Application of circulating tumor cells and circulating tumor DNA in peritoneal lavage fluid in predicting metachronous peritoneal metastasis after radical resection of gastric cancer

Abstract

The present invention relates to the field of biotechnology, and specifically to the detection of circulating tumor cells and circulating tumor DNA in peritoneal lavage fluid and their application in predicting the risk of metachronous peritoneal metastasis after radical resection of gastric cancer. How to identify micrometastases in the abdominal cavity in patients with advanced gastric cancer and delay their occurrence and development has become an urgent problem to be solved in the field of gastric surgery. The present invention provides an application in predicting the risk of metachronous peritoneal metastasis after radical resection of gastric cancer by detecting circulating tumor cells and circulating tumor DNA (ctDNA) in peritoneal lavage fluid. The present invention can screen, judge and follow up occult peritoneal metastasis at multiple stages, provide more comprehensive and sensitive information, and help guide clinical decision-making and the formulation of treatment plans.

Description

Technical Field

[0001] The present invention relates to the field of biotechnology, in particular to the detection of circulating tumor cells and circulating tumor DNA in peritoneal lavage fluid and their application in predicting the risk of metachronous peritoneal metastasis after radical resection of gastric cancer. Background Art

[0002] Peritoneal metastasis (PM) is one of the most common metastatic forms of many cancers at the end stage. The incidence of peritoneal metastasis varies with different types of cancer. In theory, peritoneal metastasis can occur in intraperitoneal tumors regardless of the tissue they originate from.

[0003] According to the "seed-soil" hypothesis, peritoneal metastasis is a multi-step process, including cancer cells separating from the primary tumor, surviving in the peritoneal microenvironment, adhering to peritoneal mesothelial cells, and infiltrating and growing. The primary cancer cell lesion penetrates the serosal layer into the peritoneal cavity to form free peritoneal cancer cells - "seeds" are the pathological basis for the occurrence of peritoneal metastasis. Therefore, how to identify detached and free peritoneal cancer cells in cancer patients has become a core issue in clinical peritoneal metastasis prevention and treatment decisions.

[0004] Peritoneal metastasis is usually a manifestation of advanced cancer, and the early detection rate is not high. Current screening methods for peritoneal metastasis include but are not limited to imaging examinations (CT / MR / PET-CT / B-ultrasound), blood tumor marker detection, ascites or peritoneal lavage fluid cytology, etc. Once peritoneal metastasis is confirmed by imaging methods, the patient's expected survival is short and the prognosis is extremely poor. Therefore, the key to preventing peritoneal metastasis is early detection and treatment of primary cancer, as well as close monitoring of high-risk patients.

[0005] Gastric cancer (GC) ranks fifth in incidence and fourth in mortality worldwide. Peritoneal metastasis is one of the important causes of death in gastric cancer patients. About 10%-30% of gastric cancer patients are found to have peritoneal metastasis at the time of initial diagnosis, and more than half of stage II-III gastric cancer patients will develop peritoneal metastasis within 5 years after radical surgery. In addition, as the distribution range of peritoneal metastasis lesions expands, the patient's survival rate decreases accordingly. A study on the prognosis of patients with peritoneal metastasis of gastric cancer showed that the 2-year survival rates of patients with peritoneal dissemination limited to the upper part of the transverse colon and patients with several nodules in the distal peritoneum below the transverse colon were 21% and 8%, respectively, while the 2-year survival rate of patients with a large number of metastatic lesions in the distal peritoneum was only 4% (reference: Yonemura Y, Bandou E, Kawamura T, et al. Quantitative prognosticindicators of peritoneal dissemination of gastric cancer[J]. Eur J Surg Oncol, 2006, 32(6):602-6.). Peritoneal metastasis is also the leading cause of death in patients with gastric cancer. Therefore, actively carrying out the prevention and treatment of peritoneal metastasis of gastric cancer will undoubtedly help improve the overall treatment level of gastric cancer in my country. Peritoneal metastasis is a multi-step process, including cancer cells separating from the primary tumor, surviving in the peritoneal microenvironment, adhering to peritoneal mesothelial cells and infiltrating and growing (see Kanda M, Kodera Y. Molecular mechanisms of peritoneal dissemination in gastric cancer. World J Gastroenterol. 2016 Aug 14;22(30):6829-40. doi: 10.3748 / wjg.v22.i30.6829. PMID: 27570420; PMCID: PMC4974582.). Therefore, how to identify micrometastases in the peritoneal cavity in patients with advanced gastric cancer and delay their occurrence and development has become an urgent problem to be solved in the field of gastric surgery.

[0006] Currently, the screening methods for peritoneal metastasis include but are not limited to imaging examinations (CT / MR / PET-CT / B-ultrasound), blood tumor marker detection, ascites or peritoneal lavage fluid cytology, etc. Once peritoneal metastasis is confirmed by imaging methods, the patient's expected survival is short and the prognosis is extremely poor. Therefore, the key to preventing peritoneal metastasis is early detection and treatment of primary cancer, as well as close monitoring of high-risk patients. Cytological examination of peritoneal lavage fluid or ascites is the gold standard for diagnosing free tumor cells in the abdominal cavity. Its positive rate is related to many factors, including whether enough peritoneal lavage fluid is retained during surgery, whether the sample is sent for examination in time, and the atypical morphology of exfoliated cells, which ultimately leads to a low positive detection rate. A meta-analysis that included 19 studies showed that the frequency of free cancer cells detected in the peritoneal lavage fluid of gastric cancer patients was 6.25-54.4%, and was affected by the depth of tumor infiltration, degree of differentiation, and the use of chemotherapy (reference: Kołomańska MM, Głuszek S. Free cancer cells in gastric cancer - methods of detection, clinical and prognostic importance (meta-analysis)[M]. Contemp Oncol (Pozn), 2020, 24(1):67-74.). Given the low positive rate of cytological examinations and the difficulty in diagnosing occult peritoneal metastasis, new molecular markers and detection technologies that can diagnose peritoneal metastasis at an early stage need to be verified.

[0007] Circulating tumor DNA (ctDNA) refers to DNA fragments released into the blood circulation by tumor cells during their growth and division. These DNA fragments carry the genetic information of tumor cells, including gene mutations, copy number variations, gene fusions, etc. Compared with the DNA released by normal cells, ctDNA released by tumor cells may contain specific tumor-related mutations.

[0008] By analyzing the gene mutation, copy number variation, gene fusion and other information in ctDNA, we can understand the genetic variation, tumor progression, drug resistance and other aspects of the tumor, providing an important basis for tumor diagnosis, monitoring treatment effects and prognosis evaluation. It has broad application prospects in early tumor screening, individualized treatment, monitoring efficacy, and predicting recurrence.

[0009] The level of ctDNA in the extracellular environment is usually low, especially in patients with local lesions, and the half-life of protein-bound DNA and naked DNA in plasma at 37°C is 26.8-161.2 minutes. This short half-life characteristic enables real-time dynamic monitoring of patients' tumor burden, but also requires highly complex detection and quantification technology for the detection of ctDNA.

[0010] The research on the detection of ctDNA in the blood has been very extensive, including the screening of early cancer, the detection of minimal residual disease (MRD), and the application in the guidance and detection of treatment of advanced diseases. However, ctDNA can also be obtained from non-blood sources, including pleural effusion, ascites, urine, and cerebrospinal fluid. A large amount of cell-free DNA (cfDNA) with a length of about 140-170bp from white blood cells in the blood will become background noise when detecting ctDNA, and the relative lack of clonal hematopoiesis in non-blood samples makes ctDNA have a relatively higher concentration and variant allele frequency, which can be a supplementary condition for the blood ctDNA detection results. WO 2023 / 130670 Al discloses a method for extracting free DNA (cfDNA) from serum and sequencing and building a library to classify and predict cancer types.

[0011] Biomarkers in peritoneal lavage fluid may be more closely related to peritoneal metastasis than blood. In 2021, Van'tErve I et al. used ddPCR to evaluate tumor ctDNA in plasma and peritoneal lavage fluid of 20 patients with colorectal cancer and peritoneal metastasis. The results showed that the ctDNA level in peritoneal lavage fluid was higher than that in plasma (p < 0.0001), so plasma ctDNA may not be a sensitive marker for detecting colorectal cancer with peritoneal metastasis. From the perspective of the pathogenesis of peritoneal metastasis, it is characterized by direct penetration of the primary tumor into the peritoneum or iatrogenic dissemination caused by surgery. On the other hand, the peritoneal-plasma barrier may limit the release of cfDNA from the peritoneum into the systemic circulation, which may explain the higher levels of ctDNA in peritoneal lavage fluid (see VAN'T ERVE I, ROVERS KP, CONSTANTINIDES A, et al. Detection of tumor-derived cell-free DNA from colorectal cancer peritoneal metastases in plasma and peritoneal fluid [J]. J Pathol Clin Res, 2021, 7(3): 203-208. HASOVITS C, CLARKE S. Pharmacokinetics and pharmacodynamics of intraperitoneal cancer chemotherapeutics [J]. Clin Pharmacokinet, 2012, 51(4): 203-24.).

[0012] Next Generation Sequencing (NGS) is a fast, accurate, and efficient DNA or RNA sequencing technology, also known as deep sequencing or high-throughput sequencing. It can obtain a large amount of genomic or transcriptome information in a short period of time by sequencing millions to billions of DNA or RNA fragments in parallel. The advantages of high-throughput sequencing include high throughput, which can sequence a large number of samples or fragments at the same time, improve sequencing efficiency, and greatly shorten sequencing cycles; high accuracy, compared with traditional sequencing technology, high-throughput sequencing has higher accuracy and resolution; diversity, which can be used to sequence DNA or RNA fragments of different lengths, and is suitable for various research fields.

[0013] Through high-throughput sequencing technology, a large amount of genomic information can be quickly obtained to reveal the structure, function and regulatory mechanism of the genome, thereby promoting the progress of disease diagnosis and treatment research.

[0014] NGS-based tissue-informed (tumor-informed) ctDNA detection is a liquid biopsy method that uses the molecular characteristics of tumor tissue samples NGS sequencing as prior information to accurately match the ctDNA data in peritoneal lavage fluid or ascites samples, thereby improving the accuracy of circulating peritoneal lavage fluid or ascites ctDNA detection and the reliability of clinical application. The main features of tissue-informed ctDNA detection include: using tumor tissue information, through molecular analysis such as genomics and transcriptomics on tumor tissue samples, to obtain the genetic characteristics and mutation information of the tumor, and establish a molecular feature map of the tumor. Based on this molecular feature map, combined with the analysis of ctDNA data in peritoneal lavage fluid or ascites, tumor-related variations in ctDNA can be more accurately identified and interpreted.

[0015] By conducting a priori ctDNA testing, we can continuously monitor the variation in ctDNA in peritoneal lavage fluid or ascites, understand the development status, treatment effect and recurrence risk of intra-abdominal tumors in real time, guide individualized treatment strategies, select more effective targeted therapy drugs, and improve treatment effects and survival rates. Summary of the invention

[0016] The purpose of the present invention is to provide a combined detection of circulating tumor cells in peritoneal lavage fluid and circulating tumor DNA in peritoneal lavage fluid and its application in predicting the risk of metachronous peritoneal metastasis after radical resection of gastric cancer.

[0017] The present invention improves the NGS-based ctDNA detection technology currently commonly used in plasma and uses it in peritoneal lavage fluid or ascites, in order to achieve early screening of occult peritoneal metastasis in malignant solid tumors when the positive rate of cytological examination is low and the diagnosis of occult peritoneal metastasis is difficult. Furthermore, the present invention combines the detection of circulating tumor DNA in peritoneal lavage fluid with the detection of circulating tumor cells (CTC) in peritoneal lavage fluid, making the risk prediction of metachronous peritoneal metastasis after radical resection of gastric cancer more accurate.

[0018] The first aspect of the present invention provides the use of a reagent for detecting circulating tumor cells in peritoneal lavage fluid and a reagent combination for detecting circulating tumor DNA content in peritoneal lavage fluid in the preparation of a kit for predicting the risk of metachronous peritoneal metastasis after radical resection of gastric cancer.

[0019] The metachronous peritoneal metastasis after radical resection of gastric cancer refers to the appearance of peritoneal metastases after the initial diagnosis of gastric cancer / after treatment of the primary lesion, rather than being discovered at the same time or when the primary lesion is diagnosed.

[0020] The reagents include: a reagent for detecting circulating tumor cells in peritoneal lavage fluid based on cytological detection, and a reagent for detecting the content of circulating tumor DNA in peritoneal lavage fluid based on the second-generation sequencing method of tumor tissue.

[0021] The method for detecting circulating tumor cells in peritoneal lavage fluid based on cytological detection is a commonly used method in the art; the method for detecting circulating tumor DNA content in peritoneal lavage fluid based on tumor tissue second-generation sequencing method comprises the following steps:

[0022] A) Centrifugation of peritoneal lavage fluid to separate the supernatant;

[0023] B) Extraction of circulating tumor DNA from peritoneal lavage fluid: Add 20 mg / mL proteinase K and 20% SDS to the supernatant, mix well, and incubate at 50-70°C for 10-30 min; add DNA binding solution and magnetic beads to allow circulating tumor DNA to fully bind to the magnetic beads; centrifuge, let stand, and remove the supernatant; add DNA washing solution, mix well, centrifuge, let stand, and remove the supernatant; rinse the magnetic beads with 70-90% ethanol; add DNA elution solution to completely release free DNA; centrifuge, let stand, collect the supernatant, and store at -20°C±5°C;

[0024] C) Library construction, capture and sequencing

[0025] The library of ctDNA samples from peritoneal lavage fluid was constructed, and the target region was captured, hybridized, eluted, and sequenced after capture;

[0026] D) ctDNA detection:

[0027] Software was used to identify variants and screen gene mutations in tumor tissues. The tissue-a priori MRD detection method was used to screen gene mutations in peritoneal lavage fluid. When defining the MRD status of the samples, the significance of differences at the sample level was corrected using the multi-variant joint confidence probability method.

[0028] The above steps A) and B) are steps for pre-treating circulating tumor DNA in peritoneal lavage fluid: centrifuging the peritoneal lavage fluid to separate the supernatant; adding 20 mg / mL proteinase K and 20% SDS to the supernatant, mixing, and incubating at 50-70°C for 10-30 min; adding DNA binding solution and magnetic beads to fully bind the circulating tumor DNA to the magnetic beads; centrifuging, standing, and removing the supernatant; adding DNA washing solution, mixing, centrifuging, standing, and removing the supernatant; rinsing the magnetic beads with 70-90% ethanol; adding DNA elution solution to completely release free DNA; centrifuging, standing, collecting the supernatant, and storing at -20°C±5°C;

[0029] Furthermore, the reagent for detecting the ctDNA content in the peritoneal lavage fluid predicts the risk of metachronous peritoneal metastasis after radical resection of gastric cancer by comparing and analyzing the quantitative detection results of ctDNA in the peritoneal lavage fluid. It is judged whether the patient has metachronous peritoneal metastasis after radical resection of gastric cancer according to the ctDNA in the peritoneal lavage fluid reaching a certain threshold level.

[0030] The ctDNA quantitative detection method can refer to the quantitative detection method of cfDNA in plasma or other body fluid samples in the prior art to obtain the ctDNA quantitative detection result.

[0031] Furthermore, the peritoneal lavage fluid is the peritoneal lavage fluid obtained by lavaging the peritoneal cavity with 250 ml of 0.9% saline during surgery or during non-surgery.

[0032] The method for collecting peritoneal lavage is as follows: during the operation, 250 ml of 0.9% normal saline is used to lavage the peritoneal cavity and then aspirate; during the non-surgical period, the patient's peritoneal cavity is punctured under the guidance of imaging techniques such as ultrasound / CT, 250 ml of 0.9% normal saline is infused and retained for an appropriate period of time before drainage.

[0033] Furthermore, the non-surgical period refers to: perioperative period, systemic treatment cycle and life survival period.

[0034] The second aspect of the present invention provides a method for detecting the ctDNA content in peritoneal lavage fluid, comprising the steps of pre-treating circulating tumor DNA in peritoneal lavage fluid:

[0035] Pre-treat the circulating tumor DNA in the peritoneal lavage fluid: centrifuge the peritoneal lavage fluid and separate the supernatant; add 20 mg / mL proteinase K and 20% SDS to the supernatant, mix well, and incubate at 50-70°C for 10-30 min; add DNA binding solution and magnetic beads to fully bind the circulating tumor DNA to the magnetic beads; centrifuge, let stand, and remove the supernatant; add DNA washing solution, mix well, centrifuge, let stand, and remove the supernatant; rinse the magnetic beads with 70-90% ethanol; add DNA elution solution to completely release the free DNA; centrifuge, let stand, collect the supernatant, and store at -20°C±5°C;

[0036] Quantitative detection of circulating tumor DNA in peritoneal lavage fluid.

[0037] The steps for pre-processing circulating tumor DNA in peritoneal lavage fluid are as follows:

[0038] A) Place the peritoneal lavage fluid in a centrifuge tube and centrifuge at 2000-4000 rpm (preferably 3000 rpm) for 10-40 min (preferably 30 min) to separate the supernatant;

[0039] B) Extraction of circulating tumor DNA from peritoneal lavage fluid:

[0040] Step A) Add 20 mg / mL proteinase K and 20% SDS to the obtained supernatant;

[0041] Invert to mix, and incubate at 50-70°C (preferably 60°C) for 10-30 min (preferably 20 min);

[0042] Add DNA binding solution and magnetic beads; shake to allow circulating tumor DNA to fully bind to the magnetic beads; centrifuge, let stand, and remove the supernatant;

[0043] Add DNA washing solution, shake and mix; centrifuge, let stand, and remove the supernatant;

[0044] Rinse the magnetic beads with 70-90% (preferably 80%) ethanol (once or more); remove the sample from the magnetic rack, add DNA elution buffer, and shake to completely release the free DNA; centrifuge, let stand, aspirate the supernatant into a new container, and store at -20℃±5℃.

[0045] The method for detecting the ctDNA content in the peritoneal lavage fluid of the present invention, and the subsequent step of ctDNA quantitative detection step can refer to the detection method of cfDNA in plasma or other body fluid samples in the prior art to obtain the ctDNA detection result.

[0046] The ctDNA test results are obtained for comparison and analysis (the library can be constructed as in the embodiment method provided in the present invention, or the sample library of gastric cancer cfDNA in plasma or other body fluid samples already available from commercial companies can be directly used). The presence or absence of ctDNA in the peritoneal lavage fluid or reaching a certain threshold level (determining whether ctDNA is positive or negative) is used to determine whether the patient has metachronous peritoneal metastasis after radical resection of gastric cancer.

[0047] In the above detection method, the proteinase K, SDS, DNA binding solution, magnetic beads, DNA washing solution and DNA elution solution used are all commercially available products.

[0048] The method for detecting the ctDNA content in the peritoneal lavage fluid of the present invention specifically comprises the following steps:

[0049] A) Place the peritoneal lavage fluid in a centrifuge tube (preferably a 50ml-100ml centrifuge tube), centrifuge at 3000rpm for 30min, and separate the supernatant;

[0050] B) Extraction of free tumor DNA from peritoneal lavage fluid:

[0051] (b1) The peritoneal lavage fluid of each patient was extracted into 3 tubes, each tube had 4 ml, and the sample number was marked;

[0052] (b2) Add 4 mL of peritoneal lavage fluid, 60 μL of proteinase K (20 mg / mL), and 200 μL of 20% SDS into each centrifuge tube in sequence;

[0053] (b3) Invert to mix, and incubate at 60°C for 20 min;

[0054] (b4) Add 5 mL of free DNA binding solution and 60 μL of magnetic beads mixed by vortexing;

[0055] (b5) Vortex for 10 seconds to mix thoroughly. Vortex for 5-10 minutes to allow free DNA to fully bind to the magnetic beads;

[0056] (b6) Briefly centrifuge to collect the liquid on the tube cap and tube wall, and place the centrifuge tube on a magnetic rack until the liquid becomes clear (3-5 minutes);

[0057] (b7) removing the supernatant;

[0058] (b8) Add 500 μL of free DNA washing solution and vortex to mix;

[0059] (b9) Briefly centrifuge to collect the liquid on the tube cap and tube wall, and place the centrifuge tube on a magnetic rack until the liquid becomes clear (1-2 minutes);

[0060] (b10) removing the supernatant;

[0061] (b11) Add 500 μL 80% ethanol and shake to mix;

[0062] (b12) Briefly centrifuge to collect the liquid on the tube cap and tube wall, and place the centrifuge tube on a magnetic rack until the liquid becomes clear (1-2 minutes);

[0063] (b13) removing the supernatant;

[0064] (b14) Repeat steps (b11) to (b13) and rinse the magnetic beads once more with 80% ethanol;

[0065] (b15) Use a 10 μL pipette to suck out the liquid remaining at the bottom of the centrifuge tube and dry it at room temperature for 3-5 minutes. Note: The magnetic beads should not be over-dried.

[0066] (b16) Remove the sample from the magnetic rack, add 30-50 μL of free DNA elution buffer, and shake for 3-5 minutes to completely release the free DNA;

[0067] (b17) Briefly centrifuge to collect the liquid on the tube cap, and place the centrifuge tube on a magnetic rack until the liquid is clear. Carefully pipette the supernatant into a new centrifuge tube and store at -20℃±5℃.

[0068] C) Library construction, capture and sequencing

[0069] (c1) Library construction

[0070] The KAPA Hyper Preparation Kit (Kapa Biosystems, USA) was used to construct the library of ctDNA samples from peritoneal lavage fluid. The concentration of the constructed library was measured using the Qubit dsDNA HS (High Sensitivity) Assay Kit (Themo Fisher, USA), and the size of the library fragment was evaluated using the Agilent Bioanalyzer 2100 (Agilent, USA).

[0071] The amount of ctDNA input into peritoneal lavage fluid was 200 ng, which was used as a reference for the input into late-stage tissues.

[0072] The library was constructed based on the 60ng input of plasma, which failed to pass the quality control. Further testing of the nucleic acid 2100 quality control fragment revealed that the cfDNA in the peritoneal lavage fluid accounted for a small proportion, and most of it was gDNA. According to the 60ng input, the actual input of cfDNA should be less than 10ng. Therefore, the present invention refers to the input of nucleic acid for library construction of late-stage tissues for library construction.

[0073] (c2) Target region capture and sequencing

[0074] The xGen Hybridization and Wash Kit (IDT, USA) was used to capture the target region. The panel used for capture covers approximately 2.2MB of the human genome, including 769 genes. After hybridization, washing, and capture, sequencing was performed using DNBSEQ-T7.

[0075] D) ctDNA detection:

[0076] (d1) Variant identification

[0077] First, Trimmomatic (v0.36) software was used to remove adapters and low-quality sequencing products (reads). Clean reads were aligned to the human hg19 reference genome using BWA aligner (v0.7.17) software. Next, Picard (v2.23.0) software was used to classify and remove duplications. SNVs and InDels were identified and detected using VarDict (v1.5.1) software, and complex mutations were identified using FreeBayes (v1.2.0). The filtering of QC data such as mutation quality and chain preference are listed in the original mutation list. In addition, mutations in low-complexity repeats and segmental duplications that matched low-mapping regions defined in ENCOD, as well as mutations in the sequencing-specific errors (SSEs) list developed and validated in-house, were removed.

[0078] (d2) Screening for gene mutations in tumor tissues

[0079] First, mutations from germline or hematopoietic sources are filtered out. Mutations that meet any of the following criteria will be filtered out: (1) the variant frequency (VAF) of the mutation from peripheral blood is not less than 5%, or (2) the VAF value of the mutation from peripheral blood is less than 5%, but the VAF value is not more than 5 times the VAF value of the same point in the matched tissue sample, or (3) the mutation can be found in the public gnomAD population database, with a smaller allele frequency (MAF) of not less than 2%. The remaining gene mutations will be further filtered by quality conditions. When screening tumor tissue mutations, each mutation is supported by at least 5 reads, the detection limit of SNV is 4%, and the detection limit of InDel is 5%, which are used as the conditions for screening tumor tissue mutations.

[0080] These threshold criteria were determined during the analytical validation of the NGS Panel performance and were measured in terms of sensitivity, specificity, repeatability, and reproducibility.

[0081] (d3) Screening for gene mutations in peritoneal lavage fluid

[0082] A tissue-a priori MRD detection method was used. First, ctDNA mutations not detected in the initial tissue sequencing were excluded, and only somatic variants detected in tumor tissue were monitored in plasma. Each monitored ctDNA mutation was statistically tested using a background noise baseline model to eliminate the impact of technical noise. The remaining variant signals represented the noise signals of each variant direction (Subtype) of each chromosome coordinate (coordinate) by removing true tumor-derived somatic variants and clonal hematopoietic (CH) mutations through paired tissue and PBL sequencing, and the Weibull distribution or inverse Gamma distribution was fitted to form a noise model with single-base resolution. The zero inflation effect in the noise baseline model was approximated by Monte Carlo sampling simulation to obtain the zero percentage weighted probability of a given variant. Specifically, 10,000 random samples were taken from the modeled distribution, including n_nonzero non-zero VAFs and n_zero zero VAFs, where n_nonzero and n_zero were calculated as the product of the non-zero and zero percentages, respectively, with a predetermined total of 1,0000. Each VAF in this list is further fed as a noise probability into a binomial test parameterized by the observed alt allele support reads and total support reads. The expected value is the average of all p values, and a threshold (0.05) is set. In order to avoid the loss of specificity caused by multiple comparisons when defining the MRD status of samples, we corrected the sample-level difference significance by the method of multi-variant joint confidence probability, applying the formula:

[0083]

[0084] n is the number of tracked mutations, k is the number of mutations with p<=0.05, and C represents the number of combinations. is the p_value of the positive site;

[0085] Finally, the significance level of the sample compared with the noise baseline was obtained. In this study, the significance threshold P of the sample level was set at 0.01. When the P value of the peritoneal lavage fluid test was lower than the threshold, the peritoneal lavage fluid sample was determined to be MRD positive.

[0086] The third aspect of the present invention provides a detection system for predicting the risk of metachronous peritoneal metastasis after radical resection of gastric cancer, wherein the detection system implements the use of a combination of a reagent for detecting circulating tumor cells in peritoneal lavage fluid and a reagent for detecting circulating tumor DNA content in peritoneal lavage fluid as described above in the preparation of a kit for predicting the risk of metachronous peritoneal metastasis after radical resection of gastric cancer.

[0087] The present invention has the advantages that:

[0088] The present invention provides an application in predicting the risk of metachronous peritoneal metastasis after radical resection of gastric cancer by monitoring the quantitative detection level of ctDNA through qualitative cytological examination of peritoneal lavage fluid and quantitative detection of circulating tumor DNA (ctDNA) based on next-generation sequencing (NGS) of tumor tissue, starting from the diagnosis of gastric cancer in patients.

[0089] Qualitative cytological examination and quantitative ctDNA detection in the whole process of management after the diagnosis of gastric cancer have improved the accuracy of predicting the risk of metachronous peritoneal metastasis after radical resection of gastric cancer, and facilitated the prediction and identification of occult peritoneal tumor metastasis during the patient's perioperative period, systemic treatment cycle and survival period, so as to formulate effective treatment plans in time, providing a fast, accurate and efficient scientific basis for extending the patient's life cycle to the greatest extent possible, and has extremely high clinical application value.

[0090] Using NGS-based methods to detect ctDNA in patients' peritoneal lavage fluid can screen, judge and follow up occult peritoneal metastasis at multiple stages, providing more comprehensive and sensitive information to help guide clinical decision-making and the formulation of treatment plans. The following are the roles of using ctDNA testing at different stages:

[0091] (1) Screening for occult peritoneal metastasis upon admission: When cancer patients with high-risk factors for peritoneal metastasis (imaging suggests possible peritoneal metastasis / imaging suggests advanced tumor stage / presence of ascites) are admitted to the hospital, ascites / peritoneal lavage fluid is obtained by direct or saline injection and retention for a certain period of time through abdominal puncture under ultrasound / CT guidance. NGS-based ctDNA detection is performed in the peritoneal lavage fluid / ascites specimens for peritoneal metastasis screening. Tumor samples for NGS sequencing can be derived from patient biopsy specimens.

[0092] (2) Intraoperative ascites / peritoneal lavage fluid to determine whether there is occult peritoneal metastasis: At the beginning of the operation and before closing the abdomen at the end of the operation, the peritoneal lavage fluid is obtained by aspirating the existing ascites or peritoneal lavage. Through the detection of ctDNA in the peritoneal lavage fluid, occult peritoneal metastasis can be determined even when it is difficult to detect with the naked eye, thereby guiding the formulation of individualized treatment strategies, including chemotherapy, targeted therapy and other intervention measures, which significantly improves the comprehensive treatment effect and quality of life of patients.

[0093] (3) Determine whether peritoneal metastasis exists during postoperative follow-up: During postoperative follow-up, peritoneal metastasis can be monitored by regular detection of ctDNA in peritoneal lavage fluid. The high sensitivity and specificity of ctDNA detection methods based on surgical specimen / biopsy specimen NGS can timely detect the occurrence or progression of occult peritoneal metastasis, so as to intervene or adjust the treatment plan in time.

[0094] By using NGS-based methods to detect ctDNA in the patient's peritoneal lavage fluid, dynamic monitoring and accurate diagnosis of peritoneal metastasis can be achieved. Combined with the detection of circulating tumor cells (CTC) in the peritoneal lavage fluid, the risk prediction of metachronous peritoneal metastasis after radical resection of gastric cancer can be more accurate. It provides important auxiliary information for clinicians, guides the formulation of individualized treatment plans, and improves the prognosis and survival rate of patients. BRIEF DESCRIPTION OF THE DRAWINGS

[0095] Figure 1 This is the prediction chart of peritoneal lavage fluid ctDNA for the survival prognosis of the experimental cohort.

[0096] Figure 2 This is the prediction chart of peritoneal lavage fluid ctDNA for survival prognosis of the validation cohort.

[0097] Figure 3 This is the ROC curve of peritoneal lavage fluid ctDNA for the prediction of peritoneal metastasis.

[0098] Figure 4 This is the ROC curve of peritoneal lavage fluid CTC for the prediction of peritoneal metastasis.

[0099] Figure 5 This is the ROC curve of peritoneal lavage fluid CTC and ctDNA for the prediction of peritoneal metastasis. DETAILED DESCRIPTION

[0100] The specific implementation methods provided by the present invention are described in detail below in conjunction with examples.

[0101] Embodiment 1:

[0102] 1. Collection and pretreatment of peritoneal lavage fluid:

[0103] During the operation, 250 ml of 0.9% normal saline was used to irrigate the abdominal cavity and then aspirate; or during the non-surgical period, the patient's abdominal cavity was punctured under the guidance of imaging techniques such as ultrasound / CT, and 250 ml of 0.9% normal saline was infused and retained for an appropriate time before drainage.

[0104] Collection and storage of peritoneal lavage fluid: Use a 50 mL centrifuge tube with EDTA anticoagulant, inject the peritoneal lavage fluid directly into the centrifuge tube, and tighten the tube cap. If it cannot be processed immediately, store it at 4-10℃ and process it within 24 hours. If the sample needs to be transported, it must be kept at 4-10℃ during transportation.

[0105] Pretreatment of peritoneal lavage fluid: Place the peritoneal lavage fluid in a 50 ml centrifuge tube, centrifuge at 3000 rpm for 30 min, and separate the supernatant.

[0106] 2. Extraction of free DNA from peritoneal lavage fluid:

[0107] (1) The peritoneal lavage fluid of each patient was extracted into three tubes, each with 4 ml, and the sample number was marked;

[0108] (2) Add peritoneal lavage fluid, proteinase K, and 20% SDS to each centrifuge tube in sequence. The amounts added are shown in Table 1.

[0109] Table 1

[0110]

[0111] (3) Invert to mix, and incubate at 60°C for 20 min.

[0112] (4) Add free DNA binding solution and oscillate to mix the magnetic beads. The amount to be added is shown in Table 2.

[0113] Table 2

[0114]

[0115] (5) Vortex for 10 seconds to mix thoroughly. Vortex for 5-10 minutes to allow free DNA to fully bind to the magnetic beads.

[0116] (6) Briefly centrifuge to collect the liquid on the tube cap and tube wall, and place the centrifuge tube on a magnetic rack until the liquid becomes clear (3-5 minutes);

[0117] (7) Remove the supernatant;

[0118] (8) Add 500 μL of free DNA washing solution and vortex to mix;

[0119] (9) Briefly centrifuge to collect the liquid on the tube cap and tube wall, and place the centrifuge tube on a magnetic rack until the liquid becomes clear (1-2 minutes);

[0120] (10) removing the supernatant;

[0121] (11) Add 500 μL of 80% ethanol and shake to mix;

[0122] (12) Briefly centrifuge to collect the liquid on the tube cap and tube wall, and place the centrifuge tube on a magnetic rack until the liquid becomes clear (1-2 minutes);

[0123] (13) removing the supernatant;

[0124] (14) Repeat steps (11) to (13) and rinse the magnetic beads once more with 80% ethanol;

[0125] (15) Use a 10 μL pipette to suck out the liquid remaining at the bottom of the centrifuge tube and dry it at room temperature for 3-5 minutes. Note: The magnetic beads should not be over-dried.

[0126] (16) Remove the sample from the magnetic rack, add 30-50 μL of free DNA elution buffer, and shake for 3-5 minutes to completely release the free DNA;

[0127] (17) Briefly centrifuge the collected liquid on the tube cap and place the tube on a magnetic rack until the liquid becomes clear. Carefully pipette the supernatant into a new centrifuge tube and store at -20°C ± 5°C.

[0128] 3. Library construction, capture and sequencing

[0129] 3.1 Library construction

[0130] The KAPA Hyper Preparation Kit (Kapa Biosystems, USA) was used to construct the library of ctDNA samples from peritoneal lavage fluid. The concentration of the constructed library was measured using the Qubit dsDNA HS (High Sensitivity) Assay Kit (Themo Fisher, USA), and the size of the library fragment was evaluated using the Agilent Bioanalyzer 2100 (Agilent, USA).

[0131] The amount of ctDNA input into peritoneal lavage fluid was 200 ng, which was used as a reference for the input into late-stage tissues.

[0132] The library was constructed based on the 60ng input of plasma, which failed to pass the quality control. Further testing of the nucleic acid 2100 quality control fragment revealed that the ctDNA in the peritoneal lavage fluid accounted for a small proportion, and most of it was gDNA. According to the 60ng input, the actual input of ctDNA should be less than 10ng. Therefore, the present invention refers to the input of nucleic acid for library construction of late-stage tissues for library construction.

[0133] 3.2 Target region capture and sequencing

[0134] The xGen Hybridization and Wash Kit (IDT, USA) was used to capture the target region. The panel used for capture covers approximately 2.2MB of the human genome, including 769 genes. After hybridization, washing, and capture, sequencing was performed using DNBSEQ-T7.

[0135] 769 genes, as follows:

[0136] ABCA13、ABCA8、ABCB1、ABCC2、ABCC9、ABL1、ACADSB、ACOT13、ACRC、ADCY8、ADGRG6、AGAP1、AK7、AKT1、AKT2、AKT3、ALDH5A1、ALG9、ALK、ALOX12B、ALS2CR11、AMBRA1、AMER1、ANAPC7、ANKRD28、ANKRD46、ANO1、APAF1、APC、APOL2、APOPT1、AQR、AR、ARAF、ARHGAP26、ARHGAP4、ARHGAP6、ARHGEF12、ARHGEF3、ARID1A、ARID1B、ARID2、ARID4A、ARID5B、ARL13B、ARL4A、ARL6IP6、ARMC5、ASB11、ASH1L、ASPH、ASXL1、ASXL2、ATG3、ATG4C、ATIC、ATM、ATP6V0A1、ATP6V0A2、ATP6V0A4、ATP6V0E1、ATP8A1、ATR、ATRX、AURKA、AURKB、AXIN1、AXIN2、AXL;

[0137] B2M、BAP1、BARD1、BCAS1、BCL2、BCL2L1、BCL2L11、BCL6、BCOR、BCR、BIRC3、BIVM-ERCC5、BLM、BMPR1A、BRAF、BRCA1、BRCA2、BRD4、BRIP1、BRMS1L、BRS3、BTF3、BTG1、BTK;

[0138] C22orf23、C5orf15、C5orf42、C7orf66、C8orf34、CAB39、CACNA1E、CACNA2D1、CALD1、CALM2、CALR、CARD11、CASP8、CAST、CBFB、CBL、CBR3、CBR4、CCDC157、CCDC18、CCND1、CCND2、CCND3、CCNE1、CD274、CD40、CD74、CD79A、CD79B、CDA、CDC73、CDCA8、CDH1、CDK12、CDK4、CDK6、CDK8、CDKL3、CDKN1A、CDKN1B、CDKN2A、CDKN2B、CDKN2C、CDO1、CEBPA、CEP120、CEP290、CFAP221、CFAP53、CHD1、CHD2、CHEK1、CHEK2、CHRM3、CHURC1-FNTB、CIC、CLASP2、CLEC16A、CLEC9A、CNKSR3、CNOT8、COL15A1、COX18、CPS1、CREBBP、CRKL、CRLF2、CSF1R、CSF3R、CTAGE5、CTCF、CTLA4、CTNNB1、CTSC、CUL3、CXCL8、CXCR4、CYBA、CYFIP1、CYLD、CYP19A1、CYP2B6、CYP2C19、CYP2C8、CYP2D6;

[0139] DARS2、DAXX、DCHS2、DDR1、DDR2、DDX19B、DDX58、DEPDC5、DHFR、DIAPH1、DIAPH2、DICER1、DIS3、DLC1、DMXL1、DNAJB1、DNAJC11、DNMT1、DNMT3A、DNMT3B、DOCK11、DOT1L、DPP6、DPYD、DSCAM;

[0140] E2F3、EBP、EED、EGFR、EIF1AX、EIF4E、EIF4G3、ELFN1、ELMOD2、EML4、ENOSF1、ENSA、EP300、EPCAM、EPG5、EPHA3、EPHA5、EPHA7、EPHB1、EPYC、ERBB2、ERBB3、ERBB4、ERCC1、ERCC2、ERCC3、ERCC4、ERG、ERI1、ERRFI1、ESR1、ETV1、ETV4、ETV5、ETV6、EWSR1、EXOSC8、EZH2、EZR;

[0141] FAM149A, FAM153B, FAM161A, FAM175A, FAM184B, FAM20A, FAM46C, FANCA, FANCC, FAN CD2, FANCF, FANCG, FAS, FAT1, FBXO11, FBXW7, FGF10, FGF16, FGF19, FGF3, FGF4, and FGF6 FGFR1, FGFR2, FGFR3, FGFR4, FH, FLCN, FLI1, FLOT1, FLT3, FLT4, FMNL2, FMO1 FMR1, FNBP4, FOLH1B, FOXA1, FOXL2, FOXO1, FOXP1, FPGT-TNNI3K, FUBP1, FUS, FXR1

[0142] GABRP、GALNT12、GALNT14、GANC、GATA1、GATA2、GATA3、GIPC1、GLI1、GMEB1、GNA11、GNA13、GNAQ、GNA S、GPAT3、GPC4、GPM6A、GRB10、GREM1、GRIK2、GRIN2A、GSK3B、GSKIP、GSTA1、GSTM1、GSTP1、GUCY1A2;

[0143] H3F3A, HAUS2, HAUS6, HCAR2, HDGFRP3, HERC6, HEY1, HGF, HIST1H1C, HIST1H3B, HLA-A, HLA-B, HLA-C HMCN1, HNF1A, HNF4A, HOMER1, HRAS, HSD17B11, HSD3B1, HSPA1B, HSPA4, HSPA5, HSPH1, HTT, HYOU1

[0144] IARS、ICOSLG、ID2、ID3、IDH1、IDH2、IGF1、IGF1R、IGF2、IKBKE、IKZF1、IL10、I L13RA1, IL7R, IMPG1, INHBA, INPP4A, INPP4B, IRF4, IRF6, IRF8, IRS2, ITGAL.

[0145] JAK1, JAK2, JAK3, JUN

[0146] KDM5A, KDM5C, KDM6A, KDR, KEAP1, KIAA1210, KIAA1841, KIT, KLF4, KMT2A, KMT2C, KMT2D, KPNA4, KPNB1, KRAS, KTN1 <h2 style=";text-align:left;direction:ltr">

[0147] <h2 style=";text-align:left;direction:ltr"> LAMA3、LATS1、LATS2、LEPR、LMO1、LNPEP、LONRF3、LRP2、LRRC16A、LRRC34、LYN;<h2 style=";text-align:left;direction:ltr"> <h2 style=";text-align:left;direction:ltr">

[0148] <h2 style=";text-align:left;direction:ltr"> MALRD1、MALT1、MAP2K1、MAP2K2、MAP2K4、MAP3K1、MAP3K13、MAP3K4、MAP4K3、MAP4K5、MAPK1、MAPKAP1、MAPKBP 1、MARK1、MARK3、MAX、MCL1、MDC1、MDM2、MDM4、MED12、MED12L、MED14、MED19、MEF2BNB-MEF2B、MEIS1、MEN1、ME T、METTL9、MITF、MLH1、MLH3、MMP16、MMP3、MPL、MRE11A、MRPL19、MS4A13、MSANTD3-TMEFF1、MSH2、MSH3、MSH6、 MTF1、MTF2、MTHFR、MTOR、MTR、MTRR、MUTYH、MYADM、MYB、MYC、MYCL、MYCN、MYD88、MYO10、MYOD1、MYOM1、MZT2A;<h2 style=";text-align:left;direction:ltr"> <h2 style=";text-align:left;direction:ltr">

[0149] <h2 style=";text-align:left;direction:ltr"> NAB1、NAMPT、NAPG、NAV1、NBAS、NBEAL1、NBN、NCOA6、NCOR1、NEDD4L、NEO1、NF1、NF2、NFE2L2、NFKBIA、NFXL1、NKAP、NKX2-1、NLRP7、 NOTCH1、NOTCH2、NOTCH3、NOTCH4、NPM1、NR1I3、NRAS、NRG1、NRG4、NSD1、NT5C2、NTHL1、NTRK1、NTRK2、NTRK3、NUDT13、NUP85、NUP93;<h2 style=";text-align:left;direction:ltr"> <h2 style=";text-align:left;direction:ltr">

[0150] <h2 style=";text-align:left;direction:ltr"> OSBP, OTOGL, OTOS;<h2 style=";text-align:left;direction:ltr"> <h2 style=";text-align:left;direction:ltr">

[0151] P2RY8, PAK1, PAK7, PALB2, PAPOLG, PAQR8, PARD6B, PARK2, PARP1, P ARP2, PARP3, PARP8, PAX3, PAX5, PBRM1, PDCD1, PDCD1LG2, PDE4D, PD GFRA, PDGFRB, PDPK1, PDS5A, PFKP, PGBD1, PGR, PGRMC2, PHF20, PIG F、PIC3C2G、PIC3C3、PIC3CA、PIC3CB、PIC3CD、PIC3CG、PIC3R1、PIC3 R2, PIK3R3, PIM1, PKHD1, PLCG2, PLEKHA1, PLEKHH2, PLXNC1, PMS1. PMS2, PNO1, POLA1, POLD1, POLE, POSTN, PPARG, PPP1R21, PPP2R1A, P RDM1, PRELID3B, PREX2, PRKAR1A, PRKCI, PRKDC, PRPF39, PRPF4, PTCH1, PTEN, PTK2, PTPN11, PTPN4, PTPRD, PTPRJ, PTPRS, PTPRT, FULL.

[0152] RAB2B, RABGAP1L, RAC1, RAD21, RAD50, RAD51, RAD51B, RAD51C, RAD51D, RAD52 AD54L, RAF1, RALGAPB, RAP2B, RARA, RASA1, RB1, RBM10, RBM27, RECQL4, REL, RET. RFC1, RFWD2, RHOA, RHOT1, RIC1, RICTOR, RIPK2, RIT1, RNF112, RNF19A, RNF43, RO BO1, ROS1, RPF2, RPRD1A, RPS6KB1, RPTOR, RRM1, RRP1B, RUNX1, RWDD1, RYBP, RYR2

[0153] <h2 style=";text-align:left;direction:ltr">SASH1、SCOC、SDHA、SDHAF2、SDHB、SDHC、SDHD、SEL1L3、SEMA3C、SEMA3E、SERTAD4、SETD2、SF3B1、SFXN4、SH2D1A、SHQ1、SH ROOM3、SIMC1、SIPA1L2、SKA3、SLC13A1、SLC22A2、SLC25A13、SLC30A5、SLC31A1、SLC35B1、SLC7A8、SLC9C2、SLCO1B1、SLC O1B3, SLIT1, SLX4, SMAD2, SMAD3, SMAD4, SMARCA4, SMARCB1, SMO, SNX6, SOCS1, SOD2, SOX17, SOX2, SOX9, SPEN, SPOP, SRC, SRSF3, SRY, STAB2, STAG2, STARD4, STAT3, STK11, STMN1, STRBP, STT3A, STYX, SUCLG1, SUFU, SUGCT, SUZ12, SYK, SYNE2;<h2 style=";text-align:left;direction:ltr"> <h2 style=";text-align:left;direction:ltr">

[0154] <h2 style=";text-align:left;direction:ltr"> TAF15、TAOK3、TARBP1、TBC1D8B、TBCD、TBX3、TECPR2、TENM3、TERT、TERT-promoter、TET1、TET2 、TFDP1、TFRC、TGFBR1、TGFBR2、TMEM126B、TMEM127、TMEM132D、TMEM67、TMPRSS15、TMPRSS2、TMT C4、TNFAIP3、TNFRSF14、TNFSF13B、TNIK、TNKS、TNRC18、TOP1、TOP2B、TP53、TP63、TPH1、TPM1、T RA2A、TRAF7、TRIM24、TRIM25、TSC1、TSC2、TSHR、TSN、TTC1、TTC6、TTN、TUBD1、TXNDC16、TXNRD1;<h2 style=";text-align:left;direction:ltr"> <h2 style=";text-align:left;direction:ltr">

[0155] <h2 style=";text-align:left;direction:ltr"> U2AF1、UBAP2L、UBE2E3、UBE4A、UBN2、UBXN7、UGT1A1、ULK2、ULK4、UMPS、UPF2、USP11、USP34、USP9Y、UTS2、UTY;<h2 style=";text-align:left;direction:ltr"> <h2 style=";text-align:left;direction:ltr">

[0156] <h2 style=";text-align:left;direction:ltr"> VEGFA, VHL, VSIG10;<h2 style=";text-align:left;direction:ltr"> <h2 style=";text-align:left;direction:ltr">

[0157] <h2 style=";text-align:left;direction:ltr"> WDR5、WHSC1、WHSC1L1、WT1;<h2 style=";text-align:left;direction:ltr"> <h2 style=";text-align:left;direction:ltr">

[0158] <h2 style=";text-align:left;direction:ltr"> XIAP、XPC、XPO1、XRCC1、XRCC2;

[0159] YAP1, YLPM1, YWHAE;

[0160] ZBBX, ZBTB40, ZDHHC17, ZDHHC20, ZMYM2, ZMYM4, ZNF195, ZNF2, ZNF280D, ZNF283, ZNF367, ZNF711, ZNF805, ZNF91, ZZZ3.

[0161] 4. ctDNA quantitative detection:

[0162] 4.1 Variant Identification

[0163] First, Trimmomatic (v0.36) software was used to remove adapters and low-quality sequencing products (reads). Clean reads were aligned to the human hg19 reference genome using BWA aligner (v0.7.17) software. Next, Picard (v2.23.0) software was used to classify and remove duplications. SNVs and InDels were identified and detected using VarDict (v1.5.1) software, and complex mutations were identified using FreeBayes (v1.2.0). The filtering of QC data such as mutation quality and chain preference are listed in the original mutation list. In addition, mutations in low-complexity repeats and segmental duplications that matched low-mapping regions defined in ENCOD, as well as mutations in the sequencing-specific errors (SSEs) list developed and validated in-house, were removed.

[0164] 4.2 Screening for gene mutations in tumor tissues

[0165] First, mutations from germline or hematopoietic sources are filtered out. Mutations that meet any of the following criteria will be filtered out: (1) the variant frequency (VAF) of the mutation from peripheral blood is not less than 5%, or (2) the VAF value of the mutation from peripheral blood is less than 5%, but the VAF value is not more than 5 times the VAF value of the same point in the matched tissue sample, or (3) the mutation can be found in the public gnomAD population database, with a smaller allele frequency (MAF) of not less than 2%. The remaining gene mutations will be further filtered under quality conditions. When screening tumor tissue mutations, each mutation is supported by at least 5 reads, the detection limit of SNV is 4%, and the detection limit of InDel is 5%, which are used as the conditions for screening tumor tissue mutations. These threshold standards have been determined during the performance analysis and validation of the NGS Panel, and have been measured in terms of sensitivity, specificity, repeatability and reproducibility.

[0166] 4.3 Screening for gene mutations in peritoneal lavage fluid

[0167] A tissue-a priori MRD detection method was used. First, ctDNA mutations not detected in the initial tissue sequencing were excluded, and only somatic variants detected in tumor tissue were monitored in peritoneal lavage fluid. Each monitored ctDNA mutation was statistically tested using a background noise baseline model to eliminate the impact of technical noise. The remaining variant signals represented the noise signals of each variant direction (Subtype) of each chromosome coordinate (coordinate) by removing true tumor-derived somatic variants and clonal hematopoietic (CH) mutations through paired tissue and PBL sequencing, and the Weibull distribution or inverse Gamma distribution was fitted to form a noise model with single-base resolution. The zero inflation effect in the noise baseline model was approximated by Monte Carlo sampling simulation to obtain the zero percentage weighted probability of a given variant. Specifically, 10,000 random samples were taken from the modeled distribution, including n_nonzero non-zero VAFs and n_zero zero VAFs, where n_nonzero and n_zero were calculated as the product of the non-zero and zero percentages, respectively, with a predetermined total of 1,0000. Each VAF in this list was further fed as a noise probability into a binomial test parameterized by the observed number of reads supporting the alt allele and the total number of reads supporting it. The expected value was the average of all p values, and the threshold was set to 0.05 to define ctDNA as positive for that variant in this study (p < 0.05).

[0168] In defining the MRD status of samples, in order to avoid the loss of specificity caused by multiple comparisons, we corrected the significance of differences at the sample level by using the multivariate joint confidence probability method, applying the formula:

[0169]

[0170] n is the number of tracked mutations, k is the number of mutations with p<=0.05, and C represents the number of combinations. is the p_value of the positive site;

[0171] Finally, the significance level of the sample compared with the noise baseline was obtained. In this study, the significance threshold P of the sample level was set at 0.01. When the P value of the peritoneal lavage fluid test was lower than the threshold, the peritoneal lavage fluid sample was determined to be MRD positive.

[0172] ROC curve of peritoneal lavage fluid ctDNA for the prediction of peritoneal metastasis Figure 3 As shown, ctDNA AUC=0.93, which shows that the prediction method provided by the present invention has much better specificity and sensitivity than carcinoembryonic antigen (CEA) AUC=0.64.

[0173] Embodiment 2:

[0174] Overall experimental process: Aims to observe and evaluate the predictive effect of ctDNA levels in peritoneal lavage fluid of patients with non-advanced gastric cancer on peritoneal metastasis. For patients with gastric cancer confirmed as stage I-III by postoperative pathology, who have not received chemotherapy, radiotherapy, targeted therapy, or immunotherapy before. The subjects completed the collection of peritoneal lavage fluid during surgery, and were screened and qualified for the study within 4 weeks after the completion of the surgery. The peritoneal lavage fluid ctDNA test was performed, and peritoneal metastasis follow-up was started. The frequency of imaging evaluation was once every 3 months. For those with follow-up time of more than 1 year, the subsequent imaging evaluation frequency was once every 6 months. All enrolled patients received standard treatment after surgery.

[0175] The above method was used to detect and follow up 34 patients (experimental cohort) for 15 months, and they were divided into ctDNA positive group and negative group according to the ctDNA test results. During the follow-up period, a total of 6 patients developed peritoneal metastasis, and all of them occurred in the ctDNA positive group. There was a statistically significant difference in the survival curves of the two groups (p=0.05), with an HR of 5.292 (95%CI 0.9489 to 29.51).

[0176] According to the results of the experimental cohort, the same method was used to divide 23 patients (validation cohort) into ctDNA positive group (predicted peritoneal metastasis group) and negative group (predicted no peritoneal metastasis group) according to the ctDNA test results. During the follow-up period, a total of 4 cases developed peritoneal metastasis, and all occurred in the predicted peritoneal metastasis group. There was a statistically significant difference in the survival curves of the two groups (p < 0.05).

[0177] Embodiment 3:

[0178] The method of the present invention was used to detect the peritoneal lavage fluid of 20 patients with confirmed peritoneal metastasis and 20 patients undergoing surgical treatment for benign diseases. The selected targets were folate receptor and epithelial cell adhesion molecule (EpCAM). The true positive rate was 80% and the false positive rate was 15%. The results are shown in Table 3.

[0179] Table 3. CTC detection results

[0180]

[0181] Embodiment 4:

[0182] Overall experimental process: Aims to observe and evaluate the predictive effect of CTC detection in peritoneal lavage fluid of patients with non-advanced gastric cancer on peritoneal metastasis. For patients with gastric cancer confirmed as stage I-III by postoperative pathology, who have not received chemotherapy, radiotherapy, targeted therapy, or immunotherapy before. The subjects completed the collection of peritoneal lavage fluid during surgery, and were screened and qualified for the study within 4 weeks after the completion of the surgery. CTC detection of peritoneal lavage fluid was performed, and peritoneal metastasis follow-up was started. The frequency of imaging evaluation was once every 3 months. For those with follow-up time of more than 1 year, the frequency of subsequent imaging evaluation was once every 6 months. All enrolled patients received standard treatment after surgery.

[0183] The above method was used to detect 36 patients and follow up for 15 months. According to the CTC test results, they were divided into CTC positive group and negative group. During the follow-up period, a total of 7 patients developed peritoneal metastasis. There was a statistically significant difference in the survival curves between the two groups (p < 0.05), with a HR of 6.746 (95% CI 1.529 to 29.76).

[0184] Embodiment 5:

[0185] The follow-up data of patients who underwent CTC and ctDNA detection in Example 2 and Example 5 were integrated. Patients with both ctDNA and CTC positive were divided into the High Risk group, and the remaining patients were Low Risk. A total of 6 cases developed peritoneal metastasis during the follow-up period. There was a statistically significant difference in the survival curves of the two groups (p=0.003), with a HR of 14.69 (95%CI2.482 to 86.90).

[0186] The combined detection of CTC and ctDNA can more accurately divide the high-risk population for peritoneal metastasis (10 / 34) and has a higher predictive effect for peritoneal metastasis.

[0187] Table 4

[0188]

[0189] The preferred embodiments of the present invention have been specifically described above, but the present invention is not limited to the embodiments. Those skilled in the art may make various equivalent modifications or substitutions without violating the spirit of the present invention. These equivalent modifications or substitutions are all included in the scope defined by the claims of this application.

Claims

1. Use of a reagent for detecting circulating tumor cells in peritoneal lavage fluid and a reagent combination for detecting circulating tumor DNA content in peritoneal lavage fluid in the preparation of a kit for predicting the risk of metachronous peritoneal metastasis after radical resection of gastric cancer, characterized in that: The reagents include: a reagent for detecting circulating tumor cells in peritoneal lavage fluid based on cytological detection, and a reagent for detecting the content of circulating tumor DNA in peritoneal lavage fluid based on the tumor tissue second-generation sequencing method; wherein the method for detecting the content of circulating tumor DNA in peritoneal lavage fluid based on the tumor tissue second-generation sequencing method comprises the following steps: A) Centrifugation of peritoneal lavage fluid to separate the supernatant; B) Extraction of circulating tumor DNA from peritoneal lavage fluid: add 20 mg / mL proteinase K and 20% SDS to the supernatant, mix well, and incubate at 50-70°C for 10-30 min; add DNA binding solution and magnetic beads to allow circulating tumor DNA to fully bind to the magnetic beads; centrifuge, let stand, and remove the supernatant; add DNA washing solution, mix well, centrifuge, let stand, and remove the supernatant; rinse the magnetic beads with 70-90% ethanol; add DNA elution solution to completely release free DNA; centrifuge, let stand, collect the supernatant, and store at -20°C±5°C; C) Library construction, capture and sequencing (c1) Library construction The KAPAHyper Preparation Kit was used to construct the library of ctDNA samples from peritoneal lavage fluid. The concentration of the constructed library was measured using the Qubit dsDNA HS Assay Kit, and the library fragment size was evaluated using the Agilent Bioanalyzer 2100. The input amount of ctDNA from peritoneal lavage fluid was greater than or equal to 200 ng. (c2) Target region capture and sequencing The xGen Hybridization and Wash Kit was used to capture the target region; the panel used for capture covered approximately 2.2MB of the human genome, including 769 genes; after hybridization, washing, and capture, DNBSEQ-T7 was used for sequencing; D)ctDNA detection: Software was used to identify variants and screen gene mutations in tumor tissues. The tissue-a priori MRD detection method was used to screen gene mutations in peritoneal lavage fluid. When defining the MRD status of the samples, the significance of differences at the sample level was corrected using the multi-variant joint confidence probability method.

2. Use of the reagent for detecting circulating tumor cells in peritoneal lavage fluid and the reagent combination for detecting circulating tumor DNA content in peritoneal lavage fluid according to claim 1 in the preparation of a kit for predicting the risk of metachronous peritoneal metastasis after radical resection of gastric cancer, characterized in that: Steps A) and B) are specifically: A) Centrifuge the peritoneal lavage fluid at 2000-4000 rpm for 10-40 min to separate the supernatant; B) Extraction of circulating tumor DNA from peritoneal lavage fluid: Step A) adding 20 mg / mL proteinase K and 20% SDS to the supernatant obtained; Invert to mix, incubate at 50-70℃ for 10-30min; Add DNA binding solution and magnetic beads; shake to allow circulating tumor DNA to fully bind to the magnetic beads; centrifuge, let stand, and remove the supernatant; Add DNA washing solution, shake and mix; centrifuge, let stand, and remove the supernatant; Rinse the magnetic beads with 70-90% ethanol; remove the sample from the magnetic rack, add DNA elution buffer, and shake to completely release the free DNA; centrifuge, let stand, aspirate the supernatant into a new container, and store at -20℃±5℃.

3. Use of the reagent for detecting circulating tumor cells in peritoneal lavage fluid and the reagent combination for detecting circulating tumor DNA content in peritoneal lavage fluid according to claim 1 in the preparation of a kit for predicting the risk of metachronous peritoneal metastasis after radical resection of gastric cancer, characterized in that: Step D) is specifically: (d1) Variant identification First, Trimmomatic v0.36 software was used to remove adapters and low-quality sequencing product reads; BWAaligner v0.7.17 software was used to align clean reads to the human hg19 reference genome; Picard v2.23.0 software was then used to classify and remove duplications; SNVs and InDels were identified and detected using VarDict v1.5.1 software, and complex mutations were identified using FreeBayes v1.2.0; the filtering of QC data such as mutation quality and chain preference were listed in the original mutation list; in addition, mutations in low-complexity repeats and fragment duplications that matched low-mapping regions defined in ENCOD, as well as mutations in the sequencing-specific error SSEs list developed and validated internally, were removed; (d2) Screening for gene mutations in tumor tissues First, mutations from germline or hematopoietic sources are filtered out. Mutations that meet any of the following criteria will be filtered out: (1) the mutation frequency VAF from peripheral blood is not less than 5%, or (2) the mutation VAF value from peripheral blood is less than 5%, but the VAF value is not more than 5 times the VAF value of the same point in the matched tissue sample, or (3) the mutation can be found in the public gnomAD population database, which has a smaller allele frequency MAF and is not less than 2%; the remaining gene mutations will be further filtered under quality conditions; when screening tumor tissue mutations, each mutation is supported by at least 5 reads, the detection limit of SNV is 4%, and the detection limit of InDel is 5%, which are used as the conditions for screening tumor tissue mutations; (d3) Screening of gene mutations in peritoneal lavage fluid A tissue-a priori MRD detection method was used. First, ctDNA mutations that were not detected in the initial tissue sequencing were excluded, and only somatic mutations detected in tumor tissue were monitored in the peritoneal lavage fluid. A statistical test was performed on each monitored ctDNA mutation using a background noise baseline model to eliminate the impact of technical noise. True tumor-derived somatic mutations and clonal hematopoietic CH mutations were removed by sequencing paired tissues and PBLs. The remaining mutation signals represented the noise signals of each mutation direction Subtype of each chromosome coordinate coordinate, and were fitted with a Weibull distribution or inverse Gamma distribution to form a single-base resolution The zero inflation effect in the noise baseline model is approximated by Monte Carlo sampling simulation to obtain the zero percentage weighted probability of a given variant; specifically, 10,000 random samples are taken from the modeled distribution, including n_nonzero non-zero VAFs, and n_zero zero VAFs, where n_nonzero and n_zero are calculated as the product of the non-zero and zero percentages respectively, with a predetermined total of 1,0000; each VAF in this list is further fed as a noise probability into a binomial test parameterized by the observed alt allele support reads and the total support reads; The expected value is the average of all p-values.

4. Use of a reagent for detecting circulating tumor cells in peritoneal lavage fluid and a reagent combination for detecting circulating tumor DNA content in peritoneal lavage fluid according to claim 1, 2 or 3 in the preparation of a kit for predicting the risk of metachronous peritoneal metastasis after radical resection of gastric cancer, characterized in that: When defining the MRD status of samples, the difference significance at the sample level was corrected by the multivariate joint confidence probability method, using the formula: n is the number of tracked mutations, k is the number of mutations with p<=0.05, C represents the number of combinations, and p k is the p_value of the positive site.

5. Use of the reagent for detecting circulating tumor cells in peritoneal lavage fluid and the reagent combination for detecting circulating tumor DNA content in peritoneal lavage fluid according to claim 3 in the preparation of a kit for predicting the risk of metachronous peritoneal metastasis after radical resection of gastric cancer, characterized in that: The threshold value P is 0.

01. When the P value of the circulating tumor DNA detection in the peritoneal lavage fluid is lower than the threshold value, the circulating tumor DNA sample in the peritoneal lavage fluid is judged to be MRD positive.

6. Use of a reagent for detecting circulating tumor cells in peritoneal lavage fluid and a reagent combination for detecting circulating tumor DNA content in peritoneal lavage fluid according to any one of claims 1 to 5 in the preparation of a kit for predicting the risk of metachronous peritoneal metastasis after radical resection of gastric cancer, characterized in that: The peritoneal lavage fluid is the peritoneal lavage fluid obtained by lavaging the peritoneal cavity with 0.9% saline during surgery or during non-surgical periods. The non-surgical period refers to the perioperative period, systemic treatment cycle, and life survival period.

7. A detection system for predicting the risk of metachronous peritoneal metastasis after radical resection of gastric cancer, characterized in that: The detection system implements the use of a combination of a reagent for detecting circulating tumor cells in peritoneal lavage fluid and a reagent for detecting circulating tumor DNA content in peritoneal lavage fluid as described in any one of claims 1-6 in the preparation of a kit for predicting the risk of metachronous peritoneal metastasis after radical resection of gastric cancer.

Images