Diagnosis of Colorectal Cancer and Advanced Adenoma Using Cancer-specific Methylation Signatures

1. Research Background 1.1 Diagnosis of colorectal cancer Colorectal cancer is a common malignant tumor of the digestive tract. According to statistics from 2015, both morbidity and mortality of colorectal cancer in China are among the top five of all cancers. Studies have shown that the five-year survival rate of patients can achieve 90% if patients are diagnosed before the spread of tumor cells. However only 40% of patients are currently diagnosed at early stage. Therefore, treatment in the early stage of the disease is a necessary means to cope with colorectal cancer. The main method used clinically for diagnosis is colonoscopy. This method is accurate, but limited by many factors, such as diarrhea, intestinal perforation, and need to be prepared in many ways before the examination. The patients need to eat fluid diet, clean the intestines, etc. Other diagnostic methods such as imaging, histopathological examination have a low detection rate, or cause damage to the organ. Therefore, it is a still challenging task to detect colorectal cancer at an early stage. 1.2 Detection of ctDNA methylation in colorectal cancer patients and clinical studies Epigenetics refers to a heritable phenotypic change without a change in the primary sequence of the DNA. DNA methylation is one of the epigenetic modification pathways, which can cause changes in chromatin structure and DNA stability, and regulate gene transcription and expression. Studies have found that DNA methylation has an close relationship with the occurrence and development of tumors. DNA methylation changes occur at numerous sites of DNA in tumor cells as compared to normal cells. Further studies show that DNA methylation occur early in the process of cancer, making it possible to use DNA methylation for early screening of cancer. Based on the research paper published by professor Kun Zhang in Nature Genetics, Singlera Genomics Inc. invented the proprietary methyl-Titan sequencing technology and developed a detection method for colorectal cancer and advanced adenoma (Adenoma/Colorectal cancer Early detection, ACE). ACE is a blood-based non-invasive diagnostic technique. It has high compliance rate compared with colonoscopy, and sampling is more convenient than stool testing. It also has much higher sensitivity compared to existing blood testing methods. Based on these advantages, the ACE technology is suitable for early colorectal cancer diagnosis. During the development stage Singlera Genomics Inc. used a high-throughput and high-coverage screening method for 4 million CpG methylation sites in the genome, and analyzed different stages of colorectal cancer, adenoma, polyp tissue samples and paired normal tissues to screen colorectal cancer-related methylation sites. Plasma samples from cancer patients and healthy people were also analyzed to select ctDNA methylation markers for colorectal cancer. The current study plans to use ACE method to analyze ctDNA in the blood for the cancer-specific DNA methylation markers to aid in the differential diagnosis of patients with colorectal cancer and adenoma. This technique will greatly reduce the discomfort in the diagnosis of suspected patients and improve the diagnosis of high-risk population of colorectal cancer. The higher detection rate is beneficial to the early detection and early treatment of colorectal cancer, which will effectively save medical costs and improve the survival rate of patients.
2. Research goals 2.1. Primary goals To establish a detection system based on plasma ctDNA methylation sequencing technology for the auxiliary diagnosis of colorectal cancer and adenoma; To assess the diagnostic value of plasma ctDNA methylation signature for colorectal cancer and adenoma. 2.2. Secondary goals To assess the association of plasma ctDNA methylation signals with colonoscopy results and pathological results of surgical specimens.
3. Research Overview 3.1 Research design and planning This study was a prospective, open, controlled, multicenter study. The enrolled samples contain a positive group (including colorectal cancer stage I-IV, advanced adenoma) and a negative group (non-advanced adenoma, patients with negative pathological examination, healthy people with negative colonoscopy results, benign hyperplasia and inflammatory colorectal disease without dysplasia). The sample sizes are: 700 cases positive and 600 cases negative. 3.2 Number of cases and grouping scheme This trial plans to enroll 1300 subjects, which will be assigned into two groups. The positive group includes 300 cases of colorectal cancer (stage I-IV) and 400 cases of advanced adenoma (high grade dysplasia, villous adenoma, tubular adenoma, serrated lesions, etc.). The negative group (600 cases in total) includes: non-advanced adenomas (size≤10mm, no less than 100 cases), inflammatory colorectal disease, benign hyperplasia, tumor-free patients upon histopathological review, and normal healthy people. 3.3 Research procedure 3.3.1 Patient screening The researcher will ask each subject for basic information, history of illness, family history, medication history, and whether or not they have participated related clinical trials to determine if they meet the inclusion criteria. Based on their records the subjects who meet the inclusion criteria are first divided in a simple positive group and a negative group. An informed consent form will be signed for each enrolled patient. All subjects who complete the initial grouping will have their peripheral venous blood collected before undergoing colonoscopy and clinical management. 20 ml of the blood sample is centrifuged according to the ctDNA detection plasma separation requirements. The supernatant plasma is collected and immediately refrigerated. 3.3.2 Clinical exam Colonoscopy is performed on all subjects (except those having surgical pathology results). For patients with intestinal lesions, colonoscopy is performed to determine whether they are positive or negative. Positive patients who need surgical resection will be further classified according to their surgical histopathological results. For negative patients, the type of lesion should be clarified. 3.3.3 ACE testing All the plasma samples of the subjects assigned to specific groups are transported to Shanghai Singlera Genomics Inc. (No. 20, Lane 500, Furonghua Road, Pudong New Area, Shanghai). cfDNA is extracted from patient plasma, and cancer-specific DNA methylation profile is analyzed. According to the results of plasma ctDNA methylation test, the risk of colorectal cancer of the enrolled subjects is scored, and combined with grouping information to assess the clinical application value of the cancer-specific methylation profile in ctDNA for early cancer and adenoma diagnosis. 3.4 End point determination The positive subjects are determined based on the results of surgical histopathological examination; the negative subjects are determined based on colonoscopy results. The ctDNA methylation profile is analyzed for positive and negative subjects.
4. Statistical analysis

4.1 Sample size estimation The results of this study are colorectal cancer-specific methylation signals from peripheral blood ctDNA. The methylation signals will all be superimposed with a certain weight, and the final score is in a single number form. Our preliminary results showed that the mean value of the control group (μ1) was 0.36, and the mean value of the test group (μ2) was 0.73. The mean difference between groups was δ = μ2 - μ1= 0.37, the standard deviation σ of the overall sample standard deviation s is 0.42. Set the check level α to 0.005 on both sides. The verification performance β is 0.005, and the t value table is queried to obtain t(α) = 2.807, t(β) = 2.576.

The following formula is used to calculate the sample size:

n = 2*((tα + tβ)*S/δ)2

The calculated result is n = 74.68, which is rounded to 75. Considering that the enrolled samples may have 25% unusable cases, the minimum test group sample size was set to 100 cases.

Positive group 1, colorectal cancer stage I-IV, no need to allocate groups in each sub-category. Considering balancing the number of samples between other groups, the sample size is set to 300 cases.

Positive group 2, advanced adenoma, according to the "Guidelines for the diagnosis and treatment of colorectal cancer", the advanced adenomas are divided into four categories. There is no need for equal distribution in different categories, with an average of 100 cases per category, for a total of 400 cases.

The negative group is divided into three categories. Non-advanced adenomas need to be differentiated from positive adenomas in positive group 2, and need a minimum sample size of 100 cases. Other categories do not require a specific sample size. Considering balancing with other groups the sample size is set to a total of 600 cases.

4.2 Statistics and data analysis Sequencing data for each sample was evaluated for sequencing quality using a variety of parameters, including whether the sequencing data reaches a preset level; whether the Q30 value is acceptable; whether the CpG site is balanced, whether the coverage is up to standard, whether the amplicon sequencing depth is as expected, and whether the sequencing uniformity is acceptable.

The cancer-specific DNA methylation signals in the sequencing results are summarized and analyzed according to the model established in our previous studies. Each signal value is superimposed with a certain weight to obtain a methylation score for each sample. Combined with the grouping information of each sample, PCA, clustering and other classification analysis are performed for each sample. The methylation signals are then converted to a colorectal cancer/adenoma risk score.

Based on the differences of risk scores between negative group and positive group, the sensitivity, specificity, and positive predicted values are calculated, and the clinical application efficacy of the cancer-specific methylation signal in plasma ctDNA is evaluated.