A non-invasive intelligent diagnostic kit for early colorectal cancer based on peripheral blood and its application

Abstract

This invention relates to the field of early cancer diagnostic reagents, specifically to a non-invasive intelligent diagnostic kit for early colorectal cancer based on peripheral blood and its application. The kit targets key gene mutation sites (MLH1 c.113del A, MSH2 c.788del A, MUTYH c.1005 G>C, PMS2 c.288 C>T, and c.780 C>G) screened in the Chinese population and includes a specific primer set, circulating tumor DNA (ctDNA) extraction, and sequencing library preparation reagents. The accompanying detection system analyzes mutation data using a logistic regression model trained on the aforementioned sites to achieve early screening. Clinical trials have shown that this method has good performance in early colorectal cancer screening in the Chinese population.

Description

Technical Field

[0001] This invention relates to the field of early cancer diagnostic reagents, specifically to a non-invasive intelligent diagnostic reagent kit for early colorectal cancer based on peripheral blood and its application. Background Technology

[0002] Early diagnosis of colorectal cancer can significantly improve patient survival rates, therefore, developing effective early screening methods is of great clinical significance. Currently, colonoscopy is the primary clinical diagnostic tool; however, it is an invasive procedure, easily leading to risks such as bleeding and infection, and patient compliance is low, resulting in low coverage. Furthermore, colonoscopy combined with immunohistochemical detection has limitations in sensitivity and specificity, and the consistency of detection for lesions in different segments of the colon needs improvement, thus restricting its widespread application in early screening.

[0003] In recent years, liquid biopsy technology based on circulating tumor DNA (ctDNA) has provided a new avenue for non-invasive early cancer diagnosis. ctDNA in the blood carries tumor-specific gene mutations, and detecting its mutation profile can provide early warning of cancer development. Studies have shown that mutations in DNA mismatch repair genes (such as MLH1, MSH2, and PMS2) and the oxidative damage repair gene MUTYH are closely related to the occurrence of colorectal cancer. However, mutation profiles vary among different populations, and currently, there is a lack of specific mutation combinations and corresponding detection products for early colorectal cancer in the Chinese population.

[0004] Therefore, it is necessary to develop a non-invasive diagnostic kit for early colorectal cancer based on peripheral blood ctDNA, suitable for the Chinese population, to make up for the shortcomings of existing technologies and improve the early detection rate and screening accessibility. Summary of the Invention

[0005] The present invention aims to provide a non-invasive diagnostic kit for early colorectal cancer in the Chinese population, along with a matching detection system and evaluation method, to address the technical problems of existing colonoscopy methods being highly invasive, having low patient compliance, and lacking specificity for the Chinese population.

[0006] To achieve the above objectives, the present invention adopts the following technical solution: In a first aspect, the present invention provides a primer set for the early diagnosis of colorectal cancer, comprising the following five pairs of primers, each pair of primers being used to specifically amplify a corresponding colorectal cancer-related gene target region: (1) Primer pair used to amplify the c.113 site region of the MLH1 gene, the sequences of which are shown in SEQ ID NO:1 and SEQ ID NO:2; (2) Primer pairs used to amplify the c.788 site region of the MSH2 gene, the sequences of which are shown in SEQ ID NO:3 and SEQ ID NO:4; (3) Primer pairs used to amplify the c.1005 site region of the MUTYH gene, the sequences of which are shown in SEQ ID NO:5 and SEQ ID NO:6; (4) Primer pairs used to amplify the c.288 region of the PMS2 gene, the sequences of which are shown in SEQ ID NO:7 and SEQ ID NO:8; (5) Primer pairs used to amplify the c.780 region of the PMS2 gene, the sequences of which are shown in SEQ ID NO:9 and SEQ ID NO:10.

[0007] Secondly, this invention provides a kit for early diagnosis of colorectal cancer containing the aforementioned primer set. This kit can be used to detect at least one of the following mutations that are frequently detected in colorectal cancer patients in the Chinese population: MLH1c.113del A, MSH2 c.788del A, MUTYH c.1005 G>C, PMS2 c.288 C>T, and PMS2 c.780 C>G. Furthermore, the kit may also include a reagent module for extracting circulating tumor DNA (ctDNA) from peripheral blood samples and a reagent module for sequencing and library preparation of the amplified products, thus constituting a complete detection solution from sample to sequencing results. This kit is suitable for in vitro detection of early colorectal cancer screening or risk assessment in the Chinese population.

[0008] Thirdly, the present invention provides a gene mutation data analysis system, comprising: a data input interface for receiving genotype data associated with a set of predetermined gene loci, wherein the set of predetermined gene loci are MLH1 gene c.113, MSH2 gene c.788, MUTYH gene c.1005, PMS2 gene c.288, and PMS2 gene c.780; a processor; and a memory coupled to the processor and storing a computer program, wherein when the computer program is executed by the processor, the system performs the following operations: based on the genotype data received through the data input interface corresponding to the set of predetermined gene loci, a risk score is generated and the risk score is output using a pre-stored logistic regression prediction model trained based on the association between the mutation status of the set of predetermined gene loci and the risk of colorectal cancer.

[0009] Fourthly, the present invention provides a computer-implemented method for analyzing gene mutation data, comprising: acquiring genotype data associated with a set of predetermined gene loci, wherein the set of predetermined gene loci are MLH1 gene c.113, MSH2 gene c.788, MUTYH gene c.1005, PMS2 gene c.288, and PMS2 gene c.780; inputting the acquired genotype data into a pre-trained logistic regression model trained based on the association between the mutation status of the set of predetermined gene loci and the risk of colorectal cancer in the Chinese population, to generate a risk score and output the risk score.

[0010] The beneficial effects of this invention are as follows: 1. Non-invasive and high compliance: By collecting peripheral blood for ctDNA testing, a completely non-invasive early screening is achieved, avoiding the invasive risks and discomfort of colonoscopy, which helps to improve screening compliance.

[0011] 2. Accuracy and Population Specificity: Based on the molecular characteristics of colorectal cancer in the Chinese population, five high-frequency mutation site combinations from four key genes were screened and validated. Joint detection and comprehensive analysis of these multi-site mutation profiles improved the accuracy of early diagnosis.

[0012] 3. Intelligent and Integrated Approach: By combining liquid biopsy with algorithmic models, an integrated analytical workflow has been constructed, encompassing ctDNA extraction, targeted sequencing, and risk assessment. This workflow is suitable for early screening and risk stratification management of large-scale populations. Attached Figure Description

[0013] Figure 1 Comparison of sequencing peaks at the c.113 site of the MLH1 gene: (a) wild type (b) mutant (c.113 del A). Figure 2 Comparison of sequencing peaks at the c.788 site of the MSH2 gene: (a) wild type (b) mutant (c.788 del A); Figure 3 Comparison of sequencing peaks at the c.1005 site of the MUTYH gene: (a) wild type (b) mutant (c.1005 G>C); Figure 4 Comparison of sequencing peaks at the c.288 site of the PMS2 gene: (a) wild type (b) mutant (c.288 C>T); Figure 5 Comparison of sequencing peaks at the c.780 site of the PMS2 gene: (a) wild type (b) mutant (c.780 C>G); Figure 6 A nomogram model for predicting colorectal cancer risk based on five mutation sites; Figure 7 Modeling set for ROC curves of model performance; Figure 8 The ROC curve validation set is used to evaluate the model's performance. Figure 9 A summary chart of gene mutation detection results for the independent validation set samples. Detailed Implementation

[0014] The present invention is illustrated by the following embodiments, but the scope of the invention is not limited thereto. Unless otherwise specified, the methods described in this invention employ conventional experimental techniques in the art, and the reagents and materials used are commercially available.

[0015] It should be noted that the primer set provided by this invention is used to specifically detect five key mutations: MLH1 c.113delA, MSH2 c.788delA, MUTYH c.1005 G>C, PMS2 c.288 C>T, and PMS2 c.780 C>G. In the gene mutation data analysis system and method, the logistic regression prediction model is specifically trained based on the association between the mutation status of the above five specific gene loci and the risk of colorectal cancer, and is an integral part of the intelligent analysis in the overall technical solution of this invention.

[0016] Raw materials and reagents The circulating tumor DNA (ctDNA) extraction module includes cell lysis buffer (such as a solution containing dissociation salts and surfactants, such as GuHCl or SDS), proteinase K, a silica membrane adsorption column or magnetic beads, a wash buffer (such as a buffer containing ethanol), and an elution buffer (such as TE buffer or sterile water). As a specific example, a commercially available circulating nucleic acid extraction kit (such as the QIAamp Circulating Nucleic Acid Kit) can be used, following its instructions.

[0017] Sequencing library preparation module: High-throughput sequencing library construction kits for the Illumina platform can be used, such as commercially available high-throughput sequencing library construction kits for the Illumina platform. These kits typically contain all the necessary components, including DNA end repair enzymes, DNA ligases, high-fidelity DNA polymerases, buffers, and adapters.

[0018] Primer synthesis: Primers with sequences as shown in SEQ ID NO:1-10 can be synthesized by any commercial service provider with DNA synthesis capabilities. The technical solution of this invention can be achieved by using the above-described or functionally equivalent commercial reagents, following their published standard operating procedures, and adapting them to the specific primer set and detection process provided by this invention.

[0019] Example 1: A kit for early diagnosis of colorectal cancer. This example provides the specific structure of the kit described in this invention.

[0020] 1. Detection of target and primer set This kit is designed to detect five high-frequency mutation sites in four key genes in colorectal cancer in the Chinese population, as shown in the table below: Table 1 High-frequency mutation sites

[0021] The specific primer sequences used to amplify the above target regions are shown in the table below: Table 2 Specific primer sequences

[0022] 2. Kit Components This kit contains the following three core modules: Circulating tumor DNA (ctDNA) extraction module: includes lysis buffer GHH, proteinase K, magnetic beads, wash buffer PWG, and elution buffer TBC.

[0023] Multiplex PCR amplification module: contains the primer mixture shown in SEQ ID NO:1-10 above, 5×FEA EnzymeMix (containing End Repair enzyme, Klenow 3'→5'exo-, T4 DNA ligase, Phusion high-fidelity polymerase), 10×FEA Reaction Buffer, and FEA Enhancer.

[0024] High-throughput sequencing library preparation module: contains universal adapters, index tags, and related buffer systems required for sequencing on the Illumina platform.

[0025] Example 2: A method for detecting colorectal cancer mutations based on peripheral blood ctDNA This embodiment details the operation procedure for detection using the kit described in Example 1.

[0026] 1. Sample processing and ctDNA extraction Collect 2 ml of peripheral venous blood from the subject into an EDTA anticoagulant tube and centrifuge at 1600×g for 10 minutes at 4°C. Carefully aspirate the cell-free plasma layer. Take 1 ml of plasma and add 1.5 ml of lysis buffer (GHH), 100 μl of proteinase K, and 25 μl of activated magnetic beads. Vortex to mix. Incubate at 25°C for 15-20 minutes, gently inverting several times during incubation. Place the centrifuge tube on a magnetic rack and let it stand for 1 minute until the magnetic beads are completely adsorbed. Discard all liquid. Resuspend the magnetic beads in 500 μl of buffer (GDF) and transfer to a new 1.5 ml centrifuge tube. Discard the liquid again using magnetic aspiration. Wash the magnetic beads twice with 500 μl of wash buffer (PWG), discarding the wash buffer completely each time. Open the tube and allow the magnetic beads to air dry at room temperature for 5-10 minutes. Add 35 μl of elution buffer (TBC) and incubate at 56°C for 5 minutes, gently tapping the tube walls during incubation. After magnetic aspiration, the supernatant containing ctDNA was transferred to a new tube and stored at -20°C for later use.

[0027] 2. Target region amplification and sequencing library construction First, a 50 μl PCR reaction mixture was prepared. The specific steps were: 2 μl primer mixture (Example 1), 10 μl 5×FEAEnzyme Mix, 5 μl 10×FEA Reaction Buffer, 5 μl FEA Enhancer, 5 μl template ctDNA, and Nuclease-Free water to a final volume of 50 μl. The PCR reaction program was then performed: 95°C pre-denaturation for 3 minutes; followed by 28 cycles (95°C denaturation for 30 seconds, 60°C annealing for 30 seconds, 72°C extension for 30 seconds); and a final extension at 72°C for 5 minutes. After purification, the PCR product was added with Illumina sequencing adapters and indexes according to standard procedures to construct a sequencing library.

[0028] 3. High-throughput sequencing and mutation analysis The library was sequenced at 150 bp paired ends on the Illumina NovaSeq platform. The sequencing data were aligned to the human reference genome (hg19) using BWA software, and variant detection was performed using tools such as GATK. The genotypes (wild-type or mutant) of the five loci (MLH1 c.113, MSH2 c.788, MUTYH c.1005, PMS2 c.288, PMS2 c.780) specified in Example 1 were finally output. A comparison of typical sequencing results for the five key mutation sites is shown in the figure below. Figures 1 to 5 As shown:

[0029] Figure 1 The sequencing peak diagrams of the c.113 site of the MLH1 gene are shown in comparison, where (a) is a clear single peak of wild type and (b) is a frameshift peak diagram caused by c.113del A, which confirms the existence of this deletion mutation.

[0030] Figure 2 The sequencing peak diagrams of the c.788 site of the MSH2 gene are shown in comparison. (a) is the wild-type sequence, and (b) shows del A at the c.788 site, which visually proves the frameshift mutation.

[0031] Figure 3 The sequencing peak diagrams of the c.1005 site of the MUTYH gene are shown in comparison. (a) shows the wild-type G base peak, and (b) clearly shows the C base peak, which proves the base substitution mutation of G>C in c.1005.

[0032] Figure 4 The sequencing peak diagrams of the PMS2 gene c.288 site are shown in comparison, where (a) shows the wild-type C base peak and (b) shows the T base peak, confirming the c.288 C>T mutation.

[0033] Figure 5 The sequencing peak diagrams of the PMS2 gene c.780 site are shown in comparison, where (a) shows the wild-type C base peak and (b) shows the G base peak, confirming the c.780 C>G mutation.

[0034] Example 3: An intelligent diagnostic system for colorectal cancer and its application This embodiment provides a risk assessment system that is compatible with the reagent kit and its usage method.

[0035] 1. System Construction We collected 129 clinical samples, including 107 pathologically confirmed colorectal cancer patients and 22 healthy controls, and their data and clinical diagnostic gold standard labels were obtained. The samples were randomly divided into a training set (90 cases) and a validation set (39 cases) at a ratio of 7:3. Using five mutation sites as characteristics (mutated as 1, wild-type as 0) and clinical diagnostic results (cancer as 1, healthy as 0) as labels, we performed multivariate logistic regression analysis using the training set data. The regression coefficients are shown in the table below.

[0036] Table 3 Results of Logistic Regression Analysis

[0037] Based on the results of the above logistic regression analysis, a predictive nomogram is constructed. This nomogram converts the regression coefficients of each mutation variable into an intuitive scoring scale, and obtains a total score by summing the scores of each variable, thereby mapping the corresponding individual's disease risk probability. The nomogram model is as follows: Figure 6 As shown, it provides a practical tool that quantifies the mutation status of five sites into an individual's total score and directly correlates it with the probability of specific disease risk.

[0038] 2. System Application The mutation status of the five sites of the new sample obtained in Example 2 is input into the system. The system calculates the risk value of colorectal cancer by comprehensively scoring the mutation characteristics of the sample based on a pre-trained Logistic regression model or the corresponding nomogram. The system outputs a diagnostic conclusion of "high risk of colorectal cancer" or "low risk of colorectal cancer" based on a preset risk threshold.

[0039] Example 4: Performance verification of the reagent kit and system of the present invention. This example verifies the effectiveness of the present invention through clinical trial data.

[0040] 1. Model Performance Evaluation Receiver Operating Characteristic (ROC) curves were plotted on the training set and the independent validation set. The results showed that the area under the curve (AUC) for the training set was 0.905 (95% CI: 0.841–0.962, P<0.001), and the AUC for the validation set was 0.843 (95% CI: 0.812–0.889, P<0.001). The AUC values ​​for the training and validation sets were 0.905 and 0.843, respectively, indicating that the model has discriminative power. The ROC curves for the modeling set and the validation set are shown below. Figure 7 and Figure 8 As shown, both curves deviate significantly from the diagonal, and the AUC values ​​are relatively high.

[0041] 2. Verification of clinical diagnostic accuracy Using the system and thresholds described in Example 3, a blinded analysis was performed on the independent validation set (39 cases, including 32 patients and 7 controls). The specific detection results of the five key mutation sites in the validation set samples are as follows: Figure 9 As shown in the heatmap, the presence of single-site or multi-site mutations is clearly evident in the colorectal cancer patient group (sample), while the healthy control group showed virtually no mutations. The multi-site mutation pattern was consistent with the clinical diagnosis. Ultimately, the model correctly identified 30 patients and all 7 controls, achieving an overall diagnostic accuracy of 94.87% (37 / 39), a sensitivity of 96.9%, and a specificity of 100%. These results demonstrate that the kit and data analysis system exhibited an overall diagnostic accuracy of 94.87% in the validation set.

[0042] The above embodiments are specific descriptions of the present invention. The scope of protection of the present invention is not limited to the embodiments described. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A primer set for early diagnosis of colorectal cancer, characterized by, The kit comprises the following five pairs of primers, each pair of primers being used for specifically amplifying a corresponding colorectal cancer related gene target region: (1) a primer pair for amplifying the region of MLH1 gene c.113 site, the sequences of which are shown in SEQ ID NO: 1 and SEQ ID NO: 2; (2) a primer pair for amplifying the region of MSH2 gene c.788 site, the sequences of which are shown in SEQ ID NO: 3 and SEQ ID NO: 4; (3) a primer pair for amplifying the region of MUTYH gene c.1005 site, the sequences of which are shown in SEQ ID NO: 5 and SEQ ID NO: 6; (4) a primer pair for amplifying the region of PMS2 gene c.288 site, the sequences of which are shown in SEQ ID NO: 7 and SEQ ID NO: 8; (5) a primer pair for amplifying the region of PMS2 gene c.780 site, the sequences of which are shown in SEQ ID NO: 9 and SEQ ID NO:

10.

2. A kit for early diagnosis of colorectal cancer, characterized by comprising the antibody of claim 1. The kit comprises the primer set of claim 1.

3. The kit of claim 2, wherein The kit is used for detecting at least one mutation selected from the group consisting of MLH1 c.113del A, MSH2 c.788del A, MUTYH c.1005 G>C, PMS2 c.288 C>T, and PMS2 c.780C>G.

4. The kit of claim 2, wherein The kit further comprises reagents for extracting circulating tumor DNA from a peripheral blood sample.

5. The kit of claim 2, wherein The kit further comprises reagents for library construction for sequencing the amplification products, the reagents comprising at least one of a DNA end repair enzyme, a DNA ligase, and a high-fidelity DNA polymerase.

6. Use of the kit of any one of claims 2 to 5 in the preparation of an in vitro detection product for early screening or risk assessment of colorectal cancer in Chinese population.

7. A gene mutation data analysis system, characterized in that, The system comprises: a data input interface configured to receive genotype data associated with a predetermined set of genetic loci, the predetermined set of genetic loci comprising MLH1 gene c.113, MSH2 gene c.788, MUTYH gene c.1005, PMS2 gene c.288, and PMS2 gene c.780; a processor; and a memory coupled to the processor and storing a computer program that, when executed by the processor, causes the system to perform the following operations: based on the genotype data corresponding to the predetermined set of genetic loci received by the data input interface, performing a calculation using a pre-stored logistic regression prediction model trained based on a mutation status of the predetermined set of genetic loci and a colorectal cancer risk association, to generate a risk score value and output the risk score value.

8. A computer-implemented method of analyzing genetic mutation data, the method comprising: The system comprises: Genotype data associated with a set of predetermined genetic loci, the set of predetermined genetic loci being c.113 of MLH1 gene, c.788 of MSH2 gene, c.1005 of MUTYH gene, c.288 of PMS2 gene and c.780 of PMS2 gene, are acquired; the acquired genotype data are input into a pre-trained logistic regression model trained based on an association between mutation states of the set of predetermined genetic loci and colorectal cancer risks of Chinese population, to generate a risk score value and output the risk score value.

Images