Application of biomarkers in diagnosis / prediction of peritoneal recurrence after surgery for locally advanced gastric cancer

Abstract

The application discloses application of biomarkers in diagnosis / prediction of postoperative peritoneal recurrence of locally advanced gastric cancer, and particularly relates to biomarkers of BUB1, CKS2, PCNA, CHEK1, NEK2, NCAPG2 or a combination thereof. Through the system, a comprehensive biomarker discovery and multi-sample verification method, we determine a six-gene group and a risk stratification model to detect the risk of high-risk postoperative peritoneal recurrence of gastric cancer, which can provide information decision for clinic and improve the treatment outcome of patients if discovered early.

Description

Technical Field

[0001] This invention belongs to the field of biotechnology and relates to the application of biomarkers in the diagnosis / prediction of peritoneal recurrence after surgery for locally advanced gastric cancer. Specifically, the biomarkers involved are BUB1, CKS2, PCNA, CHEK1, NEK2, NCAPG2 or combinations thereof. Background Technology

[0002] Gastric cancer is the second leading cause of cancer-related death worldwide. It primarily manifests at an advanced stage, resulting in a significant proportion of patients initially diagnosed with metastatic disease or experiencing recurrence after adjuvant therapy. Peritoneal dissemination is the primary mechanism of recurrence and distant metastasis in locally advanced gastric cancer (LAGC), ultimately leading to poor prognosis with a median survival of less than 12 months. The ineffectiveness of conventional systemic chemotherapy and limited treatment options are major contributing factors to the poor prognosis of patients with peritoneal metastases (PM) of gastric cancer. Furthermore, the lack of robust diagnostic tools for early detection of PM is a significant obstacle to improving patient outcomes. Staging laparoscopy offers higher diagnostic accuracy in identifying PM in gastric cancer patients compared to computed tomography (CT) or positron emission tomography-CT (PET-CT), and is therefore increasingly being used. Despite its advantages, the invasiveness, requirement of general anesthesia, and relatively high cost of staging laparoscopy hinder its widespread adoption as a standard procedure. While numerous retrospective studies have highlighted the importance of staging laparoscopy in the diagnosis of PM, consensus on its routine application in LAGC patients remains inconsistent.

[0003] A growing body of research confirms that the presence of intraperitoneal cancer cells is associated with an increased incidence of peritoneal recurrence or metastasis in gastric cancer patients. The 8th edition of the American Joint Committee on Cancer (AJCC) / Union for International Cancer Control (UICC) classification uses washed cytology to detect malignant cells, thus aiding in more precise tumor staging. Tumors with positive cytology, even without visible metastases (POCY1), are classified as advanced cancer with distant metastases. Early identification of intraperitoneal cancer cells is crucial because it necessitates treatments that significantly improve patient outcomes. Preoperative detection of peritoneal dissemination (PM) facilitates the implementation of aggressive treatments such as neoadjuvant intraperitoneal and systemic chemotherapy (NlPS), intraperitoneal hyperthermic chemotherapy (HlPEC), and cytoreductive surgery (CRS). While these treatments are not yet standardized, emerging evidence highlights their potential to significantly improve prognosis. This underscores the importance of early detection of metastatic tumors and the critical role of early detection of peritoneal dissemination in improving treatment outcomes. Therefore, developing accurate molecular biomarkers for early detection of PM, through timely intervention, has the potential to significantly reduce morbidity and mortality associated with this disease. Summary of the Invention

[0004] In order to solve the technical problems existing in the prior art, the present invention provides the following technical solution:

[0005] This invention provides the use of a substance for detecting biomarkers in a sample in any of the following:

[0006] A1. Application in the preparation of products for diagnosing postoperative recurrence of gastric cancer;

[0007] A2. Application in the preparation of products for predicting postoperative recurrence of gastric cancer;

[0008] The biomarkers are BUB1, CKS2, PCNA, CHEK1, NEK2, NCAPG2, or combinations thereof.

[0009] Furthermore, the biomarker is a combination of BUB1, CKS2, PCNA, CHEK1, NEK2, and NCAPG2.

[0010] Furthermore, the biomarkers also include invasion depth, tumor size, and vascular tumor thrombus invasion.

[0011] Furthermore, the biomarkers are a combination of BUB1, CKS2, PCNA, CHEK1, NEK2, NCAPG2, invasion depth, tumor size, and vascular tumor thrombus invasion.

[0012] Furthermore, the products include reagents, kits, test strips, primers, probes, and chips.

[0013] Furthermore, the kit may also include one or more substances selected from the group consisting of: containers, instructions for use, positive controls, negative controls, buffers, auxiliaries, or solvents.

[0014] Furthermore, the chip includes gene chips and protein chips.

[0015] In some embodiments, the gene chip includes a solid support and probes immobilized on the solid support.

[0016] In some implementations, the probe includes a probe targeting the biomarker for detecting the transcriptional level of the biomarker.

[0017] In some embodiments, the protein chip includes a solid support and an antibody immobilized on the solid support that specifically binds to a protein encoded by the biomarker.

[0018] Furthermore, the kit also includes an auxiliary reagent for detecting mRNA expression levels.

[0019] Furthermore, the mRNA expression level auxiliary detection reagent includes reaction reagents for visualizing amplicon corresponding to primers, RNA extraction reagents, reverse transcription reagents, cDNA amplification reagents, and / or standards used to prepare standard curves.

[0020] Furthermore, the reaction reagents for visualizing the amplicon corresponding to the primers include reagents used in agarose gel electrophoresis, enzyme-linked gel electrophoresis, chemiluminescence immunoassay, in situ hybridization, and fluorescence detection to visualize the amplicon.

[0021] Furthermore, the kit also includes a protein expression level detection reagent.

[0022] Furthermore, the protein expression level auxiliary detection reagent includes blocking solution, antibody dilution solution, washing buffer, and colorimetric stop solution.

[0023] Furthermore, the samples include blood, serum, plasma, peripheral blood, tissue, blood cells, bone marrow, ascites, fine needle biopsy samples, cellular body fluids, free floating nucleic acids, sputum, saliva, urine, cerebrospinal fluid, peritoneal fluid, pleural fluid, feces, lymph, skin swabs, oral swabs, nasal swabs, or lavage fluid.

[0024] Furthermore, the samples include blood, serum, plasma, peripheral blood, tissue, or fine needle biopsy samples.

[0025] In some implementations, the sample taken directly from the subject is not further processed; for example, blood may be obtained from the subject's peripheral circulatory system (e.g., via a lancet). The sample may include, for example, blood, urine, feces, saliva, cerebrospinal fluid, and sweat. Non-limiting examples of the samples include blood (or blood components—e.g., white blood cells, red blood cells, platelets) obtained from any anatomical location of the subject (e.g., tissue, circulatory system, bone marrow), cells obtained from any anatomical location of the subject, skin, heart, lungs, kidneys, exhaled breath, bone marrow, feces, semen, vaginal fluid, tissue fluid derived from tumor tissue, breast, pancreas, cerebrospinal fluid, tissue, throat swabs, biopsies, placental fluid, amniotic fluid, liver, muscle, smooth muscle, bladder, gallbladder, colon, intestine, brain, cavity fluid, sputum, pus, microbiota, meconium, breast milk, prostate, esophagus, thyroid gland, serum, saliva, urine, gastric juice and digestive juices, tears, ocular fluid, sweat, mucus, earwax, oil, glandular secretions, cerebrospinal fluid, hair, nails, skin cells, plasma, nasal swabs or nasopharyngeal washes, cerebrospinal fluid, umbilical cord blood, lymph and / or other excretions or body tissues.

[0026] Furthermore, the gastric cancer includes locally advanced gastric cancer.

[0027] Furthermore, the recurrence of gastric cancer after surgery includes peritoneal recurrence, lymph node recurrence, lung recurrence, liver recurrence, and bone recurrence.

[0028] Furthermore, the sample is derived from the object to be tested.

[0029] Furthermore, the objects to be detected include humans or non-human mammals.

[0030] Furthermore, the object to be detected is a person.

[0031] As used herein, the term "substance" may refer to any chemical class of compound or entity, including, for example, polypeptides, nucleic acids, sugars, lipids, small molecules, metals, or combinations or complexes thereof. Where appropriate, it will be clear to those skilled in the art that the term may refer to a cell or organism or its fractions, extracts, or components, or an entity containing a cell or organism or its fractions, extracts, or components. Alternatively or supplemented, as the context will make clear, the term may refer to a natural product because it is found in and / or obtained from nature. In some cases, again as will be clear from the context, the term may refer to one or more man-made entities because it is designed, engineered, and / or produced by human hand action, and / or is not found in nature. In some embodiments, the substance may be used in isolated or pure form; in some embodiments, the substance may be used in crude form. In some embodiments, potential substances may be provided as collections or libraries, for example, which may be screened to identify or characterize the active substances therein. In some instances, the term "substance" may refer to a polymer or a compound or entity containing a polymer; in other instances, the term may refer to a compound or entity containing one or more polymeric moieties. In some embodiments, the term "substance" may refer to a compound or entity that is not a polymer and / or substantially free of any polymer and / or one or more specific polymeric moieties. In some embodiments, the term may refer to a compound or entity lacking or substantially free of any polymeric moieties. In some embodiments, the substance is a compound. In some embodiments, the reinforcing agent may refer to a food-grade compound bound to a salt. The reinforcing agent binds to the salt by forming electrostatic interactions, hydrogen bonds, ionic interactions, coordination interactions, non-covalent interactions, van der Waals interactions, or any combination thereof.

[0032] The present invention provides a product comprising a detection reagent for detecting the content of a biomarker, wherein the biomarker is BUB1, CKS2, PCNA, CHEK1, NEK2, NCAPG2 or a combination thereof.

[0033] Furthermore, the biomarker is a combination of BUB1, CKS2, PCNA, CHEK1, NEK2, and NCAPG2.

[0034] Furthermore, the biomarkers also include invasion depth, tumor size, and vascular tumor thrombus invasion.

[0035] Furthermore, the biomarkers are a combination of BUB1, CKS2, PCNA, CHEK1, NEK2, NCAPG2, invasion depth, tumor size, and vascular tumor thrombus invasion.

[0036] Furthermore, the products include reagents, kits, test strips, primers, and probes.

[0037] Furthermore, the kit includes an immunohistochemistry detection kit, an immunoblotting detection kit, an immunochromatographic detection kit, a flow cytometry analysis kit, a qPCR kit, and an ELISA kit.

[0038] Furthermore, the products also include reagents commonly used in PCR reactions, RT-PCR derivatization reactions, 3SR amplification, LCR, SDA, NASBA, TMA, SYBR Green, TaqMan probes, molecular beacons, two-hybrid probes, composite probes, ISH, microarrays, Southern blotting, Northern blotting, multianalyte assays, enzyme-linked immunosorbent assays (ELISA), radioimmunoassays, immunofluorescence assays, enzyme immunoassays, immunoprecipitation assays, chemiluminescence assays, immunohistochemistry assays, dot blot assays, or narrow-line blot assays.

[0039] This invention provides a system for diagnosing whether gastric cancer has recurred after surgery, or for predicting the risk of recurrence after gastric cancer surgery, the system comprising:

[0040] The input unit is used to acquire data on the expression levels of biomarkers in the sample.

[0041] The result determination unit is used to input the data from the detection unit into the computer program for processing and determination, and to obtain the comparison result by comparing it with the set threshold value in the computer.

[0042] The output unit is used to output the results of the result determination unit to determine whether the object corresponding to the test sample has recurred gastric cancer after surgery or the probability of recurrence after gastric cancer surgery.

[0043] All units in the system are operated using computer programs or artificial intelligence.

[0044] The biomarkers are BUB1, CKS2, PCNA, CHEK1, NEK2, NCAPG2, or combinations thereof.

[0045] Furthermore, the biomarker is a combination of BUB1, CKS2, PCNA, CHEK1, NEK2, and NCAPG2.

[0046] Furthermore, the biomarkers also include invasion depth, tumor size, and vascular tumor thrombus invasion.

[0047] Furthermore, the biomarkers are a combination of BUB1, CKS2, PCNA, CHEK1, NEK2, NCAPG2, invasion depth, tumor size, and vascular tumor thrombus invasion.

[0048] The apparatus, electronic device, and non-volatile computer storage medium and method provided in the embodiments of this specification are corresponding. Therefore, the apparatus, electronic device, and non-volatile computer storage medium also have similar beneficial technical effects as the corresponding method. Since the beneficial technical effects of the method have been described in detail above, the beneficial technical effects of the corresponding apparatus, electronic device, and non-volatile computer storage medium will not be repeated here.

[0049] This invention provides a gastric cancer postoperative recurrence risk prediction model, which includes using machine learning methods to train an algorithm model for diagnosing whether gastric cancer has recurred after surgery, or predicting the risk of gastric cancer recurrence after surgery.

[0050] Furthermore, the algorithm model processes the data to obtain the postoperative recurrence assessment results of gastric cancer. The postoperative recurrence assessment results of gastric cancer include the postoperative recurrence probability of gastric cancer and important characteristic factors. The important characteristic factors reflect the correlation with postoperative recurrence of gastric cancer. The important characteristic factors are the expression level data of BUB1, CKS2, PCNA, CHEK1, NEK2, NCAPG2 or combinations thereof extracted from the information parameters of the subject to be tested.

[0051] Furthermore, the important feature factors also include invasion depth, tumor size, and vascular tumor thrombus invasion data extracted from the information parameters of the object to be detected.

[0052] Furthermore, one of the RSA models in the algorithm model is: [(3.665 × 6-mRNA biomarker combination) + (5.009 × invasion depth) + (2.009 × tumor size) + (1.417 × vascular tumor thrombus invasion) + (-12.340)]. This RSA model is used to assess the risk of recurrence after gastric cancer surgery.

[0053] Furthermore, the machine learning methods include decision tree learning, random forest learning, K-nearest neighbor algorithm learning, Naive Bayes learning, support vector machine learning, neural network learning, and Adeboost learning.

[0054] In some implementations, training data is input into the AdaBoost algorithm model to train a series of weak classifiers, which are then linearly combined into a strong classifier. Each time a weak classifier is trained, the AdaBoost algorithm alters the probability distribution of the training data, resulting in different weights for each weak classifier. Each misclassified sample receives a larger weight in the next weak classifier. Then, AdaBoost uses a weighted majority voting method to increase the weight of weak classifiers with lower classification error rates and decrease the weight of weak classifiers with higher classification error rates. These classifiers are then linearly combined to form the final classifier. The resulting classifier is tested using test data from the dataset to be processed, and the AdaBoost algorithm is continuously adjusted based on the test results. This yields an evaluation model that can be used for postoperative recurrence risk analysis of gastric cancer, enabling the prediction of postoperative recurrence risk.

[0055] The present invention provides a computer-readable storage medium having a computer program thereon, the computer program being stored in the computer-readable storage medium, and the computer program being executed by a processor to implement the aforementioned gastric cancer postoperative recurrence risk prediction model.

[0056] The various embodiments in this specification are described in a progressive manner. Similar or identical parts between embodiments can be referred to interchangeably. Each embodiment focuses on describing the differences from other embodiments. In particular, the embodiments for apparatus, electronic devices, and non-volatile computer storage media are basically similar to the method embodiments, so the descriptions are relatively simple; relevant parts can be referred to the descriptions of the method embodiments.

[0057] The systems, devices, modules, or units described in the above embodiments can be implemented by computer chips or entities, or by products with certain functions. A typical implementation device is a computer. Specifically, a computer can be, for example, a personal computer, laptop computer, cellular phone, camera phone, smartphone, personal digital assistant, media player, navigation device, email device, game console, tablet computer, wearable device, or any combination of these devices.

[0058] In some embodiments, the above-described modules or steps can be implemented using a computing device. In some embodiments, the modules or steps can be centralized on a single computing device. In some embodiments, the modules or steps can be distributed across multiple computing devices linked by a network or other means. In some embodiments, the computing device stores program code for executing the above-described modules or steps in a storage device. In some embodiments, the modules or steps can be fabricated as integrated circuits individually or combined into a single integrated circuit. No particular implementation method, whether hardware or software, is limited in the specific embodiments of this invention.

[0059] As used in this invention, the terms “subject,” “patient,” “subject,” “test subject,” or “individual” refer to any subject requiring treatment or prevention, particularly vertebrate subjects, and even more particularly mammalian subjects. Suitable vertebrates falling within the scope of this invention include, but are not limited to, any member of the subphylum Chordata, including primates (e.g., humans, monkeys, and apes, and including monkey species (e.g., genus Macaca, such as macaques like Macaca fascicularis and / or Macacamula laatta)) and baboons (Papio baboons). Marmosets (from the genus Callithrix), squirrel monkeys (from the genus Saimiri) and tamarins (from the genus Saguinus), as well as species of orangutans such as chimpanzees (Pantroglodytes), rodents (e.g., mice, rats, guinea pigs), rabbits (e.g., rabbits, hares), cattle (e.g., cattle), sheep (e.g., sheep), goats (e.g., goats), pigs (e.g., pigs), horses (e.g., horses), dogs (e.g., dogs), felines (e.g., cats), birds (e.g., chickens, turkeys, ducks, geese, companion birds such as canaries, budgerigars, etc.), marine mammals (e.g., dolphins, whales), reptiles (snakes, frogs, lizards, etc.) and fish.

[0060] This invention provides a computer-based evaluation method for assessing whether gastric cancer recurs after surgery, or the risk of gastric cancer recurrence after surgery, the method comprising:

[0061] Acquire data to be processed, wherein the data to be processed includes detection data of biomarkers such as BUB1, CKS2, PCNA, CHEK1, NEK2, NCAPG2 or combinations thereof in the sample;

[0062] The data to be processed is input into the gastric cancer postoperative recurrence risk prediction model mentioned above to obtain the assessment results of whether the gastric cancer has recurred or the risk of gastric cancer recurrence after surgery.

[0063] The results of the assessment of whether gastric cancer recurred after surgery or the risk of gastric cancer recurrence after surgery will be output from the data to be processed.

[0064] Furthermore, the data to be processed also includes detection data on the subject's invasion depth, tumor size, and vascular tumor thrombus invasion.

[0065] The foregoing has described specific embodiments of this specification. Other embodiments are within the scope of this invention. In some cases, the actions or steps described in this invention may be performed in a different order than those in the embodiments and still achieve the desired results. Furthermore, the processes depicted in the drawings do not necessarily require the specific or sequential order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0066] The term "machine learning" as used in this invention refers to algorithms that give computers the ability to learn without explicit programming, including algorithms that learn from data and make predictions about that data. The machine learning algorithms used in the embodiments disclosed herein may include (but are not limited to) Random Forest (RF), Least Absolute Shrinkage and Selection Operator (LASSO) logistic regression, regularized logistic regression, XGBoost, decision tree learning, artificial neural networks (ANN), deep neural networks (DNN), support vector machines, rule-based machine learning, etc. Algorithms such as linear regression or logistic regression can be used as part of the machine learning process.

[0067] The term "comprising / including" as used in this invention means that a composition or method comprising one or more named elements or steps is open-ended, meaning that the named element or step is necessary, but other elements or steps may be added within the scope of the composition or method. To avoid redundancy, it should also be understood that any composition or method described as "comprising / including" one or more named elements or steps also describes a correspondingly more limited composition or method "consisting substantially of the same named elements or steps," meaning that the composition or method includes the named necessary element or step and may also include other elements or steps that do not substantially affect the fundamental and novel characteristics of the composition or method.

[0068] Advantages and beneficial effects of the present invention:

[0069] We developed a risk prediction model using these genetic markers to successfully predict the risk of peritoneal recurrence in surgical, endoscopic, and blood specimens from patients with peritoneal recurrence of gastric cancer in a multicenter cohort, transitioning from invasive tissue specimens to non-invasive liquid biopsy methods. Importantly, given the association between peritoneal recurrence and metastasis, we evaluated the performance of diagnostic genomics in multiple independent detection methods for peritoneal metastasis in prospective cohorts of patients with initial metastatic or P0CY1 tumors. Through this systematic, comprehensive biomarker discovery and multi-sample validation approach, we identified a six-genome and risk stratification model to detect the risk of postoperative peritoneal recurrence in high-risk gastric cancer patients, which, if detected early, can inform clinical decision-making and improve patient outcomes. Attached Figure Description

[0070] Figure 1 This is a flowchart of the research design for discovering and validating 6-mRNA biomarkers in different LAGC patient populations;

[0071] Figure 2 This is a diagram illustrating the discovery process and preliminary validation results of candidate biomarkers for peritoneal recurrence in patients with LAGC based on transcriptomics. (A) Using appropriate transcriptomic data from the TCGA and GEO databases, and Venn diagrams of mRNA sequencing from 3 pairs of patients with and 3 pairs without peritoneal recurrence, 6 candidate mRNAs were identified; (B) The intersection of related genes from the 6 genes in the panel revealed 39 commonly associated genes; (C) 6 pairs of mRNAs... (D) Venn diagram of drug sensitivity analysis of 198 anticancer drugs in the GDSC database; (E) Comparison of the expression of 6 mRNAs (BUB1, CKS2, PCNA, CHEK1, NEK2, NCAPG2) in cancerous and normal tissues in the TCGA database; (F) Comparison of the expression of 6 mRNAs in cancerous and normal tissues in 29 fresh frozen tissue samples; (G) Comparison of the expression of 6 mRNAs in peripheral blood samples from 22 gastric cancer patients and healthy individuals; (H) Construction of PPI network of these 6 mRNAs using the online STRING database (https: / / string-db.org); (X) Correlation heatmap analysis of 6 mRNAs with common metastatic genes based on the TCGA database;

[0072] Figure 3This is a graph showing the results of the transcriptome training and validation phases for identifying peritoneal recurrence in surgical specimens from patients with LAGC. (A) A nomogram of peritoneal recurrence in LAGC patients constructed based on 6-mRNA and clinical features; (B) ROC curves for different predictors in the training set; (C) ROC curves for different predictors in the validation set; (DI) A prediction confusion matrix of the RSA model constructed using clinical features and 6-mRNA in both the training and validation sets; (J) Calibration curve of the RSA model in the training set; (K) Calibration curve of the RSA model in the validation set; (LQ) A graph showing the clinical benefit of the RSA model constructed using clinical features, 6-mRNA, and their combination in the training and validation sets; (R) Log-rank test survival curves dividing patients in the training set into low-risk and high-risk groups based on the Youden index obtained from the nomogram; (S) Log-rank test survival curves, with the Youden index obtained from the nomogram used to divide patients in the validation set into low-risk and high-risk groups.

[0073] Figure 4 This is a graph showing the results of peritoneal recurrence in endoscopic biopsy specimens from patients with LAGC, used in the transcriptome validation phase. The graph includes: (AF) Correlation analysis of six mRNAs in endoscopic biopsy specimens and paired surgically resected specimens; (GL) Comparison of the expression of six mRNAs in endoscopic biopsy specimens and paired surgically resected specimens; (M) ROC curves for different predictors in endoscopic biopsy specimens; (N) Calibration curves of the RSA model in endoscopic biopsy specimens; (O) Log-rank survival curves plotted based on the cutoff value obtained from the Youden index of the nomogram, classifying patients into low-risk and high-risk groups; (PR) A prediction confusion matrix constructed from clinical features, 6-mRNA, and the RSA model; and (SU) A graph showing the clinical benefit of using clinical features, 6-mRNA, and their combined construction of the RSA model in the endoscopic biopsy specimen validation set.

[0074] Figure 5This is a graph showing the results of peritoneal recurrence in peripheral blood samples from LAGC patients during the transcriptome validation phase. (A) A nomogram of peritoneal recurrence in LAGC patients constructed based on 6-mRNA combined with clinical features; (B) ROC curves of different predictors in the training set; (C) ROC curves of different predictors in the validation set; (D) ROC curves of different predictors in the tumor marker negative group; (EM) A prediction confusion matrix of the RSA model constructed using clinical features, 6-mRNA, and both in the training, validation, and tumor marker negative sets; (NV) An RSA model constructed using clinical features, 6-mRNA, and their combination in the training, validation, and tumor marker negative sets. Clinical benefit plot of the model; (W1) Calibration curve of the RSA model in the training set; (W2) Calibration curve of the RSA model in the validation set; (W3) Calibration curve of the RSA model in the tumor marker negative set; (X) Log-rank test survival curves plotted by dividing patients in the training set into low-risk and high-risk groups based on the Youden index obtained from the nomogram; (Y) Log-rank test survival curves plotted by dividing patients in the validation set into low-risk and high-risk groups based on the Youden index obtained from the nomogram; (Z) Log-rank test survival curves plotted by dividing patients in the tumor marker negative group into low-risk and high-risk groups based on the Youden index obtained from the nomogram.

[0075] Figure 6 These are transcriptome validation results used to identify peritoneal metastases and micrometastases in peripheral blood samples from LAGC patients; (A) ROC curves predicting peritoneal micrometastases (POCY1) using different predictor variables; (B) HE staining of free cancer cells detached from the peritoneum; (C) ROC curves predicting PM using different predictor variables; (D) Diagnostic laparoscopy showing peritoneal metastatic nodules; (EG) Confusion matrix diagram predicting peritoneal micrometastases (POCY1) using clinical features, 6-mRNA, and an RSA model constructed from both; (HJ) Confusion matrix diagram predicting peritoneal metastases using an RSA model constructed from clinical features, 6-mRNA, and their combination; (KM) Confusion matrix diagram predicting peritoneal metastases using clinical features, 6-mRNA, and their combination. Clinical benefit graphs for predicting peritoneal micrometastasis (P0CY1) using 6-mRNA and combined RSA models; (NP) Clinical benefit graphs for predicting PM using clinical characteristics, 6-mRNA, and combined RSA models; (Q) Calibration curves for RSA model prediction of peritoneal micrometastasis in (P0CY1) patients; (R) Calibration curves for RSA model prediction of PM in the cohort; (S) Log-rank test survival curves for dividing the peritoneal micrometastasis (P0CY1) patients into low-risk and high-risk groups based on the Youden index obtained from the nomogram; (T) Log-rank test survival curves, dividing the PM group into low-risk and high-risk groups based on the Youden index obtained from the nomogram; (U)

[0076] ROC curves for 6-mRNA prediction of peritoneal metastasis in colorectal cancer; (V) ROC curves for 6-mRNA prediction of peritoneal metastasis in hepatocellular carcinoma; (W) ROC curves for 6-mRNA prediction of peritoneal metastasis in pancreatic ductal carcinoma. Detailed Implementation

[0077] Example

[0078] 1. Experimental Methods

[0079] 1.1 Discovering biomarkers in whole-genome expression profiling datasets

[0080] The biomarker discovery and validation process used in this study is as follows: Figure 1 As shown, firstly, mRNA sequencing data from multiple sources, including the Gene Expression Omnibus (GEO) database, the Cancer Genome Atlas (TCGA) database, and samples matched for PM recurrence, were used for biomarker identification. For validation, we used a cohort of fresh frozen postoperative specimens, endoscopic biopsy specimens, and peripheral blood samples from multiple medical clinics to ensure thorough and reproducible validation.

[0081] During the discovery phase, stringent screening criteria for the application of the GEO database (https: / / www.ncbi.nlm.nih.gov / geo / ) led to the selection of the GSE15081 dataset for label development. mRNA sequencing data were collected from three pairs of cancer tissues that underwent radical resection of LAGC followed by adjuvant therapy and subsequently developed peritoneal metastasis, and from three pairs of non-recurrent cancer tissues at the Fourth Hospital of Hebei Medical University.

[0082] We first assessed the expression of selected mRNAs in a pilot cohort of 29 matched GC and adjacent non-malignant tissue (ANM) samples using real-time quantitative polymerase chain reaction (RT-qPCR). Peripheral blood samples were collected from 22 gastric cancer patients and 22 healthy individuals undergoing routine physical examinations during the same period to detect the expression of candidate mRNAs in the bloodstream. Samples were obtained from the Fourth Hospital of Hebei Medical University (FHHMU) between January and March 2023.

[0083] 1.2 Clinical cohorts for biomarker validation

[0084] This study collected 329 fresh frozen specimens from patients with left atrial fibrillation gastric cancer (LAGC) in two independent cohorts. The aim was to develop and validate biomarkers for predicting PM recurrence. Specimens were collected between August 2016 and March 2019, excluding patients who received neoadjuvant therapy or had residual gastric cancer after partial gastrectomy. 196 patients were from the Fourth Hospital of Hebei Medical University (FHHMU) and Shijiazhuang People's Hospital (S1ZPH), and 133 patients were from Baoding Central Hospital (BDCH), Nanjing University Jinling Hospital (INNI), and Wuhan University People's Hospital (WHPH).

[0085] Further analysis included 103 matched endoscopic biopsy specimens from 5 institutions to validate the shift of biomarkers from surgical specimens to endoscopic biopsy specimens.

[0086] In addition, we retrospectively analyzed serum samples from patients with LAGC, including those with postoperative PM recurrence and those without, to identify tissue-based biomarkers for liquid biopsy. We selected 120 LAGC patients admitted to Zhongshan Hospital affiliated with Fudan University between February 2017 and December 2019. The validation cohort included serum samples from 123 LAGC patients from four other institutions between January and December 20, 2016.

[0087] In addition, we conducted an analysis to evaluate the effectiveness of biomarkers in detecting PM and micrometastases in gastric cancer patients. Peripheral blood samples from 66 patients with LAGC (including 12 with peritoneal cytology-confirmed P0CY1 status [FHHMU.registration.nct03718624,ChiCTR1800014817] and 95 patients with LAGC (including 15 with laparoscopically confirmed PM [FHHMU, registration: NCT02555358, NCT01516944]) were used to predict P0CY1 occurrence and PM pathogenesis, respectively. Furthermore, we evaluated the specificity of our mRNA combination as a biomarker for PM in LAGC compared to other gastrointestinal malignancies (including colorectal cancer, pancreatic ductal carcinoma, and hepatocellular carcinoma) using RT-qPCR on serum samples collected from FHHMU patients between 2019 and 2021.

[0088] In accordance with gastric cancer treatment guidelines, patients were followed up for recurrence or disease progression through laboratory tests, endoscopy, and abdominal and pelvic CT scans. Tissue specimens were immediately frozen in liquid nitrogen and stored at -80°C. Surgical specimens were processed according to the guidelines of the Chinese Society of Clinical Oncology. Tumors and lymph nodes were staged according to the 8th edition of the AJCC staging system. All procedures were performed in accordance with the Declaration of Helsinki, with written informed consent obtained from all participants and approval obtained from the institutional review committees of all participating institutions.

[0089] 1.3 RNA extraction and gene expression analysis

[0090] Total RNA was isolated from freshly frozen surgical tissue using TRlzol reagent according to the manufacturer's protocol. For serum samples, 200 μL aliquots were thawed on ice, centrifuged at 3000 rpm for 5 minutes to remove cell debris, and then mixed with 5 times the TRlzol volume. Total RNA was extracted using the QiagenmiRNeasy kit. 1 / 5 volume of chloroform was added to the mixture, the mixture was inverted, and the sample was placed on ice for 5 minutes. The mixture was centrifuged at 12,000 rpm at 4°C for 15 minutes. After 15 minutes, the supernatant was aspirated, and an equal volume of isopropanol was added. The mixture was stirred and placed on ice for 10 minutes. The mixture was centrifuged again at 12,000 rpm at 4°C for 10 minutes. After discarding the supernatant, 400 μL of 75% ethanol was added, and the mixture was centrifuged at 9000 rpm at 4°C for 5 minutes. The supernatant was discarded, and after the precipitate dried, an appropriate amount of DEPC water was added to dissolve the precipitate. Reverse transcription was performed using Roche Germany, followed by qRT-PCR on an Applied Biosystems Real-Time PCR system. The pre-denaturation program was set to 95°C for 10 minutes. Then, 40 cycles were performed, each ending with denaturation at 95°C for 15 seconds, annealing at 58°C for 30 seconds, and extension at 72°C for 30 seconds. Finally, the melting curve program was set to 95°C, with the extension terminated within 15 seconds, the system temperature maintained at 60°C for 1 minute, and then increased to 95°C by 0.3°C per unit time for 15 seconds.

[0091] Several measures ensured the reproducibility of the assays, including appropriate controls, exclusion of low-quality RNA samples, and repeated assays at different time points. Gene expression was quantified using the Applied Biosystems QuantStudio 6Flex Real-Time PCR System software. The relative abundance of the target gene was determined using the 2^-ΔΔct method, with GAPDH as the internal reference standard, where ΔCT is the difference between the target gene and the GAPDH CT value.

[0092] 1.4 PPI protein interaction network analysis, pathway analysis, and chemotherapeutic drug sensitivity analysis

[0093] Protein-protein interaction (PPI) networks of target genes were constructed using the STRING database (https: / / string-db.org), retaining genes with a comprehensive score ≥0.4. The PPI networks from the STRING database were imported into Cytoscape v3.9.1 for topology analysis. Cytoscape provides 11 topology methods. This paper uses a degree-based approach to identify key genes based on the connectivity within the network. Cytoscape 3.9.1 also supports visualized PPI network representations. Furthermore, Enrichr (https: / / maayanlab.cloud / Enrichr / ) software was used to perform gene set and pathway analysis on four candidate genes.

[0094] Using Oncopredict, we identified potential therapeutic agents for each gene by utilizing data from the Genomics of Drug Sensitivity in Cancer (GDSC) database (https: / / www.cancerrxgene.org / ), which categorizes genes into high-expression and low-expression groups based on their median expression levels. We then used the Wilcoxon rank-sum test to compare differences in drug sensitivity among the groups.

[0095] 1.5 Statistical Analysis

[0096] Statistical analysis was performed using IBM SPSS 23 software, R version 3.6.3 and GraphPad Prism version 8.0. Meaningful clinicopathological and mRNA categorical variables were included as covariates in univariate and multivariate regression analyses. Variables meaningful in univariate analyses were incorporated into multivariate regression analyses. In the discovery phase, differential gene expression between the PM relapse group and the non-PM relapse group was assessed using the Wilcoxon rank-sum and Bonferroni tests. In the clinical validation phase, a gene-based risk score model was established using logistic regression analysis with inverse elimination, and model performance was evaluated using receiver operating characteristic (ROC) curves and area under the curve (AUC) values.

[0097] ROC curves were plotted using the pROC package in R, and the area under the curve (AUC) was calculated. The DeLong test was used for ROC curve comparisons. The sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), precision, and accuracy of the 6-mRNA biomarker set across all cohorts were determined using the Report ROC package as shown in the confusion matrix. The optimal cutoff value for the ROC curves was established using the Youden index in the pROC procedure. Patients were grouped into high-risk and low-risk groups based on the Youden index and median risk score, which helps predict recurrence and detect metastasis. We analyzed recurrence-free survival (RFS) using the Kaplan-Meier method, defined as the time interval from surgery to confirmed peritoneal recurrence or all-cause death, and recurrence-free patients surviving at this milestone were followed up for 5 years. Lost-to-follow-up patients without evidence of recurrence 5 years prior were assessed at their last visit. Statistical significance was set at p < 0.05.

[0098] 2. Experimental Results

[0099] 2.1 Identification of genome-wide gene expression profiles of candidate mRNAs overexpressed in patients with peritoneal recurrence of gastric cancer

[0100] In this study, we conducted preliminary biomarker identification by analyzing transcriptomic data from two publicly available gastric cancer datasets (TCGA and GSE15081). Furthermore, we performed mRNA sequencing on three pairs of gastric cancer tissues with peritoneal recurrence after radical surgery and three pairs without peritoneal metastasis. The GSE15081 dataset included 33 patients with peritoneal recurrence and 75 patients without recurrence. Through differential gene expression analysis (Wilcoxon rank-sum test for GSE15081 and EdgeR for TCGA, both P < 0.05, with Bonferroni correction) and correlation analysis (r < 0.5), we identified six differentially expressed genes (BUB1, CKS2, PCNA, CHEK1, NEK2, and NCAPG2) between patients with and without peritoneal recurrence. Figure 2 A).

[0101] Further analysis of TCGA data showed that these genes were significantly overexpressed in cancer tissues compared to ANM (P<0.05). Figure 2 D). To validate these results, we established a pilot cohort at FHHMU, including individuals undergoing gastric cancer and health screenings from January to December 2023. Analysis of 29 matched GC and ANM tissue samples by RT-qPCR confirmed high expression levels of these mRNAs in GC tissues (P<0.05). Figure 2 E). Peripheral blood analysis of 22 gastric cancer patients and 22 healthy individuals undergoing physical examinations during the same period also showed elevated expression of the candidate gene in cancer patients. Figure 2 F).

[0102] To investigate the potential association between gene expression and drug sensitivity in our panel, we used the "oncoppredict" R package. This software analyzes expression matrices and drug response data from the Genomics of Drug Sensitivity in Cancer (GDSC) database, which includes information on 198 anticancer drugs. Analysis of the GSE15081 dataset showed that high expression of six identified genes increased sensitivity to BMS-754807 and AZD8186. Figure 2 C).

[0103] Protein-protein interaction networks were plotted using the STRING database and visualized using Cytoscape 3.9.1. Figure 2 B in Figure 2 The study elucidated the potential role of genes G in gastric cancer. Furthermore, correlation analysis using Timer 2.0 (http: / / timer.cistrome.org / ) with metastasis-related genes (MMP1, MMP3, and VEGFA) showed positive correlations; other correlations are detailed in [link to relevant documentation]. Figure 2 H.

[0104] 2.2 Validating the use of 6-mRNA combinations from surgically resected specimens to predict postoperative peritoneal recurrence in patients with LAGC

[0105] We evaluated the efficacy of 6-mRNA combinations by RT-qPCR in 164 patients with LAGC without peritoneal recurrence after radical resection and 32 patients with LAGC after radical resection with peritoneal recurrence. Receiver operating characteristic (ROC) curve analysis was used to determine the accuracy of single and combined biomarkers in distinguishing cases of peritoneal recurrence. Results are as follows: Figure 3 As shown in B, while individual mRNA markers were effective, the combination of 6-mRNAs demonstrated better diagnostic performance (AUC = 0.902, 95% CI: 0.851–0.953, P < 0.001). Further clinical evaluation of the 6-mRNA classification includes analyzing its ability to detect peritoneal recurrence in conjunction with clinical variables.

[0106] We used a formula based on logistic regression coefficients and constants to estimate the probability of peritoneal recurrence [(3.665 × 6-mRNA biomarker combination) + (5.009 × depth of invasion) + (2.009 × tumor size) + (1.417 × vascular emboli invasion) + (-12.340)], which was plotted as a nomogram for visualizing the prediction of peritoneal recurrence. Figure 3A). After calibrating the model using data from the training cohort, the same statistical parameters were applied to the validation cohort. This process allowed us to stratify patients into low-risk and high-risk categories based on a cutoff value determined by the Youden index. By combining 6-mRNA with clinical variables, we developed a risk stratification assessment (RSA) model that demonstrated excellent predictive ability for peritoneal recurrence, with an AUC of 0.966 (95% CI: 0.944–0.988, P < 0.001), as shown in the figure. Figure 3 As shown in the DF (Data Degree) test, the AUC of the RSA model on the training set was higher than that of the clinical model (0.966 vs 0.869; P = 0.001) and the 6-mRNA combination model (0.966 vs 0.902; P = 0.038). The calibration curve of the model further highlights its excellent predictive accuracy. Figure 3 J).

[0107] Subsequently, the model was applied to an independent external validation cohort of 133 patients with LAGC (including 20 patients with peritoneal recurrence and 113 patients without peritoneal recurrence). This application highlighted the robust predictive power of the RSA model, with an AUC of 0.961 (95% CI: 0.931–0.991, P<0.001). Figure 3 C). In this cohort, the RSA model demonstrated unparalleled sensitivity (65.0%) and specificity (95.6%), outperforming the clinical model (sensitivity 20.0%, specificity 100.0%) and the 6-mRNA combinatorial model (sensitivity 45.0%, specificity 97.3%) in predictive performance. Figure 3 GI). Calibration curve analysis further confirmed the improvement in prediction accuracy. Figure 3 K).

[0108] We further evaluated the potential of the RSA model to improve the cost-effectiveness of clinical decision-making. In our training cohort, using existing clinical parameters, 36.7% of patients were considered high-risk for peritoneal recurrence, while the remaining 63.3% were classified as low-risk. Follow-up results showed that only 13.8% (27 / 196) of patients in the high-risk group and 2.6% (5 / 196) in the low-risk group experienced peritoneal recurrence. This outcome suggests that initial risk classification based on clinical characteristics resulted in 22.9% of patients receiving unnecessary intensive follow-up, while 2.6% of patients did not receive necessary close monitoring. Figure 3 Conversely, applying the 6-mRNA classification to the same cohort resulted in more accurate risk stratification, with 29.6% classified as high-risk and 70.4% as low-risk. Peritoneal recurrence occurred in 14.7% (29 patients) of the high-risk group and 1.5% (3 patients) of the low-risk group. Figure 3 M).

[0109] In the high-risk category of the RSA model, the incidence rate during follow-up was significantly reduced to 12.3% (24 patients) out of 342 follow-up cases. Figure 3 N). This trend is consistent in the external validation cohort; compared with other models, the RSA model significantly reduces unnecessary follow-up for high-risk patients and minimizes missed diagnoses in the low-risk group. Figure 3 This indicates a substantial improvement in clinical decision-making, reducing the likelihood of high-risk patients receiving unnecessary interventions and increasing the detection rate of low-risk patients.

[0110] Further follow-up of the enrolled patients based on the high-risk and low-risk groups of the nomogram revealed that, in the training group, the 5-year RFS was significantly worse in the high-risk group than in the low-risk group (25.0% vs. 59.3%, P<0.0001). A similar difference was observed in the validation set (28.2% vs. 57.4%, P=0.00012). Figure 3 RS).

[0111] 2.3 Validating the predictive value of 6-mRNA combination for peritoneal recurrence in patients with laminar dysplasia of the vagina (LAGC) using endoscopic biopsy specimens.

[0112] In this study, in addition to surgical resection specimens from the training and validation cohorts, we obtained 103 matched endoscopic biopsy specimens, of which 15 showed peritoneal recurrence and 88 did not. Significant high correlations were found in the expression profiles of six genes between the biopsy and surgical specimens. Figure 4 Comparison of gene expression in these matched samples showed no significant difference in the expression of any gene. Figure 4 GL). The AUC for peritoneal recurrence detected by the 6-mRNA combination was 0.899 (95% CI: 0.830-0.967, P<0.001), while the AUC of the RSA model was 0.956 (95% CI: 0.910-1.000, P<0.001), highlighting its effectiveness for preoperative biopsy specimens. Figure 4 M). Calibration curve analysis further confirmed that the RSA model has higher predictive accuracy. Figure 4 N).

[0113] In the biopsy cohort, the RSA model exhibited the highest sensitivity (73.3%) and specificity (97.7%), both exceeding those of the clinical model (sensitivity 33.3%; specificity 100.0%) and the 6-mRNA combination model (sensitivity 40.0%; specificity 97.7%). Figure 4PR). Comparative analysis of the clinical efficacy of different models showed that, compared with the clinical characteristics and 6-mRNA combination model, the RSA model increased the diagnostic rate of peritoneal recurrence in the high-risk group from 10.7% to 13.6%, while decreasing the diagnostic rate in the low-risk group from 3.9% to 1.0%. Figure 4 These findings highlight the ability of RSA models to refine clinical decision-making and reduce unnecessary interventions.

[0114] Similar to previous cohort follow-up studies, we performed a log-rank test on patients with endoscopic biopsy specimens based on the nomogram for high-risk and low-risk groups. The results showed that the 5-year RFS was significantly worse in the high-risk group than in the low-risk group (37.2% vs. 56.7%, P = 0.0056). Figure 4 O).

[0115] 2.4 Validating the prediction of peritoneal recurrence in LAGC patients using 6-mRNA combination in peripheral blood samples

[0116] The primary objective of this study was to develop a liquid biopsy-based detection method for predicting peritoneal recurrence in patients with laminar ectopic gastrointestinal (LAGC). We expanded our study from tissue samples to peripheral blood samples, using a combination of 6-mRNAs to detect mRNA expression in serum samples from 120 LAGC patients (including 102 without peritoneal recurrence and 18 with recurrence). Preliminary analysis was performed using RT-qPCR to assess the expression levels of these mRNAs and evaluate their diagnostic utility. Based on multivariate logistic regression, we constructed a nomogram predicting peritoneal recurrence. Figure 5 A) The AUC of the combination of 6 mRNA biomarkers was 0.903 (95% CI: 0.843–0.963, P<0.001); Figure 5 B), demonstrating its predictive accuracy. The probability of peritoneal recurrence risk was calculated using the logistic regression coefficient formula: [(3.873 × 6-mRNA combination) + (5.226 × invasion depth) + (2.761 × tumor size) + (1.848 × vascular tumor thrombus invasion) + (-15.657)]. Notably, in predicting lymph node metastasis, the liquid biopsy-based RSA model outperformed both the clinical model and the 6-mRNA model. Figure 5 B). After risk stratification based on the Youden index, confusion matrix analysis demonstrated its predictive power. Figure 5 EG), and calibration curve analysis confirmed its excellent performance ( Figure 5 W1). Clinical benefit comparison analysis showed that the RSA model increased the detection rate of peritoneal recurrence in high-risk patients from 11.7% to 14.2%, and decreased the detection rate in low-risk patients from 3.3% to 0.8%. Figure 5Patients were divided into low-risk and high-risk groups based on the Youden index. Follow-up showed that the 5-year RFS in the high-risk group was significantly lower than that in the low-risk group (36.7% vs. 54.9%, P = 0.016). Figure 5 X). This highlights the potential of the RSA model in improving clinical decision-making and optimizing patient management.

[0117] Applying the same statistical model and coefficients used initially in the training cohort, we evaluated an independent external validation cohort comprising 104 patients with LAGC without peritoneal recurrence and 19 patients with recurrence. The AUC was 0.960 (95% CI: 0.928–0.993, P < 0.001), confirming the ability of the RSA model to predict peritoneal recurrence. Figure 5 C). Further analysis using a confusion matrix confirmed that the RSA model was superior to the 6-mRNA combination and clinical characteristic models in predicting peritoneal recurrence. Figure 5 HJ). Calibration curve analysis also confirmed the improved predictive performance of the RSA model. Figure 5 W2). These results demonstrate that tissue-based 6-mRNA combinations were successfully applied to liquid biopsies using peripheral blood specimens, and that the RSA model, which combines serum 6-mRNA data with clinical characteristics, effectively predicted peritoneal recurrence in patients. Analysis of the clinical benefit rates of each model further confirmed the ability of the RSA model to improve the detection rate of peritoneal recurrence in the high-risk group (from 11.4% to 14.6%), while reducing the detection rate in the low-risk group (from 4.1% to 0.8%). Figure 5 QS). Using nomogram risk grouping, time-series tests were performed on the validation cohort. The results showed that the 5-year RFS of the high-risk group was significantly lower than that of the low-risk group (28.9% vs. 57.6%, P<0.0001). Figure 5 Y), highlighting the potential of the RSA model to improve clinical outcomes through more accurate risk stratification.

[0118] In current clinical practice, peripheral blood tumor markers such as CA19-9, CA72-4, and CEA are commonly used indicators for monitoring peritoneal recurrence in patients with latent peritoneal ectopic gas (LAGC). To evaluate whether our RSA model can pre-identify tumor marker-negative patients who may experience peritoneal recurrence, we analyzed peripheral blood samples from a combined training and validation cohort. Among them, 46 patients had no peritoneal recurrence, and 10 had recurrence; all patients tested negative for the aforementioned tumor markers. In the training set, the RSA model (AUC = 0.939, 95% CI: 0.870–1.000) was superior to the clinical characteristic model (AUC = 0.798; 95% CI: 0.637–0.959) and the 6-mRNA combination model (AUC = 0.911, 95% CI: 0.832–0.990) in predicting peritoneal recurrence. Figure 5D). Further analyses using confusion matrix plots and calibration curves confirmed that the RSA model was superior to other models in predicting peritoneal recurrence. Figure 5 KM, Figure 5 W3). We analyzed the clinical benefit rates among different models and again found that the RSA model was significantly superior to the combination of clinical characteristics and 6-mRNA in detecting peritoneal recurrence (W3). Figure 5 TV). Furthermore, survival analysis based on nomogram-based high- and low-risk classifications showed that in patients with negative tumor markers, the 5-year RFS was significantly lower in the high-risk group than in the low-risk group (35.3% vs. 59.0%, P = 0.043). Figure 5 Z), highlighting the potential of the RSA model in improving clinical decision-making and prognosis for patients with LAGC.

[0119] 2.5 In the diagnosis of LAGC patients, the combination of 6-mRNAs identified the presence of peritoneal metastases and micrometastases.

[0120] We used peripheral blood training set data to build a nomogram model to evaluate the early detection capability of 6-mRNA combination for peritoneal metastases and micrometastases (POCY1) in treatment-naïve LAGC patients. Unfortunately, current diagnostic techniques often overlook POCY1 tumors, making them a leading candidate for novel therapies that could significantly improve treatment outcomes in this subgroup of gastric cancer. To validate the predictive ability of the RSA model combined with 6-mRNA and clinical characteristics for POCY1, we referenced two prospective studies of POCY1 patients receiving conversion therapy (NCT03718624 and ChiCTR1800014817). Patients without distant metastases on preoperative abdominal CT were clinically considered to be in the locally advanced stage. POCY1 was detected in 12 out of 66 LAGC patients through diagnostic laparoscopy and free peritoneal cancer cell detection. RT-qPCR analysis of the levels of six mRNAs in preoperative peripheral blood serum revealed that the RSA model (AUC = 0.941, 95% CI: 0.874–1.000) was significantly superior to clinical characteristics (AUC = 0.852, 95% CI: 0.722–0.982) and the 6-mRNA combination (AUC = 0.904, 95% CI: 0.832–0.976) in predicting P0CY1. Figure 6 AB). Furthermore, confusion matrix and calibration curve analysis validated the superior prediction accuracy (AB). Figure 6 EG, 6Q). Comparative analysis of clinical benefits showed that the RSA model increased the detection rate of intraperitoneal micrometastases in high-risk patients from 13.6% to 16.7%, and reduced the misdiagnosis rate of peritoneal recurrence in low-risk patients from 4.5% to 1.5%. Figure 6Patients were divided into low-risk and high-risk groups based on the Youden index. Our follow-up results showed that the 5-year RFS in the high-risk group was significantly lower than that in the low-risk group (37.5% vs. 61.8%, P = 0.029). Figure 6 The results indicate that the RSA model has the potential to improve the prognosis of LAGC patients and guide clinical decision-making.

[0121] Next, we used serum samples from 95 patients with LAGC in two previously registered prospective cohort studies (NCT02555358 and NCT01516944), and observed that 15 patients (15.8%) developed PM after undergoing diagnostic laparoscopy prior to treatment. Using an earlier-developed nomogram model, the RSA model (AUC = 0.970, 95% CI: 0.936–1.000) showed better predictive accuracy for PM than clinical characteristics (AUC = 0.884, 95% CI: 0.736–0.964) and the 6-mRNA panel (AUC = 0.863, 95% CI: 0.701–0.941). Figure 6 CD). Further analysis using confusion matrix plots and calibration curves confirmed the excellent predictive performance. Figure 6 HJ, 6R). Notably, our analysis of the clinical benefit rates of different models showed that the RSA model increased the PM detection rate in high-risk patients from 11.6% to 14.7%, while reducing the false recognition rate of peritoneal recurrence in the overall risk group from 4.2% to 1.1%. Figure 6 (NP). Follow-up studies showed that the 5-year recurrence-free survival (RFS) rate in the high-risk group was significantly lower than that in the low-risk group (29.0% vs. 58.3%, P = 0.0045). Figure 6 (T). Overall, our systematic and thorough approach to biomarker discovery and validation has led to the identification and validation of a new gene prediction panel that significantly enhances the management of gastric cancer by improving the detection rate of peritoneal metastases, potentially improving overall survival.

[0122] Furthermore, we evaluated the diagnostic performance of the 6-mRNA combination in serum samples from patients with LAGC and assessed its efficacy in other gastrointestinal malignancies, including colorectal cancer (n=35), hepatocellular carcinoma (n=38), and pancreatic cancer (n=29). Interestingly, our 6-mRNA combination demonstrated higher diagnostic accuracy (AUC=0.879) in predicting PM than all other gastrointestinal cancers evaluated (colorectal cancer: AUC=0.756; hepatocellular carcinoma: AUC=0.642; pancreatic cancer: AUC=0.682). Figure 6DeLong's trial further confirmed that the 6-mRNA combination had high specificity for diagnosing PM in LAGC compared to other gastrointestinal cancers (P<0.001 for colorectal cancer, hepatocellular carcinoma, and pancreatic ductal carcinoma, compared to gastric cancer). In summary, these findings suggest that our 6-mRNA biomarker combination is highly specific in distinguishing LAGC patients from other gastrointestinal malignancies, even as a non-invasive blood test.

Claims

1. The use of reagents for detecting biomarkers in samples in any of the following: A1. Application in the preparation of products for diagnosing peritoneal recurrence after gastric cancer surgery; A2. Application in the preparation of products for predicting peritoneal recurrence after gastric cancer surgery; The biomarkers are a combination of BUB1, CKS2, PCNA, CHEK1, NEK2, and NCAPG2.

2. The application as described in claim 1, characterized in that, The products include reagent kits, test strips, primers, probes, and chips.

3. The application as described in claim 2, characterized in that, The kit also includes one or more substances selected from the group consisting of: containers, instructions for use, positive controls, negative controls, buffers, auxiliaries, or solvents.

4. The application as described in claim 2, wherein the chip is a gene chip.

5. The application as described in claim 2, characterized in that, The kit also includes mRNA expression level auxiliary detection reagents, which include reaction reagents for primer-corresponding amplicon visualization, RNA extraction reagents, reverse transcription reagents, cDNA amplification reagents, and / or standards used for preparing standard curves.

6. In the application as described in claim 5, the reaction reagents for visualizing the amplicon corresponding to the primer include reagents used in agarose gel electrophoresis, enzyme-linked gel electrophoresis, chemiluminescence immunoassay, in situ hybridization, and fluorescence detection to visualize the amplicon.

7. The application as described in claim 1, characterized in that, The samples include tissue, fine-needle biopsy samples, body fluids containing cells, or samples collected by skin swabs, oral swabs, or nasal swabs.

8. The application as described in claim 7, characterized in that, The cellular fluids include blood, ascites, cerebrospinal fluid, peritoneal fluid, pleural fluid, sputum, saliva, urine, feces, lymph, and lavage fluid.

9. The application as described in claim 8, characterized in that, The blood includes serum, plasma, peripheral blood, and blood cells.

10. The application as described in claim 7, characterized in that, The tissue in question is bone marrow.

11. The application as described in claim 9, characterized in that, The sample is taken from the object to be tested, which includes humans or non-human mammals.

12. The application as described in claim 11, characterized in that, The object to be tested is a person.

13. The application as described in claim 1, characterized in that, The gastric cancer mentioned is locally advanced gastric cancer.

14. A product for diagnosing peritoneal recurrence after gastric cancer surgery or predicting the risk of peritoneal recurrence after gastric cancer surgery, characterized in that, The product includes a detection reagent for detecting the content of biomarkers, wherein the biomarkers are a combination of BUB1, CKS2, PCNA, CHEK1, NEK2, and NCAPG2.

15. The product as described in claim 14, characterized in that, The products include reagent kits, test strips, primers, and probes.

16. The product as described in claim 15, characterized in that, The kit is a qPCR kit.

17. The product as described in claim 15, characterized in that, The products also include reagents commonly used in PCR reactions, RT-PCR derivatization reactions, 3SR amplification, LCR, SDA, NASBA, TMA, SYBR Green, TaqMan probes, molecular beacons, two-hybrid probes, composite probes, ISH, microarrays, Southern blotting, Northern blotting, and multianalytical spectroscopy assays.

18. A system for diagnosing peritoneal recurrence after gastric cancer surgery, or for predicting the risk of peritoneal recurrence after gastric cancer surgery, the system comprising: The input unit is used to acquire the expression level detection data of biomarkers in the sample, wherein the biomarkers are selected from a combination of BUB1, CKS2, PCNA, CHEK1, NEK2, and NCAPG2. The result determination unit is used to input the data from the detection unit into the computer program for processing and determination, and to obtain the comparison result by comparing it with the set threshold value in the computer. The output unit is used to output the results of the result determination unit, and to determine the probability of whether the object corresponding to the test sample has peritoneal recurrence after gastric cancer surgery or the probability of peritoneal recurrence after gastric cancer surgery.

19. A method for constructing a predictive model for the risk of peritoneal recurrence after gastric cancer surgery, characterized in that, The method includes: training an algorithm model using machine learning methods to diagnose whether peritoneal recurrence occurs after gastric cancer surgery, or to predict the risk of peritoneal recurrence after gastric cancer surgery; the algorithm model processes data to obtain an assessment result of peritoneal recurrence after gastric cancer surgery, the assessment result of peritoneal recurrence after gastric cancer surgery includes the probability of peritoneal recurrence after gastric cancer surgery and important characteristic factors, the important characteristic factors reflecting the correlation with peritoneal recurrence after gastric cancer surgery; the important characteristic factors are expression level data of a combination of BUB1, CKS2, PCNA, CHEK1, NEK2, and NCAPG2 extracted from the information parameters of the subject to be tested.

20. The method as described in claim 19, characterized in that, The key characteristic factors also include data on invasion depth, tumor size, and vascular tumor thrombus invasion extracted from the information parameters of the object to be detected.

21. The method as described in claim 19, characterized in that, One of the RSA models in the algorithm is: [(3.665 × 6-mRNA biomarker combination) + (5.009 × invasion depth) + (2.009 × tumor size) + (1.417 × vascular tumor thrombus invasion) + (-12.340)]. This RSA model is used to assess the risk of peritoneal recurrence after gastric cancer surgery.

22. The method as described in claim 19, characterized in that, The machine learning methods include decision tree learning, random forest learning, K-nearest neighbor algorithm learning, Naive Bayes learning, support vector machine learning, neural network learning, and Adaboost learning.

23. A computer-readable storage medium having a computer program stored thereon, characterized in that, The computer program is stored in a computer-readable storage medium, and when executed by a processor, the computer program performs the function of the gastric cancer postoperative peritoneal recurrence risk prediction model constructed by the method of claim 19.

24. A computer-based evaluation method for assessing whether peritoneal recurrence occurs after gastric cancer surgery, or the risk of peritoneal recurrence after gastric cancer surgery, characterized in that, The method includes: Acquire data to be processed, wherein the data to be processed includes the expression level detection data of biomarkers in the sample to be tested, and the biomarkers are a combination of BUB1, CKS2, PCNA, CHEK1, NEK2, and NCAPG2; The data to be processed is input into the gastric cancer postoperative peritoneal recurrence risk prediction model constructed by the method of claim 19 to obtain the assessment results of whether the gastric cancer postoperative peritoneum has recurred or the risk of gastric cancer postoperative peritoneal recurrence. The results of the assessment of whether peritoneal recurrence occurred after gastric cancer surgery or the risk of peritoneal recurrence after gastric cancer surgery are output from the data to be processed.

25. The method as described in claim 24, characterized in that, The data to be processed also includes detection data on the subject's invasion depth, tumor size, and vascular tumor thrombus invasion.

Images