Use of reagent to detect fecal microbiota in preparation of detection kit for assistance in predicting risk of gastric cancer

Abstract

The present invention relates to a use of a reagent to detect fecal microbiota in preparation of a detection kit for assistance in predicting risk of gastric cancer. Species and abundance of indicative microbes can be detected and compared by a primer pair having specific sequences of the reagent, which can assist in classifying a subject before or after Helicobacter pylori( H. pylori) eradication to a low risk group or a high risk group of gastric cancer, thereby being applied in preparation of a detection kit for assistance in predicting risk of gastric cancer.

Description

Technical Field

[0001] This invention relates to the use of a microbial community for assisting in the prediction of cancer risk, and more particularly to the use of a reagent for detecting fecal microbiota in the preparation of a test kit for assisting in the prediction of gastric cancer risk. Prior Technology

[0002] Helicobacter pylori (H. pylori) infection causes chronic active gastritis, which can lead to precancerous lesions such as atrophic gastritis (AG) and intestinal metaplasia (IM), resulting in poor regeneration, or dysplasia, and eventually gastric cancer. Persistent inflammation can destroy acid-secreting glands and cause loss of gastric acid secretion. Insufficient gastric acid can lead to colonization of bacteria other than H. pylori in the gastric mucosa. Research reports indicate that H. pylori infection causes gastritis, and the action of other gastric microorganisms [such as Streptococcus anginosus, Acinetobacter lwoffii, and Penicillium rakeb] exacerbates the progression of inflammation, ultimately leading to poor gastric regeneration and gastric cancer. This vicious cycle allows atrophic gastritis and intestinal metaplasia to progress to gastric cancer.

[0003] The gut microbiota lives in symbiosis with the host, and its composition and diversity change after Helicobacter pylori colonizes the stomach. However, it remains uncertain whether the gut microbiota changes in precancerous gastric lesions. Previous studies have suggested that insufficient gastric acid in precancerous gastric lesions alters the fecal microbiota. Nearly half the global population is infected with Helicobacter pylori. This infection alters the serum metabolome in patients with peptic ulcers, suggesting that widespread changes in the colonic microbiota may be influenced by Helicobacter pylori-related metabolomic changes, and vice versa, rather than simply by the localized effects of Helicobacter pylori in the stomach. More than 30% of Helicobacter pylori-infected individuals develop advanced atrophic gastritis and gastric mucosal intestinal metaplasia after age 50. However, research on the fecal microbiota composition of such precancerous gastric lesions is limited.

[0004] The diagnosis of gastric cancer mainly relies on gastroscopy, upper gastrointestinal X-ray, and pathological examination. However, gastroscopy and pathological examination are invasive procedures, which can easily cause patients to fear them and even delay treatment, thus missing the opportunity for intervention. Therefore, there is an urgent need to develop a non-invasive examination method to improve patients' willingness to undergo the examination, assess the risk of developing gastric cancer in subjects before and after Helicobacter pylori eradication, and even apply it to large-scale screening. Summary of the Invention

[0005] Therefore, one aspect of the present invention is to provide a reagent for detecting fecal microbiota for use in the preparation of a test kit for assisting in the prediction of the risk of gastric cancer, so as to non-invasively determine the future risk of gastric cancer in subjects before Helicobacter pylori eradication.

[0006] Secondly, another aspect of the present invention is to provide a reagent for detecting fecal microbiota for use in the preparation of a test kit for assisting in the prediction of the risk of gastric cancer, so as to non-invasively determine the future risk of gastric cancer in subjects after Helicobacter pylori eradication.

[0007] According to the above-described form of the present invention, a reagent for detecting fecal microbiota is proposed for use in preparing a test kit for assisting in the prediction of gastric cancer risk. The aforementioned test kit is used to detect a first ex vivo fecal sample from a subject before Helicobacter pylori eradication, provided that the first ex vivo fecal sample has not yet been diagnosed or identified as belonging to a low-risk or high-risk group for gastric cancer. In one embodiment, firstly, a detection step is performed on the first ex vivo fecal sample to obtain a first detection result, wherein the detection step utilizes a first primer (e.g., a forward primer) of the reagent as shown in sequence identification number (SEQ ID NO:) 1 and a second primer (e.g., a reverse primer) of the reagent to perform a polymerase chain reaction, and the aforementioned first detection result contains a plurality of 16S rRNA gene sequences, and the aforementioned 16S rRNA gene sequences correspond to the species and / or abundance of the first microbiota. Then, the species and / or abundance of the first bacterial phase of the first ex vivo fecal sample of the subject are compared with the species and / or abundance of the second bacterial phase of the second test result obtained by polymerase chain reaction of the second ex vivo fecal sample of a normal individual. When the abundance of the indicator bacterial species of the first bacterial phase is increased or decreased compared with that of the second bacterial phase, the subject corresponding to the first ex vivo fecal sample is classified into the gastric cancer risk group.

[0008] In the above embodiments, the aforementioned indicator bacteria may include, but are not limited to, Megamonas funiformis YIT 11815, Lacrimispora amygdalina, Ruminococcus callidus, Holdemanella porci, Gluceribacter canis, Blautia luti, Acidaminococcus intestini, Negativibacillus massiliensis, Amedibacillus dolichus, Fumia xinanensis, Bacteroides faecis MAJ27, Blautia wexlerae DSM 19850, and Phascolarctobacterium fecalithum. The bacteria include *Faecium*, *Lachnoclostridium phocaeense*, *Erysipelatoclostridium ramosum*, *Eisenbergiella tayi*, *Anaerotignum aminivorans*, and *Enterocloster citroniae*.

[0009] In the above embodiments, when the abundance of at least one of the aforementioned Enterococcus simulans, Marseilles-negative bacilli, Entamoeba histolytica, Fomitopsis pinnatifida, Weyclade-Broutella, Collagen foetida, Clostridium perfringens, Clostridium difficile, T. Eisenberger, aminoacidophilic bacteria, and Enterobacter citrate increases, while the abundance of at least one of Megalomonas monomorpha, Lacrimalia amygdalina, Ruminococcus rumeniformis, Holdmannii suis, Glucospermia canis, Bretschneidera sinensis, and Bacteroides fecalith decreases, the subjects corresponding to the aforementioned ex vivo fecal samples are classified as low-risk groups for gastric cancer.

[0010] In the above embodiments, when the abundance of at least one of the aforementioned *Megalomonas monomorpha*, *Lastilus alba*, *Ruminococcus rumeni*, *Holdman's swine*, *Glucobacterium canis*, *Brutella bursa-palatinea*, and *Bacteroides fecalith* increases, while the abundance of at least one of *Enteromorphococcus*, *Masseudomonas muscarinicus*, *Amoebae dolichospasmia*, *Fumigia chinensis*, *Brutella brevicla*, *Koala fecalith*, *Clostridium falciparum*, *Clostridium perfringens*, *Tasmania esculenta*, *Aminophila*, and *Enterobacter citrate* decreases, the subjects corresponding to the aforementioned ex vivo fecal samples are classified into a high-risk group for gastric cancer.

[0011] In the above embodiments, after Helicobacter pylori eradication, the abundance of Megamonas monomorpha in the ex vivo fecal samples of the aforementioned high-risk group for gastric cancer decreased, while the abundance of Weyclade-Broutella, Clostridium perfringens, and Enterobacter citrate increased.

[0012] According to another aspect of the present invention, a reagent for detecting fecal microbiota is proposed for use in the preparation of a test kit for assisting in the prediction of gastric cancer risk. The aforementioned test kit is used to detect a first ex vivo fecal sample taken by a subject after Helicobacter pylori eradication, the first ex vivo fecal sample not yet diagnosed or identified as belonging to a low-risk or high-risk group for gastric cancer. In one embodiment, firstly, a detection step is performed on the first ex vivo fecal sample to obtain a first detection result, wherein the detection step involves polymerase chain reaction using a first primer as shown in SEQ ID NO:1 and a second primer as shown in SEQ ID NO:2. The first detection result contains a plurality of 16S rRNA gene sequences, and the 16S rRNA gene sequences correspond to the species and / or abundance of the first microbiota. Then, the species and / or abundance of the first bacterial phase are compared with the species and / or abundance of the second bacterial phase of the second test result obtained by polymerase chain reaction of the second ex vivo fecal sample of a normal individual. If the first bacterial phase of the first test result contains indicator species and the abundance of the indicator species is increased or decreased compared with the second bacterial phase, the subject corresponding to the first ex vivo fecal sample is classified into the gastric cancer risk group.

[0013] In the above embodiments, the aforementioned indicator bacteria may include, but are not limited to, Bacteroides fragilis, Enterobacter bolteae, Lachnoclostridium pacaense, Dialister massiliensis, Bacteroides nordii WAL 11050 (i.e., JCM 12987), Coprococcus catus, Dielma fastidiosa, Faecalibacillus intestinalis, Roseburia hominis A2_183, Dorea phocaeensis, and Ructibacterium gallinarum.

[0014] In the above embodiments, when the abundance of at least one of the aforementioned enterobacteria, human rotavirus, focaccia dorsalis, and chicken ructoid bacillus increases, while the abundance of at least one of Bacteroides fragilis, Enterobacter baumannii, Clostridium perfringens, Alisteria massa, Bacteroides nori, Fecal cocci, and refractory Dilmurat is decreased, the subjects corresponding to the aforementioned ex vivo fecal samples are classified into the low-risk group for gastric cancer.

[0015] In the above embodiments, when the abundance of at least one of the aforementioned Bacteroides fragilis, Enterobacter baumannii, Clostridium perfringens, Alisteria massa, Bacteroides nori, Fecal cocci, and Dilmuramys refractory is increased, while the abundance of at least one of Enterobacter faecalis, Rochetomyces humanis, Focaccia dorsalis, and Rhus chinensis is decreased, the subjects corresponding to the ex vivo fecal samples are classified into the high-risk group for gastric cancer.

[0016] In the above embodiments, the abundance of at least one of Bacteroides fragilis, Enterobacter baumannii, Clostridium perfringens, Alisterella marsupialis, Bacteroides nori, Coccidia fecalithii, and Dilmuramys refractory in the ex vivo fecal samples of the aforementioned high-risk group for gastric cancer was greater than that of the low-risk group for gastric cancer.

[0017] The reagents for detecting fecal microbiota described above in this invention can be used to prepare a set of test reagents for assisting in the prediction of gastric cancer risk. It uses primers with specific sequences to detect and compare the types and abundance of indicator bacteria in the microbiota to determine whether subjects before and after Helicobacter pylori eradication are classified into a low-risk or high-risk group for gastric cancer. In the future, it can be applied to large-scale screening.

[0018] It is understood that the foregoing general description and the following detailed description are merely illustrative examples intended to provide further explanation of the claimed invention. Simple Explanation of the Diagram

[0019] To make the above and other objects, features, advantages, and embodiments of the present invention more apparent and understandable, the detailed description of the accompanying drawings is as follows: Figure 1 is a partial flowchart showing a method for detecting fecal microbiota according to an embodiment of the present invention. Figure 2A is a Venn diagram showing the fecal microbiota of a high-risk group, a low-risk group, and a control group before sterilization according to some embodiments of the present invention. Figure 2B is a stacked bar chart showing the bacterial species and abundance percentage of the top 10 fecal microbiota in each group of Figure 2A. Figure 2C is a box-and-whisker diagram showing the ratio of Firmicutes to Bacteroidetes in each group of Figure 2B. Figures 3A to 3F show box-and-whisker plots of fecal microbiota related to α-diversity in the high-risk group, low-risk group, and control group of the present invention before sterilization, including Menhinick index (Figure 3A), Shannon entropy analysis (Figure 3B), and Pielou index (Figure 3C), as well as principal coordinate analysis (PCoA) plots of Bray-Curtis distance matrix (Figure 3D), unweighted UniFrac matrix (Figure 3E), and weighted UniFrac matrix (Figure 3F) related to β-diversity. Figures 4A to 4C show abundance analysis plots of fecal microbiota in the high-risk group, low-risk group, and control group of the present invention before sterilization, including the high-risk group, low-risk group, and control group of the present invention before sterilization. The left plot shows the box-and-whisker plot of the average percentage abundance, and the right plot shows the difference in the average percentage abundance (95% confidence interval, with the color of the points indicating the group with higher abundance). Figure 5 shows the abundance trend analysis of fecal microbiota before sterilization in one embodiment of the present invention, analyzed using LEfSe≧ 2.0, from the control group to the low-risk group for gastric cancer and then to the high-risk group for gastric cancer, showing an increase or decrease in the fecal microbiota trend. Figure 6 shows the abundance analysis of fecal microbiota in the high-risk and low-risk groups for gastric cancer after sterilization in another embodiment of the present invention, where the left graph shows the mean percentage abundance box plot, and the right graph shows the difference in mean percentage abundance (95% confidence interval, the color of the dots represents the group with higher abundance). Figure 7 shows the abundance trend analysis of fecal microbiota after sterilization in another embodiment of the present invention, analyzed using LEfSe≧ 2.0, from the control group to the low-risk group for gastric cancer and then to the high-risk group for gastric cancer, showing an increase or decrease in the fecal microbiota trend. Figure 8 shows the ROC curve of fecal microbiota used to assist in predicting the high-risk group for gastric cancer in one embodiment of the present invention. Figure 9 shows the ROC curve of another embodiment of the present invention, which uses fecal microbiota to assist in predicting the high-risk group for gastric cancer. Implementation

[0020] If a cited reference provides a definition or usage of a term that is inconsistent with or contrary to the definition of that term herein, the definition herein shall apply, and the definition in the cited reference shall not apply. Secondly, unless the context otherwise defines, singular terms may include plural forms, and plural terms may include singular forms.

[0021] As mentioned above, the present invention provides a reagent for detecting fecal microbiota for use in preparing a test kit to assist in predicting the risk of gastric cancer, in order to determine the future risk of gastric cancer in subjects who have eradicated Helicobacter pylori.

[0022] The subjects mentioned here refer to patients who are Helicobacter pylori positive and have gastric cancer pathological stages OLGA (operative link for gastritis assessment) / OLGIM (operative link on gastric intestinal metaplasia) of 0 to 4. In other words, due to precancerous lesions, including atrophic gastritis, gastric intestinal metaplasia, and regenerative abnormalities, even after Helicobacter pylori eradication, these lesions are difficult to reverse and continue to progress to gastric cancer. Therefore, it is recommended that these patients undergo regular invasive examinations such as gastroscopy and pathological biopsies after Helicobacter pylori eradication to assess the risk of early gastric cancer. However, this invention proposes a non-invasive examination method using reagents for detecting fecal microbiota to assess the risk of future gastric cancer in subjects before and after Helicobacter pylori eradication.

[0023] The reagents used to detect fecal microbiota mentioned here refer to reagents used to detect the first ex vivo fecal sample of a subject before or after eradication of Helicobacter pylori (hereinafter referred to as eradication), wherein the first ex vivo fecal sample has not been diagnosed or identified as being in the low-risk group or high-risk group for gastric cancer.

[0024] In one embodiment, the reagents for detecting fecal microbiota may include reagents suitable for polymerase chain reaction (PCR) and specific primer pairs, and the detection is performed using the following method. First, a detection step is performed on a first ex vivo fecal sample to obtain a first detection result, wherein the detection step may utilize a first primer (e.g., a forward primer) of the sequence shown in SEQ ID NO: 1 and a second primer (e.g., a reverse primer) of the sequence shown in SEQ ID NO: 2 to perform polymerase chain reaction (PCR), and the aforementioned first detection result contains multiple full-length 16S rRNA gene sequences, wherein the gene fragment size obtained by amplification of the primer pair is approximately 1500 bp. The aforementioned PCR may be, for example, but not limited to, next-generation sequencing (NGS), quantitative PCR (qPCR), etc.

[0025] In the above embodiments, each primer is designed to have a 5' phosphate-modified 5' buffer sequence (GCATC), a 16-base barcode sequence, and a degenerate 16S gene-specific forward or reverse primer sequence. The forward primer (e.g., the 27F universal primer) consists of the following sequence from 5' to 3': 5' Phos / GCATC – 16bp barcode sequence (optional) – AGRGT TYGAT YMTGG CTCAG – 3'. The reverse primer (e.g., the 1492R universal primer) consists of the following sequence from 5' to 3': 5' Phos / GCATC – 16bp barcode sequence (optional) – RGYTA CCTTG TTACG ACTT – 3'. The aforementioned degenerate bases are defined as follows: R represents A, G; Y represents C, T; M represents A, C. The 16bp barcode sequence is optional and is used to indicate the source of the sequence. This is common knowledge in the technical field to which this invention pertains, and therefore will not be described in detail.

[0026] Then, compared with the second bacterial phase obtained by polymerase chain reaction from the second ex vivo fecal sample from a normal individual, the species and abundance of the first bacterial phase of the subject's first ex vivo fecal sample are compared. If the abundance of the indicator bacterial species in the first bacterial phase is increased or decreased compared with that in the second bacterial phase, the subject corresponding to the first ex vivo fecal sample is classified into the gastric cancer risk group.

[0027] The subjects mentioned above can be tested before or after sterilization treatment. Sterilization treatment referred to here refers to Helicobacter pylori eradication treatment, in one specific case, after 10 to 14 days of bismuth quadruple therapy, with a fresh stool sample taken at least 6 months later.

[0028] In the above embodiments, the aforementioned indicator bacteria species for the subjects may include, but are not limited to, Megamonas funiformis YIT 11815, Lacrimispora amygdalina, Ruminococcus callidus, Holdemanella porci, Gluceribacter canis, Blautia luti, Acidaminococcus intestini, Negativibacillus massiliensis, Amedibacillus dolichus, Fumia xinanensis, Bacteroides faecis MAJ27, Blautia wexlerae DSM 19850, and Phascolarctobacterium fecalithum. *Clostridium faecium*, *Clostridium phocaeense*, *Clostridium ramosum*, *Eisenbergiella tayi*, *Anaerotignum aminivorans*, *Enterocloster citroniae*, *Bacteroides fragilis*, *Enterocloster bolteae*, *Clostridium pacaense*, *Dialister massiliensis*, *Bacteroides nordii* (WAL 11050, JCM 12987), *Coprococcus catus*, *Dielma fastidiosa*, *Faecalibacillus intestinalis*, *Roseburia humanis* At least one of the following: hominis A2_183, Dorea phocaeensis, and Ructibacterium gallinarum.

[0029] In some specific cases, prior to Helicobacter pylori eradication, the aforementioned marker bacteria may include, but are not limited to, Megamonas funiformis YIT 11815, Lacrimispora amygdalina, Ruminococcus callidus, Holdemanella porci, Gluceribacter canis, Blautia luti, Acidaminococcus intestini, Negativibacillus massiliensis, Amedibacillus dolichus, Fumia xinanensis, Bacteroides faecis MAJ27, and Blautia wexlerae DSM. 19850, at least one of the following: Phascolarctobacterium faecium, Lachnoclostridium phocaeense, Erysipelatoclostridium ramosum, Eisenbergiella tayi, Anaerotignum aminivorans, and Enterocloster citroniae.

[0030] In other specific examples, after Helicobacter pylori eradication, the aforementioned indicator bacteria may include, but are not limited to, at least one of Bacteroides fragilis, Enterobacter bolteae, Lachnoclostridium pacaense, Dialister massiliensis, Bacteroides nordii WAL 11050 (i.e., JCM 12987), Coprococcus catus, Dielma fastidiosa, Faecalibacillus intestinalis, Roseburia hominis A2_183, Dorea phocaeensis, and Ructibacterium gallinarum.

[0031] In the above embodiments, when the abundance of at least one of the aforementioned enteroaminococci, Marseille-negative bacilli, Entamoeba dolichos, Fumigia chinensis, Weyclade-Broutella, fecal coleopterus, Clostridium perfringens, Clostridium erysipelas, Tarsus essenburger, aminoacidophilic bacteria, and Enterobacter citrate was increased before Helicobacter pylori eradication, while the abundance of at least one of the aforementioned macromonas, Lacrimala amygdalina, Ruminococcus rumeniformis, Holdmannii suis, Glucospermia canis, Bretschneidera sinensis, and Bacteroides fecalith was decreased, the subjects corresponding to the aforementioned ex vivo fecal samples were classified as low-risk groups for gastric cancer.

[0032] In the above embodiments, when the abundance of at least one of the aforementioned *Macromonas monomorpha*, *Lastilus alba*, *Ruminococcus rumeni*, *Holdman's swine*, *Glucobacterium canis*, *Brutella bursa-palatinea*, and *Bacteroides fecalith* increased before *Helicobacter pylori* eradication, while the abundance of at least one of *Enteromorphococcus spp.*, *Masseudomonas muscarinicus*, *Amoebae dolichospasmia*, *Fumigia chinensis*, *Brutella brevicula*, *Koala faecalis*, *Clostridium perfringens*, *Clostridium erysipelas*, *Tasmania esculenta*, *Aminophila*, and *Enterobacter citrate* decreased, the subjects corresponding to the aforementioned ex vivo fecal samples were classified as high-risk groups for gastric cancer.

[0033] In the above embodiments, after the subjects eradicated Helicobacter pylori, compared with the subjects before the eradication of Helicobacter pylori, the abundance of Megamonas monomorpha in the ex vivo fecal samples of the aforementioned high-risk group for gastric cancer was reduced, while the abundance of at least one of Wechslerae Broutella, Clostridium perfringens, and Enterobacter citrate was increased.

[0034] In the above embodiments, when the abundance of at least one of the aforementioned enterobacteria, human rotavirus, focaccia dorsalis, and chicken ructoid bacteria increased after the eradication of Helicobacter pylori, while the abundance of at least one of Bacteroides fragilis, Enterobacter baumannii, Clostridium perfringens, Alisterella masei, Bacteroides nodosum, Fecal cocci, and Dilmuramys refractory was decreased, the subject corresponding to the aforementioned ex vivo fecal sample was classified into the low-risk group for gastric cancer.

[0035] In the above embodiments, when the subjects eradicated Helicobacter pylori, the abundance of at least one of the aforementioned Bacteroides fragilis, Enterobacter baumannii, Clostridium perfringens, Alisterella massa, Bacteroides nori, Fecal cocci, and Dilmuramys refractory increased, while the abundance of at least one of Enterobacter faecalis, Rochetomyces humanis, Focaccia dorsalis, and Rhus chinensis decreased, and the subjects corresponding to the aforementioned ex vivo fecal samples were classified as high-risk groups for gastric cancer.

[0036] In the above embodiments, after the subjects eradicated Helicobacter pylori, the abundance of at least one of Bacteroides fragilis, Enterobacter baumannii, Clostridium perfringens, Alisteria massa della argentea, Bacteroides nori, Coccidia fecalithii, and Dilmura dystrophiae in the ex vivo fecal samples of the aforementioned high-risk group for gastric cancer was greater than that of the low-risk group for gastric cancer.

[0037] In some specific cases, the relative abundance ratio of Firmicutes to Bacteroidetes in the fecal microbiota of the pre-sterilization gastric cancer risk groups (including high-risk and low-risk groups) and the control group was significantly different. In other specific cases, the relative abundance ratio of Firmicutes to Bacteroidetes in the fecal microbiota of the pre-sterilization gastric cancer risk groups was significantly lower than that of the control group.

[0038] It should be further explained that, compared with existing invasive examinations, the reagent for detecting fecal microbiota in this invention uses primer pairs with specific sequences to detect and compare the types and abundance of indicator bacteria in the microbiota. This can accurately determine whether subjects before and after Helicobacter pylori eradication are classified into low-risk or high-risk groups for gastric cancer, and can also improve patients' willingness to undergo examination. In the future, it can be applied to large-scale universal screening.

[0039] The following examples illustrate the application of the present invention, but they are not intended to limit the invention. Those skilled in the art can make various modifications and refinements without departing from the spirit and scope of the invention.

[0040] Example 1: Evaluation Method

[0041] 1.1 Patient population

[0042] This trial was conducted at the National Cheng Kung University College of Medicine Hospital and the outpatient department of Tainan Hospital, Ministry of Health and Welfare. Patients with early or advanced gastric precancerous lesions and a Helicobacter pylori-negative control group were recruited in a 1:1:1 ratio. Informed consent was obtained from all participants before enrollment in each group. Exclusion criteria included a diagnosis of gastrointestinal cancer, use of anti-inflammatory drugs (e.g., aspirin, nonsteroidal anti-inflammatory drugs, and selective cyclooxygenase-2 inhibitors) or drugs that may affect gastrointestinal microbiota colonization and immune function (e.g., proton pump inhibitors, steroids, and / or antibiotics), or ingestion of probiotics or yogurt at least twice weekly for one month prior to enrollment. This trial was approved by the Research and Ethics Committee of National Cheng Kung University College of Medicine Hospital (Approval No.: B-BR-109-012; Trial Registration Identification Code: NCT04527055, Website: ClinicalTrials.gov).

[0043] 1.2 Gastric pathological assessment

[0044] Each subject underwent five gastric mucosal biopsies under endoscopy: two from the antrum (located within 3 cm of the pylorus at the lesser and greater curvatures, respectively), one from the incisura, and two from the corpus (located at the midpoints of the lesser and greater curvatures, respectively). A pathologist unaware of the patient's background assessed the gastric histology using the updated Sydney system and converted the results into an OLGA / OLGIM assessment. This embodiment combines both OLGA and OLGIM, with the advanced stage corresponding to OLGA or OLGIM stages three to four, and the early stage to OLGA or OLGIM stages zero to two.

[0045] 1.3 Sample Collection and Fecal Microbial DNA Extraction

[0046] Please refer to Figure 1, which is a partial flowchart illustrating a method for detecting fecal microbiota according to an embodiment of the present invention.

[0047] As shown in Method 100 of Figure 1, all subjects provided fresh stool samples upon joining their respective groups (e.g., squares 101, 103, and 105), as shown in squares 111 and 113. As shown in squares 121 and 123, Helicobacter pylori-positive subjects received 10 or 14 days of bismuth-based quadruple therapy and provided fresh stool samples at least 6 months after Helicobacter pylori eradication, which were confirmed using the 13C UBT (13C urea breath test). Fresh stool samples were collected using sample collection containers (stool collection tubes with DNA stabilizers, Invitek Diagnostics, GmbH, Berlin, Germany) and stored at -80°C. Next, as shown in box 125, total genomic DNA was extracted using a commercially available kit (QIAamp® Fast DNA Stool Mini Kit, QIAGEN GmbH, Hilden, Germany), and full-length 16S rRNA sequencing was performed to screen for fecal microbiota. The DNA concentration was then quantified using a commercially available fluorescence detection device (Qubit 4.0, Thermo Scientific) and adjusted to 1 ng / µL. All samples were stored at -80°C before processing.

[0048] It should be noted that the analysis of fecal microbiota in the high-risk group, low-risk group and control group for gastric cancer before sterilization can omit blocks 121 and 123.

[0049] 1.4 PCR amplification and purification

[0050] This embodiment uses barcoded 16S gene-specific primers to amplify the full-length 16S rRNA gene (including V1 to V9 regions) of fecal microbiota. Based on the Multiplexed SMRTbell Library (SMRTbell Library) preparation and sequencing procedures, the full-length 16S gene is amplified using barcoded primers (Pacific Biosciences, California, USA). This embodiment designs each primer to have a 5' phosphate-modified 5' buffer sequence (GCATC), a 16-base barcode sequence, and a degenerate 16S gene-specific forward or reverse primer sequence. The forward primer (27F universal primer) consists of the following sequence from 5' to 3': 5' Phos / GCATC – 16bp barcode sequence – AGRGT TYGAT YMTGG CTCAG – 3'. The reverse primer (1492R universal primer) has the following sequence from 5' to 3': 5' Phos / GCATC – 16bp barcode sequence – RGYTA CCTTG TTACG ACTT – 3'. The degenerate bases are defined as follows: R represents A, G; Y represents C, T; M represents A, C. The aforementioned 16bp barcode sequence is used to identify the source of the sequence and is common knowledge in the technical field of this invention, therefore it will not be described further. In this example, PCR was performed using 2 ng gDNA with KAPA HiFi HotStart ReadyMix (Roche), and the PCR products were analyzed using 1% agarose gel. The nucleic acid concentration and band size in the library were checked to ensure they met the requirements. Samples with bright main bands (approximately 1,500 bp) were selected, purified, and recovered using AMPure PB beads to prepare the subsequent library.

[0051] 1.5 SMRTbell Database Construction and Sequencing

[0052] This embodiment collects equal molar amounts of the aforementioned barcoded PCR products and uses 500 to 1,000 ng of the collected amplicon samples for DNA damage repair. This embodiment utilizes end repair / A-tailing and ligation steps to introduce a universal hairpin adapter into a double-stranded DNA fragment. Next, the reaction product is purified using AMPure PB beads to remove adapter dimers. Then, using Sequel II primer 3.1 and Sequel II binding kit 3.1, the SMRTbell database is used for primer binding and polymerase conjugation. Subsequently, using a PacBio Sequel IIe instrument, sequencing is performed in circular consensus sequence (CCS) mode to generate HiFi reads.

[0053] 1.6 Ordinal Data Preprocessing

[0054] This embodiment uses SMRT Link v8.0 software to mutually calibrate the subreads from a single ordering space in Example 5, obtaining a circular consensus sequence read. This sequence is a consensus sequence generated by aligning subreads in a single zero-mode waveguide through at least three passes. Random ordering errors on the subreads can be corrected through the consensus sequence. This embodiment selects Q30 HiFi reads, whose read quality (Q) or Phred quality score is greater than 30, to analyze the full-length 16S ribosomal genes of the bacterial community, achieving an accuracy of 99.9%.

[0055] 1.7 Bioinformatics Analysis

[0056] This embodiment uses DADA2 (version 1.20) to process HiFi reads after demultiplexing to obtain amplicon reads with single nucleotide resolution. The DADA2 workflow includes quality filtering, deduplication, learning dataset-specific error models, amplicon sequence variant (ASV) inference, and chimera removal. The aforementioned pruning and filtering criteria allow for a maximum of two expected errors per read (maxEE = 2). After DADA2 analysis, accurate ASVs are generated, with a near-zero single nucleotide resolution error rate for the full-length 16S rRNA gene. Next, based on information retrieved from the NCBI 16S ribosomal RNA database, each representative sequence is annotated and classified using the feature classifier and the VSEARCH algorithm in QIIME2 (v2022.11; https: / / qiime2.org / ). Then, QIIME2 alignment MAFFT was used to perform multiple sequence alignment analysis on the sequence similarity and diversity among different ASVs, and a set of phylogenetic trees representing ASV sequences was constructed using QIIME2 phylogeny FastTree.

[0057] Next, QIIME was used to calculate alpha diversity indices, including the Menhinick and Margalef indices for abundance, the Shannon and Simpson indices for diversity, and the Pielou indices for evenness. The microeco tool in the R suite was used to generate beta diversity analysis and dissimilarity matrices (called Bray-Curtis), as well as unweighted and weighted UniFrac distances. Permutational multivariate analysis of variance (PERMANOVA) was used to determine differences in microbial community structure between and within groups. Then, dimensionality reduction methods, including principal component analysis (PCA) and principal coordinate analysis (PCoA), were used to visualize the multidimensional data. Finally, PCA and PCoA were performed using the microeco and vegan tools in the R suite, and plots were generated using ggplot2 and plotly in the R suite (version v4.0.5).

[0058] Subsequently, linear discriminant analysis (LDA) effect size (LEfSe) was used to select bacterial species with statistically significant differences between groups. The Kruskal-Wallis test and the Wilcoxon rank-sum test were used to perform LEfSe to identify bacterial taxa with significant differences in relative abundance between groups. Taxa with an LDA score (log 10) greater than 2 were considered to have significant differences.

[0059] 1.8 Statistical Analysis

[0060] Hypothesis testing was performed to assess differences in fundamental characteristics among study groups using analysis of variance and the chi-square test. The Kruskal-Wallis test was used to assess differences in alpha diversity among groups, and permutational multivariate analysis of variance (PERMANOVA) was used to assess differences in beta diversity. Hypothesis testing was performed to assess differences in microbial abundance among the three groups using either the Kruskal-Wallis test or analysis of variance. The Wilcoxon rank-sum test or Welch's t-test was used to assess differences in microbial abundance between two groups, and the Wilcoxon signed-rank test was used to assess differences in microbial abundance between paired samples. Multiple comparisons were performed using the Benjamini-Hochberg procedure for correction. According to the Wilcoxon rank-sum test, if the effect size, alpha, and statistical power are 0.75, 0.05, and 0.9, respectively, the recommended sample size is 41 participants per group. Considering the missing data rate of approximately 10%, 45 participants were recruited for each group. All tests were two-tailed tests, and p < 0.05 was considered statistically significant.

[0061] Based on the maximum Youden index determined by the receiver operating characteristic curve (ROC), an optimal cutoff value for the relative abundance of fecal microbiota was obtained, and a gastric cancer microbial risk scoring system was established accordingly to better distinguish between the high-risk and low-risk groups for gastric cancer. A Poisson distribution was assumed to be used to calculate the 95% confidence intervals (CIs) for various ratios for precise testing.

[0062] Example 2: Basic characteristics of the subjects

[0063] Please refer to Table 1, which lists the basic characteristics of the subjects. Among the Helicobacter pylori-positive subjects, 45 patients who met the criteria for OLGA and OLGIM combined with advanced stage were classified as the high-risk group (High risk bf). In addition, 45 Helicobacter pylori-positive subjects who met the criteria for OLGA and OLGIM combined with early stage and were matched with the high-risk group in terms of age and sex, and 45 Helicobacter pylori-negative subjects were classified as the low-risk group and the control group, respectively, as shown in Figure 1.

[0064] Next, please refer to Table 1, which lists the basic information of the subjects. The results in Table 1 show that there were no significant differences in mean age and gender distribution among the three groups. The ages of the high-risk group, low-risk group, and control group were 60.3 ± 9.2 years, 60.2 ± 10.2 years, and 59.7 ± 10.8 years, respectively (P = 0.96). Females accounted for 53.3% of the participants in all three groups (P>0.99). There were no significant differences in the distribution of height, weight, body mass index (BMI), smoking status, alcohol consumption, and diet among the three groups. The distributions of OLGA stage (P = 0.1), OLGIM stage (P = 0.11), smoking status (P = 0.50), alcohol consumption (P > 0.99), and diet (P > 0.99) showed no significant differences between the low-risk group and the control group.

[0065] Table 1: Basic Characteristics of Subjects N ± SD, (%) High-risk group (n = 45) Low-risk group (n = 45) Controls (n = 45) P Age (years) 60.3 ± 9.2 60.2 ± 10.2 59.7 ± 10.8 0.96 Gender (female) 24 (53.3) 24 (53.3) 24 (53.3) > 0.99 Height (cm) 162.9 ± 7.9 162.3 ± 8.7 163.4 ± 7.3 0.81 Weight (kg) 64.1 ± 11.7 65.4 ± 12.7 63.5 ± 12.2 0.76 Body Mass Index (BMI) 24.1 ± 3.5 24.7 ± 4.0 23.6 ± 3.4 0.36 Smoking status 2 (4.4) 6 (13.3) 4 (8.9) 0.33 Alcohol consumption 8 (17.8) 4 (8.9) 4 (8.9) 0.32 Diet (Non-vegetarian:Vegetarian:Vegan) 44:1:0 43:2:0 43:2:0 0.81 Diabetes 14 (31.1) 12 (26.7) 8 (17.8) 0.33 Helicobacter pylori infection 45 (100%) 45 (100%) 0 (0) < 0.001 OLGA stage (0:I:II:III:IV) 1:3:1:29:11 14:18:13:0:0 24:13:8:0:0 < 0.001 OLGIM Phase (0:I:II:III:IV) 19:1:4:17:4 32:3:10:0:0 38:4:3:0:0 < 0.001 OLGA / OLGIM Combined Phase (0-II vs. III-IV) 0:45 45:0 45:0 < 0.001

[0066] In Table 1, all participants in the high-risk (bf) and low-risk (bf) groups were grouped together as the pre-cure Helicobacter pylori infection group (Hp infect bf). The 35 participants in the Hp infect bf group provided fresh stool samples after eradication. The median time between eradication and subsequent stool sample collection was 318 days, with an interquartile range (IQR) of 273 to 404 days. These 35 participants were then classified as the post-cure Helicobacter pylori infection group (Hp infect af).

[0067] Example 3: Analysis of differences in fecal microbiota

[0068] Please refer to Figure 2A, which is a Venn diagram showing the fecal microbiota of a high-risk group (high risk bf), a low-risk group (low risk bf), and a control group before sterilization in some embodiments of the present invention. In Figure 2A, the high-risk group (high risk bf) includes blocks 201, 211, 213, and 221; the low-risk group (low risk bf) includes blocks 203, 211, 215, and 221; and the control group includes blocks 205, 213, 215, and 221. The relevant values ​​are shown in Table 2.

[0069] Table 2. Block ASV Ratio and Remarks: Block 201 4804 25.7% High-risk group; Block 203 4923 26.3% Low-risk group; Block 205 5385 28.8% Control group; Block 211 714 3.8% Intersection of high-risk and low-risk groups; Block 213 773 4.1% Intersection of high-risk and control groups; Block 215 865 4.6% Intersection of low-risk and control groups; Block 221 1251 6.7% Intersection of high-risk, low-risk, and control groups.

[0070] The Venn diagram in Figure 2A shows that the number of amplicon sequence variants (ASVs) specific to the high-risk group (high-risk bf) of gastric cancer is 4,804, the number of ASVs specific to the low-risk group (low-risk bf) of gastric cancer is 4,923, and the number of ASVs specific to the control group is 5,385. There are a total of 1,251 common ASVs among the three groups (Figure 2A).

[0071] Please refer to Figure 2B, which is a stacked bar graph 230 showing the species and abundance percentage of the top 10 fecal microbiota in each group of Figure 2A. The results in Figure 2B show that the distribution of the top 10 most abundant bacterial species in the high-risk group (high risk bf) and the low-risk group (low risk bf) is different from that in the control group.

[0072] Please refer to Figure 2C, which is a box plot 240 showing the Firmicutes / Bacteroidetes ratio for each group in Figure 2B. Figure 2C shows that there were significant differences in the Firmicutes / Bacteroidetes ratio among the three groups (P < 0.001). There were significant differences between the high-risk group (high risk bf) and the control group (P = 0.002), and also between the low-risk group (low risk bf) and the control group (P < 0.001), but no significant difference was found between the high-risk group (high risk bf) and the low-risk group (low risk bf) (P = 0.72).

[0073] Example 4: Differences in α and β diversity among populations

[0074] Please refer to Figures 3A to 3F, which respectively show the box-and-whisker plot 301 (Figure 3A) of the fecal microbiota and α-diversity related Menhinick index, the box-and-whisker plot 302 (Figure 3B) of Shannon entropy analysis, the box-and-whisker plot 303 (Figure 3C) of Pielou index, the principal coordinates analysis (PCoA) 304 (Figure 3D) of Bray-Curtis distance matrix related to β-diversity, the principal coordinates analysis 305 (Figure 3E) of unweighted UniFrac matrix, and the principal coordinates analysis 306 of weighted UniFrac matrix (Figure 3F) for a high-risk group (high-risk bf) of gastric cancer, a low-risk group (low-risk bf) of gastric cancer, and a control group before sterilization according to an embodiment of the present invention.

[0075] Figures 3A to 3C show that, comparing the intragroup α diversity among the three groups, the Menhinick index (P = 0.62, Figure 3A) and Margalef index (P = 0.7) represent abundance, the Shannon entropy (P = 0.44, Figure 3B) and Simpson analysis (P = 0.42) represent diversity, and the Pielou index (P = 0.59, Figure 3C) represents evenness. These results indicate that there are no significant differences among the groups.

[0076] Figures 3D to 3F show the β-diversity of bacterial species complexity differences among the three groups, using PCoA based on the Bray-Curtis dissimilarity matrices (Figure 3D), unweighted UniFrac distance (Figure 3E), and weighted UniFrac distance (Figure 3F). A significant difference was found between the high-risk group (bf) and the control group using weighted UniFrac distance (PERMANOVA, r² = 0.070, adjusted P = 0.002). A significant difference was also found between the low-risk group (bf) and the control group using weighted UniFrac distance (r² = 0.082, adjusted P = 0.002). However, Bray-Curtis and unweighted UniFrac distances showed no significant differences (High-risk bf vs. control group, Bray-Curtis, r² = 0.012, adjusted P = 0.108; unweighted UniFrac, r² = 0.016, adjusted P = 0.093; Low-risk bf vs. control group, Bray-Curtis, r² = 0.011, adjusted P = 0.833; unweighted UniFrac distance (r² = 0.016, adjusted P = 0.093). There was no significant difference between the High-risk bf and Low-risk bf groups (Bray-Curtis, r² = 0.011, adjusted P = 0.833). 0.833; unweighted UniFrac,r² = 0.001, adjusted P = 0.396; weighted UniFrac,r² = 0.003, adjusted P = 0.981).

[0077] Example 5: Taxonomic units with varying abundance among different populations of Helicobacter pylori before eradication

[0078] Please refer to Figures 4A to 4C. Figure 4A shows the abundance analysis of fecal microbiota in the high-risk group (Bf) and control group of gastric cancer before sterilization according to an embodiment of the present invention (Figure 401). Figure 4B shows the abundance analysis of fecal microbiota in the low-risk group (Bf) and control group of gastric cancer before sterilization according to an embodiment of the present invention (Figure 403). Figure 4C shows the abundance analysis of fecal microbiota in the high-risk group (Bf) and low-risk group (Bf) of gastric cancer before sterilization according to an embodiment of the present invention (Figure 405). The left graph of Figures 4A to 4C shows the mean percentage abundance of the two groups in a box plot, and the right graph shows the difference in the mean percentage abundance of the two groups (95% confidence interval; the color of the dots represents the group with higher abundance).

[0079] Figures 4A to 4C show the relative abundance of strains in the high-risk group (BF), low-risk group (BF), and control group before Helicobacter pylori eradication. Compared with the control group, Prevotella copri DSM 18205 was significantly increased in both the high-risk group (Figure 4A) and low-risk group (Figure 4B), while Bacteroides stercoris ATCC 43183 was significantly decreased (Prevotella copri DSM 18205, P = 0.0004 and 0.0002; Bacteroides stercoris ATCC 43183, P = 0.005 and 0.004, respectively).

[0080] Furthermore, there are taxonomic units with significant differences in abundance between the high-risk group (BF) and the low-risk group (BF) for gastric cancer. Figure 4C shows that in the high-risk group (high-risk bf) for gastric cancer, the levels of *Megamonas funiformis* YIT 11815 (P = 0.029), *Lacrimispora amygdalina* (P = 0.036), *Ruminococcus callidus* (P = 0.046), *Holdemanella porci* (P = 0.008), *Gluceribacter canis* (P = 0.002), and *Blautia luti* (P = 0.025) were significantly increased. However, the levels of *Acidaminococcus intestini* (P = 0.032), *Negativibacillus massiliensis* (P = 0.047), and *Amedibacillus dolichospasmica* were significantly increased. The abundance of *Fumia dolichus* (P = 0.020) and *Fumia xinanensis* (P = 0.042) was significantly reduced. The relative abundance of *Revoirius copripulae* DSM 18205 (P = 0.874) and *Bacteroides fecalis* ATCC 43183 (P = 0.989) did not differ significantly between the high-risk and low-risk groups for gastric cancer.

[0081] Please refer to Figure 5. Figure 5 shows the abundance trend analysis of fecal microbiota before sterilization in an embodiment of the present invention. Using LEfSe≧ 2.0 analysis, the trend of fecal microbiota in the control group, the low-risk group (low risk bf) and the high-risk group (high risk bf) was determined to be increasing or decreasing. The abundance of *Blautia wexlerae* DSM 19850 (P = 0.021), *Phascolarctobacterium faecium* (P = 0.014), *Lachnoclostridium phocaeense* (P = 0.033), *Erysipelatoclostridium ramosum* (P = 0.002), *Eisenbergiella tayi* (P = 0.033), *Anaerotignum aminivorans* (P = 0.040), and *Enterocloster citroniae* (P = 0.002) varied from the control group to the low-risk group (Bf) and then to the high-risk group (Bf). bf) showed a gradually decreasing trend, while Bacteroides faecis MAJ27 (P = 0.011) showed a gradually increasing trend.

[0082] Example 6: Taxonomic units with varying abundance among different populations after Helicobacter pylori eradication

[0083] Please refer to Figure 6. Figure 6 shows the abundance analysis of fecal microbiota in a high-risk group and a low-risk group for gastric cancer after sterilization according to another embodiment of the present invention. The left figure shows the mean percentage abundance box plot, and the right figure shows the difference in mean percentage abundance (95% confidence interval, and the color of the dots represents the group with higher abundance).

[0084] Figure 6 shows the relative abundance of strains between the high-risk group and the low-risk group for gastric cancer after Helicobacter pylori eradication. In the high-risk group for gastric cancer, the levels of Bacteroides fragilis (P = 0.004), Enterobacter bolteae (P = 0.018), Lachnoclostridium pacaense (P = 0.028), Dialister massiliensis (P = 0.016), Bacteroides nordii WAL 11050 (i.e. JCM 12987, P = 0.011), Coprococcus catus (P = 0.022), and Dielma fastidiosa (P = 0.033) were significantly increased.

[0085] Please refer to Figure 7, which shows the abundance trend analysis of fecal microbiota after sterilization according to another embodiment of the present invention. Using LEfSe≧ 2.0 analysis, the trend of increasing or decreasing fecal microbiota from the control group to the low-risk group for gastric cancer (Low risk af) to the high-risk group for gastric cancer (High risk af) is analyzed. The abundance of *Faecalibacillus intestinalis* (P = 0.010), *Roseburia hominis* A2_183 (P = 0.041), *Dorea phocaeensis* (P = 0.001), and *Ructibacterium gallinarum* (P = 0.006) gradually decreased from the control group to the low-risk group for gastric cancer (Low risk af) to the high-risk group for gastric cancer (High risk af).

[0086] Example 7: Indicator bacteria used as biomarkers to distinguish between high-risk and low-risk groups for gastric cancer before Helicobacter pylori eradication.

[0087] This embodiment evaluates whether the top 10 most abundant bacterial species in Figure 4C have the potential to serve as biomarkers for assisting in predicting and distinguishing between the high-risk and low-risk groups for gastric cancer (hereinafter referred to as microbial risk score 1).

[0088] Please refer to Table 3, which shows the optimal cutoff values ​​for the relative abundance of fecal microbiota species to distinguish between the high-risk bf and low-risk bf groups for gastric cancer, as well as the definition of the scoring system.

[0089] Table 3. Cutoff values ​​for relative abundance of fecal bacteria species (0 points, 1 point): Megamonas funiformis YIT 11815 0.0006615945 < ≥ Lacrimispora amygdalina 0.0016900495 < ≥ Ruminococcus callidus 0.0000863035 < ≥ Holdemanella porci 0.0002719795 < ≥ Gluceribacter canis 0.0005336830 < ≥ Blautia luti 0.0000551330 < ≥ Acidaminococcus intestini 0.0003289255 ≥ < Negativibacillus massiliensis 0.0002715880 ≥ < Amedibacillus dolichus 0.0001236705 ≥ < Fumia xinanensis 0.0000875810 ≥ <

[0090] In Table 3, a score of 1 is defined as a positive correlation between the bacterial species and the high-risk group (high risk bf) for gastric cancer, in which case the relative abundance of the microorganism is greater than or equal to (≥) the cutoff value; if the bacterial species is negatively correlated with the high-risk group (high risk bf), then the relative abundance of the microorganism is less than the cutoff value. In the above embodiment, the microbial risk score 1 was established by summing the scores of 10 fecal bacteria species (also known as indicator species). The predictive value of the microbial risk score 1 gradually increases from the total score of 1 to 10 (0% [0 / 1], 0% [0 / 3], 0% [0 / 9], 32.0% [8 / 25], 50.0% [9 / 18], 70.6% [12 / 17], 90.0% [9 / 10], 100% [5 / 5], 100% [1 / 1] and 100% [1 / 1], P < 0.001). In addition, if the total score of the microbial risk score 1 is greater than or equal to (≧) 5, it can effectively improve the sensitivity and specificity of predicting the high-risk group for gastric cancer.

[0091] Table 4. Microbial Risk Score 1 for Differentiating Gastric Cancer Risk Score 1 N (%) High risk group for gastric cancer (n = 45) Low risk group for gastric cancer (n = 45) Total ≥ 5 37 15 52 < 5 8 30 38

[0092] Please refer to Figure 8, which shows the ROC curves of a fecal microbiota analysis method used in an embodiment of the present invention to assist in predicting the high-risk group for gastric cancer. Curve 801 represents the ROC curve of microbial risk score 1, and reference line 803 represents the diagonal reference line. The results in Figure 8 show that the area under the receiver operating characteristic curve (AUROC) for microbial risk score 1 is 0.84 (95% CI, 0.76–0.92), and the optimal value for the ROC curve of microbial risk score 1 is 5.5. Furthermore, Table 5 shows that when the cutoff value of microbial risk score 1 is ≥ 5, its sensitivity is 82.2% (95% CI, 68.7%–90.7%), its specificity is 66.7% (95% CI, 52.1%–78.6%), and the likelihood ratios for positive and negative tests are 3.38 and 0.37, respectively, indicating that it can predict the risk of high-risk precancerous lesions of the stomach. When the prevalence of high risk of gastric cancer was 50%, 33.3%, and 20%, the predictive accuracy of the microbial risk score 1 was 74.4%, 71.9%, and 69.8%, respectively.

[0093] Furthermore, from the top 10 most abundant bacterial species in Figure 4C, the top 5 most abundant bacterial species in either the high-risk (bf) or low-risk (bf) group for gastric cancer (hereinafter referred to as Microbial Risk Score 2) were further selected, including *Megamonas funiformis* YIT11815, *Lacrimispora amygdalina*, *Ruminococcus callidus*, *Negativibacillus massiliensis*, and *Acidaminococcus intestini* to distinguish between the high-risk (bf) and low-risk (bf) groups for gastric cancer. The ROC curve of Microbial Risk Score 2 showed an optimal value of 2.5. Secondly, the results in Table 5 show that when the cutoff value of the microbial risk score 2 is ≥ 2, its sensitivity is 91.1% (95% CI, 78.8%–97.5%), its specificity is 24.4% (95% CI, 12.9%–39.5%), and the likelihood ratios for positive and negative tests are 2.05 and 0.62, respectively, which seems to predict the risk of high-risk precancerous lesions of the stomach. However, when the prevalence of high-risk stomach cancer is 50%, 33.3%, and 20%, respectively, the predictive accuracy of the microbial risk score 2 gradually decreases to 57.8%, 46.7%, and 37.8%, respectively, which is lower than that of the microbial risk score 1.

[0094] Please refer to Figure 9, which shows the ROC curves for predicting high-risk gastric cancer using fecal microbiota analysis according to another embodiment of the present invention. Curve 901 represents the ROC curve for predicting high-risk gastric precancerous lesions using the top 10 most abundant indicator bacteria (i.e., microbial risk score 1), curve 903 represents the ROC curve for predicting high-risk gastric precancerous lesions using the top 5 most abundant indicator bacteria (i.e., microbial risk score 2), and reference line 905 represents the diagonal reference line. The results in Figure 9 show that microbial risk score 1 has a higher AUROC than microbial risk score 2.

[0095] Additionally, please refer to Table 5, which compares the prediction of high-risk gastric precancerous lesions using microbial risk score 1 and microbial risk score 2. The results in Table 5 show that a microbial risk score 1 score greater than or equal to (≥) 5 is significantly more effective in predicting the high-risk group for gastric cancer prognosis than a microbial risk score 2 score greater than or equal to (≥) 2. This includes higher specificity, likelihood ratio, and accuracy of microbial risk score 1 compared to microbial risk score 2. Particularly when the disease prevalence decreases from 50% to 33.3% or 20%, microbial risk score 1 still maintains approximately 70% accuracy, while the accuracy of microbial risk score 2 falls below 50%. Therefore, the aforementioned microbial risk score 1 (using the top 10 most abundant indicator bacteria) is more accurate in predicting high-risk gastric precancerous lesions than microbial risk score 2 (using the top 5 most abundant indicator bacteria).

[0096] Table 5. Comparison of Microbial Risk Score 1 and Microbial Risk Score 2 in predicting high-risk gastric precancerous lesions. Microbial Risk Score 1 (using the top 10 most abundant indicator bacteria) Microbial Risk Score 2 (using the top 5 most abundant indicator bacteria) AUROC 0.84 (95% CI, 0.76–0.92) 0.74 (95% CI, 0.64–0.85) Optimal cutoff value ≥ 5 ≥ 2 Sensitivity 82.2% (95% CI, 68.7%–90.7%) 91.1% (95% CI, 78.8%–97.5%) Specificity 66.7% (95% CI, 52.1%–78.6%) 24.4% (95% CI, 12.9%–39.5%) Likelihood ratio for positive test 3.38 2.05 For negative tests: 0.37, 0.62. Accuracy: Prevalence 50%: 74.4%, 57.8%; Prevalence 33.3%: 71.9%, 46.7%; Prevalence 20%: 69.8%, 37.8%.

[0097] Previous literature has shown an association between the human gut microbiota and various diseases. However, the above examples identified a correlation between the differential abundance of fecal microbiota and the severity of precancerous lesions in the stomach. The abundance of *Macromonas monomorpha* YIT 11815 increased in the high-risk group for gastric cancer, but significantly decreased after *Helicobacter pylori* eradication, showing no significant difference from the control group. Notably, as precancerous lesions progressed from the control group to the low-risk group and then to the high-risk group, the abundance of *Weikorla-Brutella* DSM 19850, *Clostridium perfringens*, and *Enterobacter citrate* gradually decreased; however, after *Helicobacter pylori* eradication, the abundance of *Weikorla-Brutella*, *Clostridium perfringens*, and *Enterobacter citrate* significantly increased, eventually reaching levels comparable to the control group. Since *Helicobacter pylori* eradication therapy can reduce the risk of gastric cancer, the fluctuations in these gut microbiota after *Helicobacter pylori* eradication therapy represent a possible role of the gut microbiota in the development of gastric cancer. Therefore, these gut microbiota could be used as biomarkers to predict the occurrence of gastric cancer in the future. A microbial risk score of 1 based on these 10 bacterial strains, with a total score greater than or equal to (≥) 5, can effectively predict a high-risk group for gastric cancer. However, the abundance of *Bacteroides fowleri* MAJ27, which increased in the high-risk group for gastric cancer, did not change significantly after *Helicobacter pylori* eradication, indicating the persistent presence of *Bacteroides fowleri* after *Helicobacter pylori* eradication therapy. This may be related to the fact that some patients still develop gastric cancer after eradication. These results suggest that these fecal microbiota may be involved in gastric cancer formation, and that *Helicobacter pylori* eradication may have prevented gastric cancer formation, leading to an increase in potentially protective microorganisms. Furthermore, *Bacteroides fowleri* MAJ27 is a potentially harmful microorganism that was not suppressed after *Helicobacter pylori* eradication; its role in gastric cancer development should be evaluated in the future.

[0098] Literature has revealed the characteristics of gastric mucosal microbiota at various stages of gastric cancer development. This literature identified eight gastric bacterial communities as universal biomarkers to distinguish between gastric cancer and superficial gastritis, with an area under the curve (AUC) of 0.85. Bacterial biomarkers have the potential to predict disease; however, such biomarkers still require endoscopic mucosal biopsy to obtain gastric microbiota. Even in individuals at risk for gastric cancer, large-scale screening using endoscopic mucosal biopsies is not feasible. The embodiments of this invention utilize the top 10 most abundant indicator bacterial species as biomarkers to assist in predicting the prognostic risk of gastric cancer, and the results are shown in Table 5. Table 5 shows that the AUC for microbial risk score 1 is 0.84, effectively distinguishing between high-risk and low-risk precancerous lesions of the stomach. Furthermore, the likelihood ratios for positive and negative tests are 3.38 and 0.37, respectively. If the prevalence of high-risk groups for gastric cancer is 50%, 33.3%, and 20%, respectively, the predictive accuracy would be 74.4%, 71.9%, and 69.8%. Combined with Helicobacter pylori fecal antigen, the aforementioned fecal microbiota can serve as a biomarker and may be applied to large-scale universal screening in the future.

[0099] In summary, the above embodiments demonstrate that the reagent for detecting fecal microbiota of the present invention can indeed be used in the preparation of a test kit for assisting in the prediction of the risk of gastric cancer. It can determine the risk of future gastric cancer in subjects before Helicobacter pylori eradication in a non-invasive manner, and can also improve patients' willingness to undergo examination. It can be applied to large-scale universal screening in the future.

[0100] It should be added that this invention uses specific indications, specific bacterial species, specific sequences, or specific evaluation methods to illustrate the use of the reagent for detecting fecal microbiota in the preparation of a test kit for assisting in the prediction of gastric cancer risk. However, those skilled in the art will understand that other indications, other bacterial species, other sequences, or other evaluation methods can also be used to illustrate the use of the reagent for detecting fecal microbiota in the preparation of a test kit for assisting in the prediction of gastric cancer risk without departing from the spirit and scope of this invention, and are not limited to those described above. For example, the primer pair in Example 1 can be replaced with other label sequences according to actual needs, or other sequences can be selectively added to the 5' or 3' ends.

[0101] According to the above embodiments, the reagent for detecting fecal microbiota of the present invention is used in the preparation of a test kit for assisting in the prediction of the risk of gastric cancer. Its advantage lies in the use of primer pairs with specific sequences to detect and compare the types and abundance of indicator bacteria in the microbiota. This can accurately determine whether subjects before and after Helicobacter pylori eradication are classified into low-risk or high-risk groups for gastric cancer in a non-invasive manner, and can also improve patients' willingness to undergo examination. In the future, it can be applied to large-scale universal screening.

[0102] While the invention has been disclosed above with reference to several specific embodiments, other embodiments are also possible. Therefore, the spirit and scope of the appended claims should not be limited to the embodiments described herein.

[0103] 100: Method 101,103,105,111,113,121,123,125: Squares Blocks 201,203,205,211,213,215,221: 230: Stacked bar chart 240, 301, 302, 303: Box-whisker diagram 304, 305, 306: Principal Coordinate Analysis Chart 401, 403, 405, 601: Abundance Analysis Chart 501,701: Abundance Trend Analysis Chart 801, 901, 903: Curves 803, 905: Reference lines

[0104] Domestic storage information (please note in order of storage institution, date, and number): None. International storage information (please note in order of storage country, institution, date, and number): None. A0101_OR_PSEQ.xml

Claims

1. The use of a reagent for detecting fecal microbiota in the preparation of a test kit for assisting in the prediction of gastric cancer risk, wherein the test kit is used to detect a first ex vivo fecal sample of a subject before Helicobacter pylori eradication, and the first ex vivo fecal sample has not been diagnosed or identified as belonging to a low-risk group or a high-risk group for gastric cancer; a detection step is performed on the first ex vivo fecal sample to obtain a first detection result, wherein the detection step utilizes a first primer of the reagent as shown in SEQ ID NO: 1 and a second primer of the reagent as shown in SEQ ID NO: 2 for polymerase chain reaction, and the first detection result contains a plurality of 16S rRNA gene sequences, and the 16S rRNA gene sequences correspond to the abundance of a species of a first microbiota; and the abundance of the species of the first microbiota in the subject is compared with a cutoff value for the species, and it is determined whether the abundance of the species of the first microbiota meets the definition of a scoring system, wherein the definition of the scoring system includes: Megamonas monomorpha The first relative abundance of *Lactobacillus funiformis* YIT 11815 is greater than or equal to a first cutoff value; the second relative abundance of *Lacrimispora amygdalina* is greater than or equal to a second cutoff value; the third relative abundance of *Ruminococcus callidus* is greater than or equal to a third cutoff value; the fourth relative abundance of *Holdemanella porci* is greater than or equal to a fourth cutoff value; the fifth relative abundance of *Gluceribacter canis* is greater than or equal to a fifth cutoff value; the sixth relative abundance of *Blautia luti* is greater than or equal to a sixth cutoff value; the seventh relative abundance of *Acidaminococcus intestini* is less than a seventh cutoff value; and the eighth relative abundance of *Negativibacillus massiliensis* is less than an eighth cutoff value. The relative abundance of Amedibacillus dolichus was less than the 9th cutoff value;The tenth relative abundance of *Fumia xinanensis* is less than the tenth cutoff value, wherein the subject corresponding to the first ex vivo fecal sample is classified into the high-risk group for gastric cancer if at least five of the first bacterial phases meet the definition of the scoring system, and the first, second, third, fourth, fifth, sixth, seventh, eighth, ninth, and tenth cutoff values ​​are obtained by determining the species and abundance of the second bacterial phase in a normal individual, the species and abundance of the third bacterial phase in a high-risk individual for gastric cancer, and the species and abundance of the fourth bacterial phase in a low-risk individual for gastric cancer according to the maximum Youden index determined by the receiver operating characteristic curve, the normal individual does not have *Helicobacter pylori*, the high-risk individual for gastric cancer has *Helicobacter pylori* and meets the OLGA and OLGIM criteria for combined advanced stage, and the low-risk individual for gastric cancer has *Helicobacter pylori* and meets the OLGA criteria. Merging earlier standards with OLGIM.

2. The use of the reagent for detecting fecal microbiota as described in claim 1 in the preparation of a test kit for assisting in the prediction of the risk of gastric cancer, wherein the first cutoff value is 0.0006615945, the second cutoff value is 0.0016900495, the third cutoff value is 0.0000863035, the fourth cutoff value is 0.0002719795, the fifth cutoff value is 0.0005336830, the sixth cutoff value is 0.0000551330, the seventh cutoff value is 0.0003289255, the eighth cutoff value is 0.0002715880, the ninth cutoff value is 0.0001236705, and the tenth cutoff value is 0.0000875810.

3. The use of the reagent for detecting fecal microbiota as described in claim 1 or claim 2 in the preparation of a test kit for assisting in the prediction of the risk of gastric cancer, wherein when fewer than 5 of the first microbiota meet the definition of the scoring system, the subject corresponding to the first ex vivo fecal sample is classified as the low-risk group for gastric cancer.

Images