CN-122025135-A - MUC2 pancreatic cancer cell exosome ACOT 7-based early pancreatic cancer screening method for new-onset diabetes patients

CN122025135ACN 122025135 ACN122025135 ACN 122025135ACN-122025135-A

Abstract

The invention relates to the technical field of medical diagnosis and artificial intelligence aid decision making, and discloses a MUC2 pancreatic cancer cell exosome ACOT 7-based early pancreatic cancer screening method for a newly developed diabetes patient group, which comprises the following steps: for new diabetic population, detecting the expression levels of three biomarkers, namely ACOT7 protein, MUC2 protein and serum saccharide antigen CA19-9, derived from pancreatic cancer exosomes in blood, inputting a pancreatic cancer risk assessment model based on a three-level sequential feature refining framework, outputting integrated risk likelihood, and finally generating a structured screening report containing layering risk decision categories and clinical suggestions. The invention realizes accurate, efficient and automatic early risk stratification of pancreatic cancer of the newly developed diabetes high-risk group, and remarkably improves early detection rate of pancreatic cancer.

Inventors

ZHOU YU
CHEN RUFU
Anwar Yomar Abula
YANG JIABIN
GONG YANYU
CAO BINGXIN
ZHONG HAOMING
HUANG QIAOZHI

Assignees

广东省人民医院

Dates

Publication Date: 20260512
Application Date: 20260202

Claims (9)

1. A method for early screening pancreatic cancer in a new diabetic population based on MUC2 pancreatic cancer exosomes ACOT7, the method comprising: S1, obtaining a triple traceable marker spectrum of a subject, wherein the triple traceable marker spectrum comprises quantitative readings of ACOT7 protein carried by related exosomes of pancreatic cancer cells of the subject, quantitative readings of MUC2 protein from the same exosome source and quantitative readings of saccharide antigen CA19-9 in serum of the subject; s2, respectively mapping quantitative readings of the triple traceability marker spectrum into independent risk indication values based on respective biological meanings through a three-level sequential feature refining framework; and S3, calculating the integrated risk likelihood of the pancreatic cancer of the subject, dynamically mapping the hierarchical risk decision category and generating a screening report.
2. The method for early screening of pancreatic cancer in a group of new diabetics based on MUC2 pancreatic cancer exosomes ACOT7 according to claim 1, characterized in that: the step of obtaining the triple traceability marker spectrum of the subject comprises the following steps: dividing a blood sample of a subject of a new-born diabetes patient group into a first sub-sample and a second sub-sample; quantitatively detecting a quantitative reading of serum carbohydrate antigen CA19-9 for the first subsample; Enriching a second sub-sample by adopting an immunoadsorption material aiming at pancreatic cancer cell exosome surface specific antigen, and quantitatively reading ACOT7 protein and MUC2 protein; The quantitative readings obtained for ACOT7, MUC2 and CA19-9 were converted to percentile ranks based on the background of the new diabetic population.
3. The method for early screening of pancreatic cancer in a group of new diabetics based on MUC2 pancreatic cancer exosomes ACOT7 according to claim 1, characterized in that: the step of mapping quantitative readings of the triple traceability marker spectrum into independent risk indication values based on respective biological meanings through a three-level sequential feature refining framework comprises the following steps of: Performing compensation treatment based on exosome enrichment efficiency on quantitative readings of ACOT7, and outputting a first independent risk indication value; Calculating the ratio of the quantitative reading of MUC2 to the first independent risk indication value, performing inhibition treatment based on serum amylase level, and outputting a second independent risk indication value; The quantitative readings of CA19-9 are weighted based on the pre-detected subject Lewis antigen phenotype, bilirubin levels, and diabetes progression, and a third independent risk indicator is output.
4. The method for early screening for pancreatic cancer in a group of newly developed diabetics based on MUC2 pancreatic cancer exosomes ACOT7 according to claim 3, wherein: the step of performing compensation processing based on exosome enrichment efficiency on the quantitative readings of ACOT7 and outputting a first independent risk indication value comprises the following steps: Using a standard reference substance with known ACOT7 theoretical abundance, and inversely calculating the real-time process capture rate of the immunomagnetic beads capturing exosomes in the detection flow according to the actual measurement quantitative reading of the standard reference substance and the known theoretical abundance of the standard reference substance and combining with a preset detection gain coefficient; Comparing the real-time process capture rate with a reference capture rate calibrated in advance through a standard operation flow, and calculating a process deviation correction coefficient; Based on the reference capture rate and the system detection gain coefficient, reversely deducing theoretical background abundance of the sample to be detected under ideal capture conditions according to ACOT7 quantitative readings of the sample to be detected; and correcting the theoretical background abundance through the process deviation correction coefficient to obtain a first independent risk indication value.
5. The method for early screening for pancreatic cancer in a group of newly developed diabetics based on MUC2 pancreatic cancer exosomes ACOT7 according to claim 3, wherein: the step of calculating the ratio of the first independent risk indication value to the quantitative reading of MUC2, performing inhibition treatment based on serum amylase level, and outputting a second independent risk indication value comprises the steps of; Calculating the ratio of the MUC2 expression level to the first independent risk indicator value, and decoupling a homology signal representing the specific change of MUC2 relative to ACOT 7; and (3) inputting the homologous signal and the serum amylase level into a biomarker specific classifier together, and calculating to obtain a second independent risk indication value.
6. The method for early screening for pancreatic cancer in a group of newly developed diabetics based on MUC2 pancreatic cancer exosomes ACOT7 according to claim 3, wherein: the step of weighting the quantitative readings of CA19-9 based on the pre-detected subject Lewis antigen phenotype, bilirubin level and diabetes mellitus course, and outputting a third independent risk indicator value comprises: Taking CA19-9 quantitative reading as a core, and constructing a multidimensional evidence vector by combining the Lewis antigen phenotype, the total bilirubin level and the diabetes course of a subject; Inputting the multi-dimensional evidence vector into a pre-trained credibility evaluation network and outputting a situation weight between 0 and 1; and inputting the CA19-9 quantitative reading into a reference risk quantizer to obtain a basic risk indicated value, and multiplying the basic risk indicated value by the situation weight to obtain a third independent risk indicated value.
7. The method for early screening of pancreatic cancer in a group of new diabetics based on MUC2 pancreatic cancer exosomes ACOT7 according to claim 1, characterized in that: The step of calculating an integrated risk likelihood of a subject suffering from pancreatic cancer and generating a screening report indicative of the risk of pancreatic cancer in the subject comprises: weighting and fusing the independent risk indication values of the markers, and calculating the integrated risk likelihood of the pancreatic cancer of the subject; Calculating the corresponding statistical uncertainty range by utilizing a Bootstrap resampling technology while outputting the integrated risk likelihood; Dynamically judging a hierarchical risk decision type according to the relation between the statistical uncertainty range and a dynamically adjusted risk action trigger value; the independent risk indicator value, the integrated risk likelihood, and the stratified risk decision category for each marker are output as a screening report indicative of the pancreatic cancer risk of the subject.
8. The method for early screening of pancreatic cancer in a group of newly developed diabetics based on MUC2 pancreatic cancer exosomes ACOT7 according to claim 7, characterized in that: The step of dynamically determining the hierarchical risk decision category according to the relation between the statistical uncertainty range and a dynamically adjusted risk action trigger value comprises the following steps: Calculating the risk action trigger value by taking the independent risk indication value as a dynamic variable and combining the age and the diabetes course of the subject through a risk threshold regression method; And calculating the membership degrees of the three risk decision states belonging to the low risk, the medium risk and the high risk according to the statistical uncertainty range and the risk action trigger value, and determining the final layered risk decision category according to the maximum membership degree principle.
9. The method for early screening of pancreatic cancer in a group of newly developed diabetics based on MUC2 pancreatic cancer exosomes ACOT7 according to claim 7, characterized in that: For the same subject, when they have two or more consecutive screening records, the step of generating a screening report further comprises: calculating risk variation trend indexes of the subject, including absolute variation and relative variation rate of integrated risk likelihood; when the absolute change amount is greater than or equal to a first preset threshold value or the relative change rate is greater than or equal to a second preset threshold value, increasing the conclusion that the pancreatic cancer risk is in a rapid ascending trend in the screening report, and suggesting that the clinical imaging examination is performed immediately.

Description

MUC2 pancreatic cancer cell exosome ACOT 7-based early pancreatic cancer screening method for new-onset diabetes patients Technical Field The invention relates to the technical field of medical diagnosis and artificial intelligence aid decision making, in particular to a method for screening pancreatic cancer early in a newly developed diabetes patient group based on MUC2 pancreatic cancer exosome ACOT 7. Background Pancreatic cancer is one of the digestive tract tumors with the highest malignancy degree, early symptoms are hidden and lack of specific markers, and at present, the serum saccharide antigen CA19-9 widely used clinically has certain auxiliary diagnostic value, but has poorer specificity and is easily interfered by various non-cancer factors, and especially in patients with new diabetes mellitus, the false positive rate of CA19-9 is obviously increased, so that the accuracy and reliability of early screening are seriously insufficient. Epidemiological studies in recent years show that new diabetes is an important high-risk factor of pancreatic cancer, however, a special early screening scheme for the high-risk group is lacking at present, and the existing screening means rely on imaging examination, so that the cost is high, radiation exposure is complicated, the operation is difficult to be used for primary screening of large-scale groups, and a large number of potential early pancreatic cancer patients are missed or delayed. The exosomes serve as important carriers for intercellular communication, the content of the exosomes can reflect the pathological state of the source cells, the exosomes become research hotspots of tumor liquid biopsy, the traditional detection method is mostly based on tissue samples or whole blood exosomes, the specific enrichment and the accurate quantification of the exosomes of pancreatic cancer cells are lacked, and a multi-marker collaborative analysis system combined with CA19-9 is not established, so that the clinical transformation of the exosomes is limited. At present, a simple threshold judgment or linear weighting model is mostly adopted for pancreatic cancer risk prediction based on biomarkers, interference of confounding factors on the marker signals cannot be effectively stripped, dynamic calibration on individual features is also lacking, in addition, the monitoring and early warning of longitudinal risk trend are mostly ignored in the existing method, and real individuation and dynamic early risk layering are difficult to realize. Disclosure of Invention The invention aims to solve the problems in the prior art and provides a method for screening early pancreatic cancer of a newly developed diabetes patient group based on MUC2 pancreatic cancer exosome ACOT 7. In order to achieve the aim, the invention adopts the following technical scheme that the method for screening early pancreatic cancer of a newly developed diabetes patient group based on MUC2 pancreatic cancer cell exosome ACOT7 comprises the following steps: S1, obtaining a triple traceable marker spectrum of a subject, wherein the triple traceable marker spectrum comprises quantitative readings of ACOT7 protein carried by related exosomes of pancreatic cancer cells of the subject, quantitative readings of MUC2 protein from the same exosome source and quantitative readings of saccharide antigen CA19-9 in serum of the subject; s2, respectively mapping quantitative readings of the triple traceability marker spectrum into independent risk indication values based on respective biological meanings through a three-level sequential feature refining framework; and S3, calculating the integrated risk likelihood of the pancreatic cancer of the subject, dynamically mapping the hierarchical risk decision category and generating a screening report. The technical scheme provided by the invention has the beneficial effects that at least: According to the invention, a special screening path is designed for a new-onset diabetes patient group, and a linked marker system is constructed by jointly detecting ACOT7 and MUC2 proteins in exosomes derived from pancreatic cancer cells and combining serum CA19-9, so that the detection sensitivity and specificity of early-stage pancreatic cancer are remarkably improved, the screening blank of the new-onset diabetes high-risk group is filled, and a noninvasive, convenient and targeted early-stage early-warning tool is provided for clinic. According to the invention, through designing a three-level sequential characteristic processing flow of 'ACOT 7 enrichment efficiency compensation-MUC 2/ACOT7 ratio combined amylase inhibition-CA 19-9 multi-factor dynamic weighting', the influence of non-cancer factors such as exosome capture deviation, pancreatic inflammation interference, biliary tract diseases, individual genetic differences and the like is systematically stripped, so that the finally output risk score has more tumor specificity, and the false positive and false negati