2 in stool specimens as an integrated biomarker for colorectal cancer early detection. \u003cem\u003eFront Genet\u0026nbsp;\u003c/em\u003e2020; 11: 643. doi: 10.3389/fgene.2020.00643\u003c/li\u003e\n \u003cli\u003eOh TJ, Oh HI, Seo YY, Jeong D, Kim C, Kang HW, et al. Feasibility of quantifying sdc2methylation in stool dna for early detection of colorectal cancer. \u003cem\u003eClin Epigenetics\u0026nbsp;\u003c/em\u003e2017; 9: 126. doi: 10.1186/s13148-017-0426-3\u003c/li\u003e\n \u003cli\u003eNassar FJ, Msheik ZS, Nasr RR, Temraz SN. Methylated circulating tumor dna as a biomarker for colorectal cancer diagnosis, prognosis, and prediction. \u003cem\u003eClin Epigenetics\u0026nbsp;\u003c/em\u003e2021; 13: 111. doi: 10.1186/s13148-021-01095-5\u003c/li\u003e\n \u003cli\u003eKakar S, Deng G, Cun L, Sahai V, Kim YS. Cpg island methylation is frequently present in tubulovillous and villous adenomas and correlates with size, site, and villous component. \u003cem\u003eHum Pathol\u0026nbsp;\u003c/em\u003e2008; 39: 30-6. doi: 10.1016/j.humpath.2007.06.002\u003c/li\u003e\n \u003cli\u003eZhao G, Li H, Yang Z, Wang Z, Xu M, Xiong S, et al. Multiplex methylated dna testing in plasma with high sensitivity and specificity for colorectal cancer screening. \u003cem\u003eCancer Med\u0026nbsp;\u003c/em\u003e2019; 8: 5619-28. doi: 10.1002/cam4.2475\u003c/li\u003e\n \u003cli\u003eBagheri H, Mosallaei M, Bagherpour B, Khosravi S, Salehi AR, Salehi R. Tfpi2 and ndrg4 gene promoter methylation analysis in peripheral blood mononuclear cells are novel epigenetic noninvasive biomarkers for colorectal cancer diagnosis. \u003cem\u003eJ Gene Med\u0026nbsp;\u003c/em\u003e2020;22: e3189. doi: 10.1002/jgm.3189\u003c/li\u003e\n \u003cli\u003eCao Y, Zhao G, Yuan M, Liu X, Ma Y, Cao Y, et al. KCNQ5 and C9orf50 Methylation in Stool DNA for Early Detection of Colorectal Cancer. Front Oncol 2021; 29;10:621295. doi: 10.3389/fonc.2020.621295.\u003c/li\u003e\n \u003cli\u003eZhang Y, Wu Q, Xu L, Wang H, Liu X, Li S, et al. Sensitive detection of colorectal cancer in peripheral blood by a novel methylation assay. Clin Epigenetics 2021;13(1):90. doi: 10.1186/s13148-021-01076-8.\u003c/li\u003e\n \u003cli\u003eJensen S\u0026Oslash;, \u0026Oslash;gaard N, \u0026Oslash;rntoft MW, Rasmussen MH, Bramsen JB, Kristensen H, et al. Novel DNA methylation biomarkers show high sensitivity and specificity for blood-based detection of colorectal cancer-a clinical biomarker discovery and validation study. Clin Epigenetics 2019;11(1):158. doi: 10.1186/s13148-019-0757-3.\u003c/li\u003e\n \u003cli\u003eBarault L, Amatu A, Siravegna G, Ponzetti A, Moran S, Cassingena A, et al. Discovery of methylated circulating DNA biomarkers for comprehensive non-invasive monitoring of treatment response in metastatic colorectal cancer. Gut 2018; 67(11):1995-2005. doi: 10.1136/gutjnl-2016-313372.\u003c/li\u003e\n \u003cli\u003eHuang H, Cao W, Long Z, Kuang L, Li X, Feng Y, et al. DNA methylation-based patterns for early diagnostic prediction and prognostic evaluation in colorectal cancer patients with high tumor mutation burden. Front Oncol 2023;12:1030335. doi: 10.3389/fonc.2022.1030335.\u003c/li\u003e\n \u003cli\u003eLadabaum U, Dominitz JA, Kahi C, Schoen RE. Strategies for colorectal cancer screening. \u003cem\u003eGastroenterology\u0026nbsp;\u003c/em\u003e2020; 158: 418-32. doi: 10.1053/j.gastro.2019.06.043\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":true,"highlight":"","institution":"The First Affiliated Hospital of Zhengzhou University","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"colorectal cancer, methylation, NTMT1, MAP3K14-AS1 ","lastPublishedDoi":"10.21203/rs.3.rs-4838443/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-4838443/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003e\u003cstrong\u003eBackground\u003c/strong\u003e: Screening for colorectal cancer (CRC) using plasma cell-free DNA (cfDNA) methylation is more challenging than stool testing due to the low abundance of cfDNA. Therefore, the development of signal amplification assays based on appropriate markers is essential to increase sensitivity.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eMethods\u003c/strong\u003e: A total of 17 existing 450K microarray datasets including tissue, healthy white blood cell (WBC) and plasma cfDNA data from public databases were used to identify differentially methylated CpGs (DMCs) common to CRC and adenoma. The methylation status of candidate DMCs was confirmed by Sanger sequencing with CRC and normal tissues. A sense-antisense and dual MGB probe (SADMP) assay was then developed. Subsequently, the biomarkers were validated in 712 plasma samples using the SADMP method.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eResults\u003c/strong\u003e: A total of 2237 DMCs showed overlap between the cancer vs. normal and adenoma vs. normal groups. Of these, 75 were hypomethylated in 30 other non-CRC cancers. After LASSO regression, this number was reduced to eight. Two of these, \u003cem\u003eNTMT1\u003c/em\u003e and \u003cem\u003eMAP3K14-AS1\u003c/em\u003e, were identified as promising candidate markers following WBC validation and primer/probe design evaluation. The SADMP technology demonstrated the ability to amplify the detection signal to approximately twice the original level. Overall, the dual-target SADMP assay demonstrated a sensitivity of 84.8% for CRC (stage I: 75.0%), a sensitivity of 32.0% for advanced adenomas (AA), and a specificity of 91.5% in controls.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConclusions\u003c/strong\u003e: The dual-target assay demonstrated high performance for CRC and AA detection in plasma-based tests, suggesting that it may serve as a promising noninvasive tool for CRC detection.\u003c/p\u003e","manuscriptTitle":"Discovering methylation markers and development of a sense-antisense and dual-MGB probe PCR assay in plasma for colorectal cancer early detection","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2024-08-05 07:41:26","doi":"10.21203/rs.3.rs-4838443/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"da38d593-eb13-4b13-a224-b1ded46b73e8","owner":[],"postedDate":"August 5th, 2024","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":35477169,"name":"Cancer Biology"}],"tags":[],"updatedAt":"2024-08-05T07:41:26+00:00","versionOfRecord":[],"versionCreatedAt":"2024-08-05 07:41:26","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-4838443","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-4838443","identity":"rs-4838443","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}