Pancreatic cancer screening method based on methylation scoring and grading

Through a method based on methylation scoring and grading, using methylated CpG island short tandem amplification sequencing and machine learning algorithms to screen key sites, establish a risk scoring formula, and solve the lack of sensitivity and specificity of pancreatic cancer screening methods. Problems, high-accuracy screening and grading have been achieved, false positives and false negatives have been reduced, and extended to screening for other diseases.

CN119061141APending Publication Date: 2024-12-03BEIJING SHIJITAN HOSPITAL CAPITAL MEDICAL UNIVERSITY
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411223585.3
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-09-03
Publication Date
2024-12-03

AI Technical Summary

Technical Problem

The existing pancreatic cancer screening methods are not sensitive and specific, difficult to diagnose early and accurately, and lack the ability to classify screening results, resulting in frequent occurrence of false positive and false negative results.

Method used

Using a method based on methylation scoring and grading, 14 key methylation sites were screened through methylated CpG island short tandem amplification sequencing and machine learning algorithms, and a risk scoring formula was established for screening of pancreatic cancer. and grading.

Benefits of technology

It significantly improves the accuracy and grading ability of pancreatic cancer screening, reduces false positive and false negative results, provides digital diagnostic assessment, facilitates comparison and analysis of results at different time points, and can be generalized to screening of other benign and malignant diseases. .

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119061141A_ABST
    Figure CN119061141A_ABST
Patent Text Reader

Abstract

The invention relates to the technical field of biological detection, and discloses a pancreatic cancer screening method based on methylation scoring and grading, and the pancreatic cancer screening method comprises the following steps: S1, scoring and grading; s2, a detection kit. According to the pancreatic cancer screening method based on methylation scoring and grading, pancreatic cancer patients are relatively comprehensively and accurately recognized, the system comprises 14 gene methylation sites, different methylation sites have different action coefficients, and a final result is obtained by adding products of the 14 sites and the coefficients of the 14 sites. And the occurrence frequency of false positive and false negative results is greatly reduced.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of biological detection technology, and in particular to a pancreatic cancer screening method based on methylation scoring and grading. Background Art

[0002] Pancreatic adenocarcinoma (PAAD) is the 12th most common cancer in the world, and its disease-related mortality ranks fourth among malignant tumors. It is known as the "king of cancers." Since the early symptoms of pancreatic cancer are not obvious, generally only manifesting as mild abdominal pain or poor blood sugar control, it is not easy to attract people's attention. Imaging examinations are also difficult to diagnose early pancreatic cancer. Therefore, most patients are diagnosed only when they develop serious symptoms. At this time, pancreatic cancer has often developed to the middle and late stages, which greatly affects the patient's prognosis.

[0003] As the only serum biomarker approved by the U.S. Food and Drug Administration (FDA), carbohydrate antigen-199 (CA-199) is currently the most commonly used diagnostic marker for pancreatic cancer in clinical practice, but its sensitivity (80%, 95% confidence interval [CI] = 72-86%) and specificity (75%, 95% CI = 68%-80%) are not high. The effects of other new serum biomarkers, such as mucin 5AC (MUC5AC), used alone or in combination with CA-199 remain to be observed. Using a single marker to screen pancreatic cancer is not comprehensive, and the workload of combined screening of multiple markers will be magnified several times. It is crucial to achieve a relatively balanced result in the selection of the quantity and quality of markers. Most of the current research focuses on the combined use of markers, but does not give the contribution ratio and effect size of each marker in the screening process, resulting in frequent false positive and false negative results.

[0004] However, tumor screening is different from benign diseases. It is not a simple binary result of yes or no. While screening, we need to grade the screening results and make different diagnosis and treatment suggestions for high-risk groups according to the severity of the results. So far, no published results or patents have made progress in these aspects. Therefore, a pancreatic cancer screening method based on methylation scoring and grading is proposed to solve the above problems. Summary of the invention

[0005] The technical problem to be solved by the present invention is to provide a pancreatic cancer screening method based on methylation scoring and grading in view of the deficiencies in the above-mentioned prior art.

[0006] In order to solve the above technical problems, the technical solution adopted by the present invention is: a pancreatic cancer screening method based on methylation scoring and grading, comprising the following steps:

[0007] S1. Scoring and grading methods;

[0008] S1.1. Prepare two groups of experimental cases, including pancreatic cancer patients and normal controls, each group of experimental cases includes 33 experimental tissues and 50 experimental plasmas. All samples were randomly divided into two groups, training cohort and validation cohort, in a ratio of 8:2 in advance;

[0009] S1.2. Perform short tandem amplification sequencing of methylated CpG islands on two groups of experimental cases in the training cohort at the same time, conduct comparative analysis to find differentially methylated sites, and calculate the result data of differentially methylated sites;

[0010] S1.3. In the result data of differentially methylated sites in S1.2, further screen the gene methylation sites that can be used for screening according to the additive positive method;

[0011] S1.4. Through machine learning algorithms, select the optimal methylation frequency combination and the optimal methylation site as markers for subsequent disease diagnosis and screening, and establish a disease rating formula;

[0012] S1.5. Further verify the effect of gene methylation sites in S1.3 on distinguishing pancreatic cancer patients from normal subjects in the validation cohort;

[0013] S2, detection kit;

[0014] The detection steps and specific implementation process are mainly based on three detection kits, kit A, kit B, and kit C;

[0015] S2.1. Sample collection: A total of 33 samples of cancer tissues and paired normal tissues were collected, and 50 samples of plasma were collected from pancreatic cancer patients and normal controls;

[0016] S2.2, DNA extraction, bisulfite conversion, and library expansion;

[0017] S2.3. DNA sequencing library purification.

[0018] Preferably, in S1.2, a total of 4615 differentially methylated CGCGCGG-CpGs were identified (two-tailed Mann-Whitney U test, false discovery rate <0.05, average methylation fold change>2). In addition, principal component analysis also showed that most PAAD cancer patients (29 out of 33 cases) could be distinguished from normal controls. These two analysis steps were completed by the R packages "limma", "tidyverse" and "ggplot2".

[0019] Preferably, in S1.3, among the above 4615 differentially methylated sites, gene methylation sites that can be used for screening are further screened according to the additive positive method.Figure 1 Figure A is the additive positive method. All sequencing sites are divided into fre0-fre10, a total of 6 groups. Fre0 means that the frequency of this group of gene sites in normal human plasma is 0, and fre10 means that the frequency of this group of gene sites in normal human plasma is 10%. The Brier score on the left represents the judgment error rate. The smaller the value, the smaller the error rate, and the better the effect of distinguishing cancer patients from normal people. The results show that in the fre0 group, the diagnostic combination composed of 120 sites has the lowest error rate; Figure 1 Figure B is a comparison of the expression differences of the 120 gene methylation sites screened out in cancer tissues and normal tissues.

[0020] The genomic information of the following 120 methylation sites was screened by the additive positive method:

[0021]

[0022]

[0023]

[0024]

[0025]

[0026]

[0027]

[0028]

[0029] Preferably, S1.4 includes: using the LASSO regression algorithm to select the optimal methylation frequency combination and the optimal methylation site, and using them as markers for subsequent disease diagnosis and screening. By fitting 14 preferred methylated CpG sites through logistic regression, a pancreatic cancer methylation score and grading formula is established: Risk score = -23.35 + 2.84 * TWIST2-17229_1_1 + 1.23 * ST8SIA3-9198_2_3–0.69 * SLC10A4-19814_1_4 + 1.74 * PDE10A-21786_2_1–0.70 * OTX2-4797_1_9–33.95 * FBXL21-2037 6_1_4+19.06*KCNS1-14770_1_3–0.54*17547_1_2+3.60*18198_1_6+10.57*19013 _2_3+6.06*19330_5_3+7.24*19348_2_3+16.04*23126_3_5+0.79*LHX5-2647_1_2.

[0030] Preferably, S1.5 includes: respectively verifying the effect of the above 120 gene methylation sites on pancreatic cancer patients and normal subjects in the validation cohort and observing them in the training cohort, and evaluating the diagnostic performance by calculating the number of positives and the area under the curve (AUC) value. The results in the training cohort are shown (e.g. Figure 2 As shown in A in the figure, the positive counts of 120 markers in both early and late PAAD patients were significantly increased compared with controls (the median of the control group was 0, and the median of stages I & II and III & IV were 37 and 31, respectively, P < 0.0001, two-tailed Mann-Whitney U test). ROC curve analysis showed (as shown in Figure 2 In the training group, the AUC value of CA-199 was 0.85 (95% CI = 0.77-0.93), with a sensitivity and specificity of 70% and 100%. In the validation cohort (as shown in Figure 2B), the AUC value of the combination of 120 mCGCGCGG-CpG for diagnosing PAAD was 0.99 (95% CI = 0.99-1.00), with a sensitivity and specificity of 97% and 100%. Figure 2As shown in C in the figure, the number of positive mCGCGCGG-CpG in pancreatic cancer plasma was significantly increased compared with the control group (the median number of positive numbers in patients with stage I & II and stage III & IV were 29 and 34, respectively, and 0 in the control group, P < 0.001, two-tailed Mann-Whitney U test). The AUC values ​​of the diagnostic combination of these 120 mCGCGCGG-CpG and CA-199 were 0.98 (95% CI = 0.96-1.00) and 0.85 (95% CI = 0.75-0.95), respectively. The sensitivity and specificity were 86%, 70%, 100%, and 100%, respectively (as shown in C in the figure). Figure 2 After merging the training set and the validation set (as shown in D in Figure 2 (As shown in Figures E and F), the AUC value of the combination of 120 mCGCGCGG-CpGs for diagnosing PAAD was 0.99 (95% CI = 0.99-1.00), and the AUC value of CA-199 for diagnosing PAAD was 0.85 (95% CI = 0.77-0.93). The sensitivity and specificity were 92%, 78%, 98%, and 100%, respectively.

[0031] Preferably, S2.1 includes: first, sample collection, 33 cancer tissues and paired normal tissues were collected, and 50 plasmas were collected from pancreatic cancer patients and normal controls. Each tissue specimen weighing about 20 mg was washed with physiological saline and directly frozen in a -80 degree refrigerator, and the centrifugation steps of the plasma sample were as follows: 10 ml of peripheral venous blood was collected in an anticoagulant tube containing ethylenediaminetetraacetic acid (EDTA), which can be temporarily placed in a 4°C refrigerator for storage, but the centrifugation must be completed within 6 hours to separate blood cells and plasma. The blood sample was centrifuged at a low speed of 1350g for 13 minutes, and the plasma layer, white blood cells and platelet layer, and red blood cell layer were separated from top to bottom. After the plasma layer was completely sucked out, it was centrifuged again at 1350g for 13 minutes, and then the plasma except the centrifugal precipitate was sucked out again and centrifuged at 13500g for 5 minutes. The separated plasma was numbered and archived and immediately frozen at -80°C for storage.

[0032] Preferably, S2.2 includes: tissue specimen DNA extraction uses consumable No. 1 in kit A, and plasma free DNA extraction uses consumable No. 2 in kit A, and the operations are respectively performed according to the instructions; the extracted DNA needs to be converted to bisulfite, during the conversion process, the methylated cytosine C remains unchanged, and the unmethylated cytosine C is converted to uracil U, and this step uses consumable No. 3 in kit A; library expansion is carried out according to the MCTA-Seq technical process, which is divided into three steps in total, the first step is PCR reaction using Primer A in kit B and NE Buffer 2 and Klenow Fragment in kit C, which can specifically amplify the methylated CpG region in DNA, the second step is PCR library construction reaction using Primer B in kit B, and further selecting DNA fragments starting with CGCGCGG, and the third step is PCR library construction reaction using Primer C and Primer D in kit B, and adding sequencing adapter sequences.

[0033] Preferably, S2.3 includes: at least 3 purifications are required before sequencing the DNA library to ensure that the content of impurity DNA is low and does not occupy sequencing space. The amplification PCR reaction in this step requires the use of Primer Q-P1 and Primer Q-P2 in Kit B, and the amplified product uses agarose gel electrophoresis to obtain a band in the range of 180-250bp (using DL2000 DNAMarker and 20bp DNA Ladder in Kit C to provide molecular weight standards), and agarose gel recovery and DNA cleaning and purification require the use of consumables No. 5 and No. 4 in Kit A. Low melting point agarose powder and TAE buffer must be used when configuring agarose electrophoresis fluid, and SYBR Safe DNA Gel Stain is used as the gel developer.

[0034] The present invention adopts the above technical solution to bring the following beneficial effects:

[0035] 1. This pancreatic cancer screening method based on methylation scoring and grading can relatively comprehensively and accurately identify pancreatic cancer patients. This system contains 14 gene methylation sites. Different methylation sites have different action coefficients. The final result is obtained by adding the product of these 14 sites and their coefficients, which greatly reduces the frequency of false positive and false negative results.

[0036] 2. A pancreatic cancer screening method based on methylation scoring and grading, which grades the risk level of the subjects according to the screening results. Subjects with negative screening results can be excluded from the risk of disease, subjects with positive screening results but low grades need to be followed up for observation, and subjects with positive screening results and high grades need to be clinically diagnosed as soon as possible;

[0037] 3. This pancreatic cancer screening method based on methylation scoring and grading has digital screening results. It does not simply give a binary judgment of yes or no. It gives a digital assessment of the possibility of pancreatic cancer. The results are easy to read and stable, which is convenient for before and after comparison analysis at different time points.

[0038] 4. This pancreatic cancer screening method based on methylation scoring and grading can be extended to the high-risk screening of other benign and malignant diseases, such as liver cancer, gastric cancer, colorectal adenoma, etc. It is applicable to a wide spectrum of diseases and has diverse screening methods. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] Figure 1 It is a first schematic diagram of the present invention;

[0040] Figure 2 is a second schematic diagram of the present invention;

[0041] Figure 3 is a third schematic diagram of the present invention;

[0042] Figure 4 is a fourth schematic diagram of the present invention;

[0043] Figure 5 is a fifth schematic diagram of the present invention;

[0044] Figure 6 This is the sixth schematic diagram of the present invention. DETAILED DESCRIPTION

[0045] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0046] The specific implementation process of how to perform pancreatic cancer methylation scoring and grading on a new patient with an unknown disease state or a healthy person is further described:

[0047] First, the peripheral blood samples of the subjects were centrifuged to separate the plasma, and then the free DNA in the plasma was extracted, bisulfite converted, the three-step PCR library construction reaction of DNA, and the library expansion and purification were completed in sequence. When the library DNA concentration was >2.0ng / ul, it was sent out for methylation group sequencing. The sequencing instrument was Illumina HiSeq 2000 / 2500 sequencer. After the sequencing data was returned, the data was first quality controlled and cleaned to remove useless adapter sequences, and the uMEPM values ​​of the 10 CpG sites immediately following CGCGCGG were calculated according to the FMM (Fully Methylated Molecular) analysis process. The methylation degree of the gene site in plasma is expressed as the uMEPM (unique methylated alleles per million mapped reads) value.

[0048] Secondly, the uMEPM values ​​of the 120 CpG sites screened by the additive positive method are used to directly judge the risk of pancreatic cancer in the subject. Compared with normal healthy people, high methylation levels of one or more of the sites indicate that the individual is very likely to suffer from pancreatic cancer.

[0049] In addition, the methylation score is calculated according to the scoring formula established by the 14 CpG sites. The total score is the sum of the coefficients of the 14 CpG sites multiplied by the uMEPM value. If the total score is greater than 0, it means that the subject is at extremely high risk of pancreatic cancer, close to 100% clinical diagnosis; if the total score is less than or equal to 0, the subject can almost completely rule out the risk of pancreatic cancer, and the accuracy rate is also close to 100%. For subjects with a total score greater than 0 who are at extremely high risk of pancreatic cancer, we grade their risk level according to the numerical value of the total score. Figure 3 As shown in Figure D, when the total score is between 0 and 150, it can be estimated that pancreatic cancer is in clinical stage I and II, which is an early stage of the disease and can generally be effectively treated with surgery. When the total score is greater than 150, it can be estimated that the subject's pancreatic cancer is in clinical stage III and IV, which is an advanced stage of the disease. It is difficult to cure with surgery alone, and chemotherapy or immunotherapy is required after surgery. In particular, Figure 3 As shown in Figure C, the non-zero number of the 14 CpG sites that make up the scoring system also has obvious differences in stage I, II and stage III, IV of the disease. When the non-zero number (i.e., the number of positive CpG sites) is less than or equal to 8, the subject is likely to be in stage I and stage II of the disease. When the non-zero number is greater than 8, the subject is likely to be in the late stage of the disease, i.e., stage III or stage IV. In this way, we have completed the grading based on pancreatic cancer methylation score, which can be effectively used for the screening and evaluation of pancreatic cancer.

[0050] Implementation Case 1: A subject (id=80) was enrolled. 10 ml of peripheral blood was collected from the subject upon admission. After the peripheral blood free DNA was extracted, a methylation sequencing library was constructed according to the MCTA-Seq technical process. After the sequencing results were returned, the uMEPM values ​​of the 14 screened methylated CpG sites were substituted into the above risk score formula: Risk score=-23.35+2.84*4.52+1.23*0.65–0.69*0+1.74*1.94–0.70*0–33.95*0.65+19.06*1.30–0.54*3.23+3.60*1.30+10.57*1.30+6.06*1.30+7.24*20.01+16.04*1.93+0.79*0=196.76. A risk score greater than 0 indicates that the subject is very likely to be a pancreatic cancer patient, and the total score is greater than 150. The TNM stage of pancreatic cancer is very likely to be in stage III or IV. In addition, there are only 3 "0" values ​​in the 14 CpG sites, which further proves that the stage of pancreatic cancer in the subject is already in the advanced stage. These results are consistent with the clinical pathological data of the subject.

[0051] Implementation Case 2: A subject (id=60) was enrolled. 10 ml of peripheral blood was collected from the subject upon enrollment. After free DNA from the peripheral blood was extracted, a methylation sequencing library was constructed according to the MCTA-Seq technical process. After the sequencing results were returned, the uMEPM values ​​of the 14 screened methylated CpG sites were substituted into the above risk score formula: Risk score=-23.35+2.84*0+1.23*0–0.69*0+1.74*0–0.70*0–33.95*0+19.06*0–0.54*4.19+3.60*0+10.57*0+6.06*0+7.24*0+16.04*0+0.79*0=-25.61. A risk score less than 0 indicates that the subject has a nearly 0 probability of developing pancreatic cancer. The evaluation result is consistent with the actual situation of the subject.

[0052] It is worth noting that methylated CpG island short tandem amplification sequencing (MCTA-Seq) is an existing technology, which includes PrimerA—Primer D as well as Primer Q-P1 and Primer Q-P2, and will not be elaborated on here.

[0053] The present invention provides a pancreatic cancer screening method based on methylation scoring and grading. There are many methods and approaches to implement the technical solution. The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications should also be regarded as the protection scope of the present invention. All components not specified in this embodiment can be implemented by existing technologies.

Claims

1. A pancreatic cancer screening method based on methylation scoring and grading, characterized in that: The following steps are involved: S1. Scoring and grading methods; S1.

1. Prepare two groups of experimental cases, including pancreatic cancer patients and normal controls, each group of experimental cases includes 33 experimental tissues and 50 experimental plasmas. All samples were randomly divided into two groups, training cohort and validation cohort, in a ratio of 8:2 in advance; S1.

2. Perform short tandem amplification sequencing of methylated CpG islands on two groups of experimental cases in the training cohort at the same time, conduct comparative analysis to find differentially methylated sites, and calculate the result data of differentially methylated sites; S1.

3. In the result data of differentially methylated sites in S1.2, further screen the gene methylation sites that can be used for screening according to the additive positive method; S1.

4. Through machine learning algorithms, select the optimal methylation frequency combination and the optimal methylation site as markers for subsequent disease diagnosis and screening, and establish disease scoring and grading formulas; S1.

5. Further verify the effect of gene methylation sites in S1.3 on distinguishing pancreatic cancer patients from normal subjects in the validation cohort; S2, detection kit; The detection steps and specific implementation process are mainly based on three detection kits, kit A, kit B, and kit C; S2.

1. Sample collection: A total of 33 samples of cancer tissues and paired normal tissues were collected, and 50 samples of plasma were collected from pancreatic cancer patients and normal controls; S2.2, DNA extraction, bisulfite conversion, and library expansion; S2.

3. DNA sequencing library purification.

2. A pancreatic cancer screening method based on methylation scoring and grading according to claim 1, characterized in that: In S1.2, a total of 4615 differentially methylated CGCGCGG-CpGs were identified, and principal component analysis showed that most pancreatic cancer patients could be distinguished from normal controls.

3. A pancreatic cancer screening method based on methylation scoring and grading according to claim 2, characterized in that: The S1.3 specifically includes: taking the plasma sequencing results of normal controls as a reference, directly calculating the different distribution and different frequencies of each CpG site in the two groups of subjects to determine whether the methylation site is associated with the disease diagnosis, and the methylation levels of 120 methylation sites obtained by the additive positive method can be used for the diagnosis of pancreatic cancer.

4. A pancreatic cancer screening method based on methylation scoring and grading according to claim 3, characterized in that: In S1.4, the least absolute convergence and selection operator regression method was used to select 14 optimal methylation sites to form a disease rating model, and a disease scoring and grading formula was established.

5. A pancreatic cancer screening method based on methylation scoring and grading according to claim 4, characterized in that: The S1.5 includes: further verifying the discrimination effect of the above 120 gene methylation sites on pancreatic cancer patients and normal subjects in the validation cohort, and evaluating the discrimination performance by calculating the number of positives and the area under the curve.

6. A pancreatic cancer screening method based on methylation scoring and grading according to claim 1, characterized in that: The S2.1 includes: each tissue specimen weighing about 20 mg is washed with physiological saline and directly frozen in a -80°C refrigerator. The blood sample is required to be at least 10 ml of peripheral blood. The separated plasma must be numbered, filed and stored in a -80°C refrigerator after centrifugation within 6 hours.

7. A pancreatic cancer screening method based on methylation scoring and grading according to claim 6, characterized in that: The S2.2 includes: tissue specimen DNA extraction uses consumable No. 1 in kit A, and plasma free DNA extraction uses consumable No. 2 in kit A; the extracted DNA needs to be converted to bisulfite, and this step uses consumable No. 3 in kit A; library expansion is carried out according to the MCTA-Seq technical process, which is divided into three steps in total. The first step of PCR reaction uses Primer A in kit B and NE Buffer 2 and Klenow Fragment in kit C, the second step of PCR reaction uses Primer B in kit B, and the third step of PCR reaction uses Primer C and Primer D in kit B.

8. A pancreatic cancer screening method based on methylation scoring and grading according to claim 7, characterized in that: The S2.3 includes: the DNA library needs to be purified at least 3 times before sequencing, the amplification reaction in this step needs to use Primer Q-P1 and Primer Q-P2 in Kit B, the amplification product is electrophoresed by agarose gel to obtain a band in the range of 180-250bp, and agarose gel recovery and DNA cleaning and purification require the use of consumables No. 5 and No. 4 in Kit A.