Prediction model for identifying alcoholic hepatitis and alcoholic cirrhosis and application

Abstract

The invention discloses a prediction model for identifying alcoholic hepatitis and alcoholic cirrhosis and application, the prediction model takes intestinal bacteria, fungi and age information as combined markers, the model is especially suitable for clinical application scenarios, can effectively identify potential ALC patients, can avoid excessive diagnosis of AH to the greatest extent, and has good application prospects. The comprehensive improvement of the model performance proves the strategy value of integrating clinical parameters and microbiological markers, and a new thought is provided for developing a more accurate ALD diagnostic tool.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of disease prediction model, in particular to a prediction model for identifying alcoholic hepatitis and alcoholic liver cirrhosis and application thereof. BACKGROUND

[0002] Alcohol-related liver disease (ALD) includes a series of liver pathological changes such as simple fatty degeneration, alcoholic hepatitis (AH), alcoholic liver cirrhosis (ALC) and hepatocellular carcinoma (HCC), which poses a great threat to human life and health. Epidemiological studies have shown that more than 90% of alcoholics have liver steatosis, of which about 10%-35% will develop into AH; among AH patients, about 10%-20% will progress to ALC; and among ALC patients, about 3%-10% will eventually develop into HCC. The clinical diagnosis of ALD relies on drinking history, liver disease signs, biochemical abnormalities and imaging features, and other liver disease causes need to be ruled out. However, more than 90% of early patients lack typical symptoms, resulting in most patients being diagnosed at the irreversible cirrhosis stage, and liver transplantation being the only effective treatment for end-stage.

[0003] It is worth noting that the core pathological mechanism of ALD involves alcohol-induced oxidative stress, lipid toxicity, and a vicious cycle of the intestinal-liver axis: alcohol damages the intestinal barrier function and triggers dysbiosis (such as a decrease in Firmicutes and an overproliferation of cytolysate Enterococcus faecalis), prompting bacterial / toxins to migrate to the liver, activating the inflammatory cascade and directly damaging hepatocytes, and thus feeding back to exacerbate intestinal lesions. Although intervention strategies targeting the intestinal microecosystem (such as probiotics, fecal transplantation, and phage therapy) show therapeutic potential, the mechanism of bacterial-fungal interaction in the evolution of AH to ALC is not clear, and its value as a non-invasive biomarker or intervention target needs to be verified. Given that abstinence is a decisive factor for improving prognosis (damage is reversible in mild to moderate patients), developing a new non-invasive model based on the microbiome to achieve early identification of AH and ALC and block disease progression has become a key requirement for current clinical translation. SUMMARY

[0004] The present application integrates the basic clinical information of patients, intestinal bacterial community characteristics and intestinal fungal flora structure to construct a multi-dimensional identification system. This model provides an innovative and practical molecular marker combination scheme for the accurate differential diagnosis of AH and ALC through the joint analysis of intestinal microecological flora (bacteria + fungi) and patient baseline data.

[0005] Therefore, the scheme of the present application is as follows: The first aspect of the present application is to propose a biomarker for identifying alcoholic hepatitis and alcoholic liver cirrhosis, which includes Lactobacillus plantarum (Lactobacillus plantarum) and Candida albicans (Candida albicans). s Lactiplantibacillus plantarum), unidentified species of the genus Bacillus ( s unclassified g Paenibacillus Mediterranean bacteria that decompose glycyrrhizic acid ( s__ Mediterraneibacter glycyrrhizinilyticus ), Pseudomonas aeruginosa ( s Zygosaccharomyces pseudobailii ) and unclassified fungi ( s unclassified k Fungi ).

[0006] Furthermore, the biomarkers also include age data.

[0007] A second aspect of the invention is to propose the application of a detection reagent for the biomarker described in the first aspect in the preparation of a predictive product for differentiating between alcoholic hepatitis and alcoholic cirrhosis.

[0008] Furthermore, the detection reagent is an abundance detection reagent, including primers for specific amplification of the V3-V4 variable region of the 16S rRNA gene for detecting bacteria, and primers for specific amplification of the ITS region for detecting fungi.

[0009] Preferably, the nucleotide sequences of the primers used for detecting bacteria are: upstream primer: 5'-AGRGTTYGATYMTGGCTCAG-3', downstream primer: 5'-RGYTACCTTGTTACGACTT-3'.

[0010] Preferably, the nucleotide sequences of the primers used for detecting fungi are: upstream primer: 5'-CTTGGTCATTTAGAGGAAGTAA-3'; downstream primer: 5'-TCCTCCGCTTATTGATATGC-3'.

[0011] Furthermore, the predicted product is a reagent kit, detection system, or instrument.

[0012] A third aspect of the invention is to provide a kit for differentiating alcoholic hepatitis from alcoholic cirrhosis, characterized in that the kit comprises a microbial abundance detection reagent, a prediction formula, and a prediction standard; the microbial abundance detection reagent is used to detect the abundance of the biomarker of claim 1; the prediction formula is: P(ALC)=1 / (1+exp(-(-9.234-14478.4)). s Lactiplantibacillus plantarum +3977114.078 s__ unclassified g Paenibacillus +61.995 s Mediterraneibacter glycyrrhizinilyticus -8323.763 s Zygosaccharomyces pseudobailii -16.92 s__ unclassified k Fungi+0.133 Age represents the patient's age; the predictive criterion is that when P(ALC) is greater than 0.369, the patient is diagnosed with alcoholic cirrhosis (ALC), and otherwise, the patient is diagnosed with alcoholic hepatitis (AH).

[0013] Furthermore, the kit also includes reagents for extracting the biomarkers from the sample.

[0014] A fourth aspect of the present invention is to provide an apparatus for differentiating between alcoholic hepatitis and alcoholic cirrhosis, comprising a detection unit and a data analysis unit, wherein: The detection unit is used to acquire samples, detect and determine the abundance of microorganisms in the samples; the microorganisms are the biomarkers described in the first aspect. The data analysis unit is used to analyze the detection results of the detection unit in conjunction with case information to make identification results.

[0015] Furthermore, the detection unit performs detection and obtains detection results according to the following operations: Test fecal samples to obtain DNA data; The data analysis unit is used to analyze and process the detection results of the detection unit, including: sequencing the DNA data of the fecal sample to obtain intestinal microbial genome data; The relative abundance information of each biomarker was obtained by analyzing the gut microbiome genome data. Based on the relative abundance information, gut microbiota characteristic data are determined.

[0016] Compared with the prior art, the present invention has the following beneficial effects: The combined biomarkers provided by this invention include a variety of intestinal bacteria and fungi, which are of great significance for differentiating alcoholic hepatitis from alcoholic cirrhosis.

[0017] The kit provided by this invention includes a predictive model that combines biomarkers with age information. This model is particularly suitable for clinical applications, effectively identifying potential ALC patients while minimizing overdiagnosis of AH. The comprehensive improvement in model performance demonstrates the value of the strategy of integrating clinical parameters and microbiome biomarkers, providing new ideas for developing more accurate ALD diagnostic tools. Attached Figure Description

[0018] Figure 1 This is a bar graph showing the Shannon and Invsimpson indices and the Wilcoxon rank-sum test results for fecal bacteria from the four groups of people described in the example.

[0019] Figure 2The results of the β-diversity analysis of fecal bacteria in the four groups of people described in the examples are shown.

[0020] Figure 3 These are the main components of fecal bacteria in the four groups of people described in the examples at the phylum and genus levels.

[0021] Figure 4 These are the main components of fecal bacteria at the species level in the four groups of people described in the examples.

[0022] Figure 5 This is a bar graph showing the Shannon and Invsimpson indices and the Wilcoxon rank-sum test results for fecal fungi in the four groups of people described in the examples.

[0023] Figure 6 The results of the β-diversity analysis of fecal fungi in the four groups of people described in the examples are shown.

[0024] Figure 7 These are the main components of the fecal fungi in the four groups of people described in the examples, at the phylum and genus levels.

[0025] Figure 8 These are the main components of the fecal fungi in the four groups of people described in the examples at the species level.

[0026] Figure 9 In this example, the bacteria ranked 10th in terms of the greatest influence on AH and ALC classification were selected by random forest feature importance screening.

[0027] Figure 10 In this example, the fungi ranked 10th in terms of the greatest influence on AH and ALC classifications were selected by random forest feature importance screening.

[0028] Figure 11 The AUC curves of the logistic regression model established by the combined bacteria in the training set (11A) and the AUC curves of the logistic regression model established by the combined bacteria in the training set and the validation set (11B) are shown in the examples.

[0029] Figure 12 The AUC curves of the logistic regression model established by the combined fungi in the training set (12A) and the AUC curves of the logistic regression model established by the combined fungi in the training set and the validation set (12B) are shown in the examples.

[0030] Figure 13 The AUC curves (13A) of the logistic regression model established by bacteria and fungi in the training set in the examples are shown; and the AUC curves (13B) of the logistic regression model established by bacteria and fungi in the training set and the validation set are shown.

[0031] Figure 14The AUC curves (14A) of the logistic regression model established by combining bacteria and fungi with patient age in the training set are shown in the examples; and the AUC curves (14B) of the logistic regression model established by combining bacteria and fungi with patient age in the training set and validation set are shown in the examples. Detailed Implementation

[0032] The technical solution of the present invention will now be clearly and completely described in conjunction with preferred embodiments. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0033] This invention relates to a bibliographical table with an English-Chinese equivalent, as follows:

[0034] This invention relates to a glossary of fungal classifications in English and Chinese, as shown below:

[0035] Note: "Unclassified fungi" refers to fungi that can only be identified at a certain level and cannot be further classified into a specific known category, such as... unclassified k Fungi This mainly refers to the fact that the sequencing sequence can only confirm that the microorganism belongs to the kingdom Fungi, but cannot be further classified. Additionally, "sp" is an abbreviation for species, indicating that the microorganism belongs to that genus, but a specific species has not been identified. Saccharomyces sp This indicates that the microorganism belongs to the genus *Yeast*, but no specific species has been identified. The relative abundance of major unclassified fungi = (number of unclassified sequences / total number of fungal sequences) × 100%.

[0036] The sequencing method involved in this invention is as follows: 1.1 Materials 1) Study subjects: This study included patients with alcoholic liver disease (ALD) and healthy volunteers. The inclusion criteria were: (1) age greater than 18 years; (2) healthy controls (HC): no history of drinking or serious illness; (3) patients with alcohol use disorder (AUD), alcoholic hepatitis (AH), or alcoholic cirrhosis (ALC) with a history of drinking for more than 5 years, with a daily ethanol intake of ≥40 grams for men and ≥20 grams for women. The exclusion criteria were: (1) co-infection with hepatitis A, B, C, D, or E virus or human immunodeficiency virus; (2) co-infection with non-alcoholic fatty liver disease, drug-induced liver injury, autoimmune liver disease, congenital liver disease, etc.; (3) presence of primary liver cancer or liver metastasis; (4) presence of serious organic diseases affecting other organs; (5) pregnancy or lactation; (6) history of antibiotic or probiotic use in the past 3 months. This study has been approved by the Ethics Review Committee of Union Hospital, Tongji Medical College, Huazhong University of Science and Technology and registered in the Chinese Clinical Trial Registry ((NCT05448144)). All study participants have signed informed consent forms.

[0037] 2) Sample collection: Stool samples were collected from the subjects upon admission, placed in sterile plastic cups, and stored at -80°C until extraction.

[0038] 3) Main reagents and instruments: FastPure Stool DNA Isolation Kit (MJYH, Shanghai, China); FastPfu Polymerase (Beijing TransGen Biotech Co., Ltd.); T100 Thermal Cycler (BIO-RAD, USA); JY600C bistable electrophoresis apparatus (Beijing Junyi Oriental Electrophoresis Equipment Co., Ltd.). High-throughput sequencing was performed by Shanghai Meiji Biopharmaceutical Technology Co., Ltd.

[0039] 4) Clinical indicator measurement: Basic characteristics such as gender, age, and weight were collected from healthy volunteers and patients. Blood routine, biochemical and imaging data were obtained by consulting clinical data.

[0040] 1.2 Methods

[0041] 1) Sample DNA extraction: Total genomic DNA of microbial communities was extracted from fecal samples from the disease group and the healthy group according to the instructions of the FastPure Stool DNAIsolation Kit (MJYH, shanghai, China). The integrity of the extracted genomic DNA was detected by 1% agarose gel electrophoresis, and the DNA concentration and purity were determined by NanoDrop2000 (ThermoScientific, USA).

[0042] 2) PCR amplification and sequencing library construction: Using the extracted DNA as a template, the full-length 16S rRNA gene of bacteria was amplified using primers 27F (5'-AGRGTTYGATYMTGGCTCAG-3') and 1492R (5'-RGYTACCTTGTTACGACTT-3') with barcodes, and the full-length ITS PCR amplification of fungi was performed using primers ITS1F (5'-CTTGGTCATTTAGAGGAAGTAA-3') and ITS4R (5'-TCCTCCGCTTATTGATATGC-3'). The PCR reaction system was as follows: 4 μL of 5×FastPfu buffer, 2 μL of 2.5 mM dNTPs, 0.8 μL of upstream primer (5 uM), 0.8 μL of downstream primer (5 uM), 0.4 μL of FastPfu polymerase, 0.2 μL of BSA, 10 ng of template DNA, and the volume was increased to 20 μL. Each sample was replicated in triplicate. The amplification program was as follows: pre-denaturation at 95℃ for 3 min, 27 cycles (denaturation at 95℃ for 30 s, annealing at 60℃ for 30 s, extension at 72℃ for 30 s), followed by a stable extension at 72℃ for 10 min, and finally storage at 4℃ (PCR instrument: T100 Thermal Cycler PCR, USA). After detection by 2% agarose gel electrophoresis, the product was purified by magnetic beads and quantified using a Qubit 4.0 (Thermo Fisher Scientific, USA). The products were then mixed in appropriate proportions according to the sequencing volume requirements for each sample.

[0043] Library construction was performed using the SMRTbell prep kit 3.0: (1) DNA damage repair; (2) end repair; (3) adapter ligation. Sequencing was performed using the PacBio Sequel IIe System (Shanghai Meiji Biotechnology Co., Ltd.). HiFi reads were generated from the sequenced subreads using the CCS mode of SMRT-Link v11.0 for subsequent data analysis.

[0044] 3) Sequencing Data Analysis: Data from each sample was differentiated based on the barcode sequence, and length filtering and orientation correction were performed, retaining sequences of 1000-1800 bp (bacteria) / 300-900 bp (fungi). Using the software platform Uparse (version 7.0.1090 http: / / drive5.com / uparse / ), OTU clustering was performed on sequences based on 97% similarity, and chimeras were removed. Sequences annotated to chloroplasts and mitochondria were removed from all samples (removal is recommended if chloroplast and mitochondrial contamination is present). To minimize the impact of sequencing depth on subsequent Alpha and Beta diversity data analysis, the sequence count of all samples was flattened to the minimum sample sequence count. After flattening, the average sequence coverage (Good's coverage) for each sample still reached 99.09%. Bacterial OTU taxonomic annotation was performed by comparing the Silva 16S rRNA gene database (v138) with the RDP classifier (http: / / rdp.cme.msu.edu / , version 2.11), and fungal OTU taxonomic annotation was performed by comparing the fungal database with the Unie (Release 8.0 http: / / unite.ut.ee / index.php) database. The confidence threshold was 70%, and the community composition of each sample was statistically analyzed at different species classification levels. Species abundance information of bacteria and fungi was extracted from the sequencing data, and the following indicators were calculated: (1) Relative abundance of a specific bacterial species (percentage of total bacterial community abundance) and (2) Relative abundance of a specific fungal species (percentage of total fungal community abundance).

[0045] 4) Statistical Analysis: Microbial diversity analysis was conducted on the MajorBio Cloud Platform (https: / / cloud.majorbio.com), specifically as follows: Mothur software (http: / / www.mothur.org / wiki / Calculators) was used to calculate the Alpha diversity knowledge Shannon index, etc.; PCoA analysis (principal coordinate analysis) based on the Bray-Curtis distance algorithm was used to examine the similarity of microbial community structure among samples; based on the taxonomic analysis results, R language was used to obtain the species composition of different groups (or samples) at various taxonomic levels (such as domain, kingdom, phylum, class, order, family, genus, species, OTU, etc.).

[0046] Stratified random sampling of research subjects was implemented using the `createDataPartition` function of the `caret` package in R (version 4.4.2), dividing the dataset into training and validation sets in a 7:3 ratio. Based on the relative abundance of bacteria or fungi in the training set, random forest analysis was performed using R (version 4.4.2) to screen for bacteria or fungi that play an important role in differential diagnosis. ROC curve analysis was used to continuously optimize the multi-marker combined diagnostic model for bacteria, fungi, and clinical indicators with excellent diagnostic efficacy. The results were then validated on the training set to ensure their reliability.

[0047] Example 1: Clinical and gut microbiota sequencing results of healthy controls and ALD patients

[0048] 1. Experimental Objective: To collect basic information, clinical indicators, and fecal samples from healthy controls (HC), alcohol-only individuals (AUD), AH patients, and ALC patients, and to analyze the differences in fecal microbiota between ALD patients and healthy controls through sequencing analysis, in order to clarify whether the intestinal bacteria and fungi in ALD patients have changed.

[0049] 2. Experimental methods: Basic information, blood routine, biochemical and other clinical indicators of hospitalized patients (including HC, AUD, AH and ALC) were prospectively collected.

[0050] Inclusion criteria were: (1) age greater than 18 years; (2) healthy controls: no history of drinking or serious illness; (3) drinkers who were single drinkers, patients with AH and ALC with a history of drinking for more than 5 years, with daily ethanol intake ≥40g for men and ≥20g for women.

[0051] Exclusion criteria are: (1) co-infection with hepatitis A, B, C, D, or E virus or human immunodeficiency virus; (2) co-infection with non-alcoholic fatty liver disease, drug-induced liver injury, autoimmune liver disease, congenital liver disease, etc.; (3) presence of primary liver cancer or liver metastasis; (4) presence of serious organic diseases affecting other organs; (5) being pregnant or lactating; (6) history of antibiotic or probiotic use in the past 3 months.

[0052] Stool samples were collected from the subjects upon admission, placed in sterile plastic cups, and stored at -80°C until extraction.

[0053] 3. Experimental Results: 3.1 Drinking alcohol can cause liver damage.

[0054] Table 1 shows the basic information and clinical data of the patients. Patients with AH and ALC had significantly higher levels of TB, ALT, AST, ALP, and GGT than patients with HC and AUD (Table 1). Patients with ALC were older than those with AH (58.2±11.3 vs 48.3±13.9, P=0.0063), had higher TB levels, and lower ALB levels, while patients with AH had higher ALT and AST levels, consistent with the clinical presentations of both AH and ALC patients (Table 2).

[0055] Table 1. Patient Basic Information and Clinical Data

[0056] Table 2. Post-hoc comparison results of AH and ALC

[0057] 3.2 Alcohol consumption leads to intestinal bacterial imbalance

[0058] Figure 1 To investigate the effects of alcohol consumption on human gut microbiota, this study systematically analyzed the impact of alcohol consumption on the gut microbiota using 16S rRNA gene sequencing technology. The results showed that alcohol intake significantly altered the structure and composition of the gut microbiota. Regarding α-diversity, compared with healthy controls, patients with ALD had significantly higher Shannon index (α-diversity). Figure 1 A) and the inverse Simpson exponent ( Figure 1 B) Both showed a decreasing trend, indicating that the richness and evenness of the gut microbiota significantly decreased with increasing disease severity. β-diversity analysis showed significant differences in gut microbiota composition between ALD patients of different severities and healthy controls. Figure 2 At the phylum level, the dominant bacterial groups included Bacillota, Pseudomonadota, Bacteroidota, Actinomycetota, and Verrucomicrobiota. The relative abundance of Bacillota and Bacteroidota decreased with disease progression, while Pseudomonadota showed an increasing trend. Figure 3 A). At the genus level, the number of potentially pathogenic bacteria such as *Escherichia* and *Klebsiella* increases with the progression of ALD, while Faecalibacterium When beneficial bacteria decrease ( Figure 3 B). Species-level analysis further confirmed this trend, such as... Escherichia coli and Klebsiella pneumoniae As the number of pathogenic bacteria increases with the progression of ALD, Faecalibacterium prausnitzii When beneficial bacteria decrease ( Figure 4These findings suggest that alcohol consumption leads to gut microbiota dysbiosis, characterized by reduced microbial diversity, an increase in potentially pathogenic bacteria, and a decrease in beneficial bacteria, playing a significant role in the development and progression of ALD.

[0059] 3.3 Alcohol consumption leads to intestinal fungal dysbiosis

[0060] Figure 2 To investigate the effects of alcohol consumption on human gut fungi, this study systematically analyzed the impact of alcohol consumption on the human gut fungal community using ITS sequencing technology. The results showed that, although there were no significant differences in gut fungal α-diversity (including Shannon and Simpson indices) between ALD patients of different severities and HC patients, Figure 5 AB), but β-diversity analysis showed a clear separation of fungal community structure among the groups ( Figure 6 This suggests that alcohol consumption significantly alters the composition of gut fungi. At the phylum level, the gut fungal community is mainly composed of Ascomycota, Basidiomycota, unclassified_k__Fungi, and Fungi_phy_Incertae_sedis. Figure 7 A). Notably, the fungal community exhibits significant dynamic changes as alcoholic liver disease (ALD) progresses: Candida The relative abundance of the genera progressively increases, while Cutaneotrichosporon The genus gradually decreases ( Figure 7 B). At the microbial level, this trend is even more pronounced, particularly among opportunistic pathogens. Candida albicans It was significantly enriched in patients with AH and ALC, while potentially beneficial bacteria Cutaneotrichosporon curvatum Then significantly reduced ( Figure 8 These findings indicate that alcohol consumption specifically alters the composition of the gut fungal community, manifesting as an increase in opportunistic pathogens and a decrease in potentially beneficial fungi. This change in fungal community structure may be closely related to the development of alcoholic liver disease, providing a new perspective for a deeper understanding of the impact of alcohol on the gut microbiota.

[0061] 4. Experimental Conclusions

[0062] This study found significant differences in liver damage caused by alcohol consumption across different disease stages. Patients with acute hepatocellular injury (AH) primarily exhibited significantly elevated ALT and AST levels, indicating acute hepatocellular injury; while patients with acute liver collapse (ALC) showed more severe liver decompensation, characterized by elevated TB and decreased ALB levels, and were significantly older than the AH group. Regarding gut microbiota, the study observed a close correlation between alcohol-induced bacterial-fungal dysbiosis and ALD progression. The AH stage already showed significant gut microbiota imbalance, characterized by decreased bacterial diversity, increased pathogenic bacteria, and reduced beneficial bacteria, accompanied by changes in fungal community structure, including increased opportunistic pathogens and reduced beneficial fungi. This microbiota imbalance further worsened as the disease progressed to the ALC stage. These findings provide important evidence for a deeper understanding of the pathogenesis of ALD and the development of targeted intervention strategies.

[0063] Example 2: Establishment of the AH vs. ALC Discrimination Model

[0064] 1. Data Processing

[0065] This study included 159 participants. Stratified random sampling was implemented using the `createDataPartition` function of the `caret` package in R. The 159 participants (including 26 HC, 37 AUD, 56 AH, and 40 ALC) were divided into a training set (n=113) and a validation set (n=46) in a 7:3 ratio. The training set contained 19 HC, 26 AUD, 40 AH, and 28 ALC, while the validation set contained 7 HC, 11 AUD, 16 AH, and 12 ALC. The proportions of each group in the training and validation sets were balanced (HC 73.1% / 26.9%, AUD 70.3% / 29.7%, AH 71.4% / 28.6%, ALC 70.0% / 30.0%). All subjects underwent 16S rRNA gene sequencing and ITS sequencing. The data underwent rigorous quality control to ensure that the validation set had statistical power while maintaining a sufficient training sample size, thus providing a reliable data foundation for subsequent microbial community analysis and machine learning modeling.

[0066] 2. Differential Diagnostic Model for AH and ALC Intestinal Bacteria

[0067] 1) In the training set, we evaluated the importance of gut microbiota features using the random forest algorithm. Based on MeanDecrease Gini index analysis, we screened out the 10 key bacteria with the most discriminative power for AH and ALC classification: s__ Lactiplantibacillus plantarum , s unclassified g Paenibacillus , s__ Mediterraneibacter glycyrrhizinilyticus , s Rosenbergiella nectarea , s__unclassified g Aeromonas , s Paenibacillus sp. 6083 , s Lacrimispora xylanolytica , s Paenibacillus sp. FSL H8-457 , s Levilactobacillus brevis , s__ Lacticaseibacillus paracasei ( Figure 9 This provides a new perspective for understanding the bacterial characteristics of AH and ALC, and also provides a theoretical basis for developing diagnostic biomarkers and intervention targets based on specific bacteria.

[0068] 2) Based on the results of random forest feature importance analysis, we further evaluated the diagnostic efficacy of the top five key bacteria. Receiver operating characteristic (ROC) curve analysis showed that these bacteria exhibited excellent differential diagnostic value for ALC: s Lactiplantibacillus plantarum (AUC=0.891, 95%CI: 0.798-0.984), s unclassified g Paenibacillus (AUC=0.911, 95%CI: 0.838-0.983), s__ Mediterraneibacter glycyrrhizinilyticus (AUC=0.918, 95%CI: 0.836-1), s__ Rosenbergiella nectarea (AUC=0.893, 95%CI: 0.815-0.97), s__unclassified_g__ Aeromonas( AUC = 0.89, 95% CI: 0.811–0.97). It is noteworthy that in single bacteria… s__ Mediterraneibacter_glycyrrhizinilyticus It exhibited the best diagnostic performance (AUC=0.918), with a sensitivity of 0.821 and a specificity of 0.95. Figure 11 A, Table 3). These results suggest that specific gut bacteria not only have important pathophysiological significance, but may also serve as non-invasive biomarkers for the auxiliary diagnosis and staging of alcoholic liver disease.

[0069] Table 3. Characteristics of the differential diagnostic model of AH and ALC intestinal bacteria

[0070] 3) Based on ROC curve analysis, we optimized a multi-bacterial biomarker combined diagnostic model for bacterial species with excellent diagnostic efficacy. The study found that, based on s... __Lactiplantibacillus_plantarum , s__unclassified_ g__Paenibacillus and s__Mediterraneibacter_glycyrrhizinilyticus The combined diagnostic model consisting of three bacteria demonstrated optimal disease prediction capability. Figure 11 A). This model integrates... s__ Lactiplantibacillus_plantarum (Optimal threshold 0.1806%)s__unclassified_g__ Paenibacillus (Optimal threshold 0.0018%) and s__Mediterraneibacter_glycyrrhizinilyticus Abundance characteristics (optimal threshold 0.0185%) were obtained through logistic regression, using the formula P(ALC) = 1 / (1+exp(-(-1.727 -8019.461)). s__Lactiplantibacillus_plantarum +265019.739 s__unclassified_ g__Paenibacillus -8.984 s__Mediterraneibacter_glycyrrhizinilyticus) The probability of ALC was calculated. When P(ALC) > 0.575, the patient can be diagnosed with ALC. The model showed excellent diagnostic efficacy (AUC = 0.932, sensitivity 82.1%, specificity 100%, positive predictive value 100%, negative predictive value 88.9% (Table 3)).

[0071] 4) We validated the established multi-bacterial biomarker combined diagnostic model on an independent validation set (16 AH patients and 12 ALC patients). The validation results showed that the combined diagnostic model based on three characteristic bacterial biomarkers exhibited excellent discriminative performance on the independent validation set. ROC curve analysis showed that the model achieved an AUC of 1.00 on the validation set. Figure 11 B). Confusion matrix analysis showed that the model's predicted classification for all 28 samples was completely consistent with the actual clinical diagnosis (all 16 cases in the AH group were correctly classified, and all 12 cases in the ALC group were correctly identified), achieving an accuracy of 100% (95% confidence interval: 0.8766-1.0000, P=1.567×10⁻⁶). -7 All evaluation indicators of the model reached ideal levels: sensitivity, specificity, positive predictive value, and negative predictive value were all 100%, and the Kappa coefficient was 1, indicating that the model's prediction results had perfect consistency with the actual diagnosis. Based on an ALC prevalence of 42.86%, the model demonstrated excellent clinical recognition ability, with a balanced accuracy of 100%. These results fully confirm that this multi-bacterial biomarker combined diagnostic model can provide a reliable basis for the clinical differential diagnosis of ALD.

[0072] It is worth noting that although bacterial markers have shown good diagnostic value, considering the complex interactions between bacteria and fungi in the gut microbiota, we further explored the changing characteristics of fungal communities and their diagnostic potential.

[0073] 3. Differential Diagnostic Models for AH and ALC Intestinal Fungi

[0074] 1) In the training set, we evaluated the importance of gut microbiota features using the random forest algorithm. Based on MeanDecrease Gini index analysis, we selected the 10 key fungi with the most discriminative power for AH and ALC classification: s__ Zygosaccharomyces_pseudobailii , s__unclassified_k__Fungi , s__Aspergillus_ cibarius , s__Wallemia_sp , s__Moesziomyces_sp , s__Wickerhamomyces_ subpelliculosus , s__Xeromyces_bisporus , s__Penicillium_citrinum , s__ Saccharomyces_sp , s__Aspergillus_penicillioides ( Figure 10 This provides a new perspective for understanding the fungal characteristics of AH and ALC, and also provides a theoretical basis for developing diagnostic biomarkers and intervention targets based on specific fungi.

[0075] 2) Based on the results of random forest feature importance analysis, we further evaluated the diagnostic efficacy of the top five key fungi. ROC curve analysis showed that these fungi exhibited excellent differential diagnostic value for ALC: s__ Zygosaccharomyces_pseudobailii (AUC=0.77, 95%CI: 0.672-0.867), s__unclassified_ k__Fungi (AUC=0.692, 95%CI: 0.563-0.822), s__Wallemia_sp (AUC=0.761, 95%CI:0.663-0.858), s__Aspergillus_cibarius (AUC=0.669,95%CI:0.521-0.818), s__ Moesziomyces_sp (AUC=0.72, 95%CI: 0.623-0.816). It is noteworthy that in single fungi... s__ Zygosaccharomyces_pseudobailii It exhibited the best diagnostic performance (AUC=0.77), with a sensitivity of 0.571 and a specificity of 0.975. Figure 12 A (Table 4); the AUC values ​​of the remaining fungal markers ranged from 0.669 to 0.761, slightly lower than the diagnostic efficacy of bacterial markers, but suggesting a significant association between fungal community changes and ALC. These findings not only expand our understanding of ALD-related gut microbiota dysbiosis but also provide a new direction for developing a diagnostic system based on fungal-bacterial combined markers. Further research can focus on exploring... s__Zygosaccharomyces_pseudobailii The synergistic diagnostic value of characteristic fungi and identified bacterial markers.

[0076] 3) Based on the preliminary screening results, we further optimized and constructed a combined diagnostic model for three fungi. This model integrates... s__Zygosaccharomyces_pseudobailii (Optimal threshold 0.0019%) s__unclassified_k__Fungi (Optimal threshold < 0.0315%) and s__Wallemia_sp Abundance characteristics (optimal threshold 0.0019%) were obtained through logistic regression equation P(ALC) = 1 / (1 + exp(-(0.417 + 4724.228)). s__Zygosaccharomyces_pseudobailii -10.134 s__unclassified_k__Fungi -19245.368 s __Wallemia_sp The model calculates the probability of ALC (Alzheimer's disease), defining a patient as having ALC when P(ALC) > 0.429. This model demonstrates excellent diagnostic efficacy (AUC = 0.905), with significantly higher sensitivity (92.9%) and specificity (75.0%) than single fungal markers. The positive predictive value is 72.2%, and the negative predictive value is 93.8%. Figure 12 A, Table 4).

[0077] 4) We validated the established multi-fungal biomarker combined diagnostic model on an independent validation set (16 AH patients and 12 ALC patients). The validation results showed that the combined diagnostic model based on three characteristic fungal biomarkers exhibited good discriminative performance on the independent validation set (accuracy 75.0%, 95% CI: 0.5513-0.8931). The model correctly predicted 13 AH patients as AH, 8 ALC patients as ALC, 4 ALC patients as AH, and 3 AH patients as ALC, with an AUC of 0.701, sensitivity and specificity of 66.7% and 81.2%, respectively. Analysis showed 7 misclassifications (4 ALC cases misclassified as AH, 3 AH cases misclassified as ALC), but the McNemar test confirmed that the misclassifications had no directional bias (P=1.000). Figure 12 B, Table 4).

[0078] Table 4. Characteristics of the differential diagnostic models of AH and ALC intestinal fungi

[0079] 5) It is worth noting that although the diagnostic efficacy of this fungal combination (AUC=0.905) was slightly lower than that of the previously established bacterial biomarker model (AUC=0.932), the two have complementary value: 1) the fungal model has higher sensitivity for ALC identification (92.9% vs 82.1%); 2) the bacterial model has better specificity (100% vs 75%). These findings suggest that changes in the gut fungal community and bacterial dysbiosis jointly participate in the pathogenesis of ALD. The combined application of characteristic fungal biomarkers and bacterial biomarkers may further improve diagnostic efficacy, providing an important basis for developing a precision diagnostic system for ALD based on "bacterial-fungal" multidimensional biomarkers. Further research can further optimize the combination strategy and verify its clinical application value.

[0080] 4. Differential Diagnostic Model of Intestinal Bacteria and Fungi Combined with AH and ALC

[0081] 1) Next, by integrating dominant bacterial and fungal biomarkers, an innovative multi-kingdom microbial joint diagnostic model was established. During the training set construction process, we used machine learning methods to screen for the most diagnostically valuable combinations of microbial biomarkers, including three types of bacteria (…). s__Lactiplantibacillus_plantarum , s__unclassified_g__ Paenibacillus , Mediterraneibacter glycyrrhizinilyticus ) and two fungi ( s__ Zygosaccharomyces bailii , unclassified k__ Fungi The diagnostic equation established through logistic regression analysis is: P(ALC) = 1 / (1 + exp(-(-1.268-7911.395)). Lactiplantibacillus Lactiplantibacillus plantarum +6825460.118 unclassified g__ Paenibacillus +2.341 s__ Mediterraneibacter glycyrrhizinilyticus -14605.089 Zygosaccharomyces Zygosaccharomyces bailii -15.853 unclassified k__ Fungi The variables correspond to the standardized abundance of the aforementioned microbial markers. Specifically, when the calculated P(ALC) > 0.224 based on the combined abundance of the three fungi, the patient is diagnosed with ALC. This model demonstrated excellent diagnostic performance on the training set, with an AUC curve area of ​​0.954 (95% CI: 0.905-1), sensitivity of 85.7%, and specificity of 100%. The positive predictive value (100%) and negative predictive value (90.9%) were both at ideal levels. Figure 13 A, Table 5).

[0082] Table 5. Characteristics of the differential diagnostic model of AH and ALC intestinal bacteria combined with fungi

[0083] 2) In the independent validation set (16 cases of AH and 12 cases of ALC), the combined model demonstrated excellent generalization ability and clinical applicability. All 15 cases of AH were correctly predicted (high specificity), and 11 cases of ALC were correctly predicted (high sensitivity), with an overall diagnostic accuracy of 92.86% (26 / 28), and a 95% confidence interval of 0.765-0.9912, significantly higher than random guessing. Only one case of ALC was misdiagnosed as AH (extremely low false negative rate), and only one case of AH was misdiagnosed as ALC (low false positive rate). McNemar's test showed no statistically significant misdiagnosis (P=1). ROC curve analysis showed that the validation set AUC was as high as 0.974 (95% CI: 0.926-1.000), significantly better than a single microbial model. Figure 13 B, Table 5).

[0084] 3) Comparative analysis revealed that the diagnostic efficacy of this multi-kingdom microbial model was significantly superior to that of single bacterial or fungal models. The combination of microbial biomarkers in the model reflected the characteristic gut microbiota dysbiosis pattern of ALD: reduced probiotics (e.g., Lactiplantibacillus plantarum An increase in opportunistic pathogens (such as...) unclassified g__ Paenibacillus ) and characteristic changes in fungal communities (such as Zygosaccharomyces bailii (Additional information). These results not only confirm that integrating bacterial and fungal biomarkers can construct a more precise ALD diagnostic system, but also provide a new perspective for understanding the role of multi-kingdom gut microbiota interactions in liver disease progression. This model provides an innovative microbiome solution for clinical practice.

[0085] 5. Age-based differential diagnostic model for patients with AH and ALC based on combined gut bacteria and fungi.

[0086] 1) In the analysis of basic patient information, we found that ALC patients were older than AH patients (58.2±11.3 vs 48.3±13.9, P=0.0063) (Table 1, Table 2). Therefore, based on the previously established microbial biomarker model, we further integrated age as an important clinical variable to construct an innovative "microbial-age" multi-parameter diagnostic model. The model was established through logistic regression analysis, with the formula: P(ALC)=1 / (1+exp(-(-9.234-14478.4)). s__ Lactiplantibacillus plantarum +3977114.078 unclassified g__ Paenibacillus +61.995 Mediterraneibacter glycyrrhizinilyticus-8323.763 s__ Zygosaccharomyces bailii -16.92 unclassified k__ Fungi +0.133 Age ))).

[0087] The model demonstrated near-perfect diagnostic performance on the training set (AUC=0.985, 95%CI: 0.965-1.000), with a sensitivity of 96.4%, specificity of 97.1%, and positive and negative predictive values ​​both exceeding 95%. Specifically, when the abundance of three bacteria, two fungi, and the patient's age were combined to calculate a P(ALC) greater than 0.369, the patient was diagnosed with ALC. Figure 14 A, Table 6).

[0088] Table 6. Characteristics of age-based differential diagnostic models for patients with combined gut bacteria and fungi in AH and ALC

[0089] 2) In the independent validation set (16 cases of AH and 12 cases of ALC), this multiparameter model demonstrated excellent clinical applicability. Figure 14 (B) All 16 cases of AH were correctly predicted (high specificity), 10 cases of ALC were correctly predicted (high sensitivity), only 2 cases of ALC were misdiagnosed as AH (extremely low false negative rate), and only 0 cases of AH were misdiagnosed as ALC (low false positive rate). ROC curve analysis showed that the validation set AUC reached 0.948 (95% CI: 0.881-1.000), with sensitivity and specificity of 0.833 and 1, respectively. The Kappa coefficient was 0.8511, indicating that the model prediction was highly consistent with the actual diagnosis.

[0090] 3) Notably, this multimodal joint diagnostic model exhibits significant performance improvements: in the training set, the model's AUC reached 0.985 (95% CI: 0.965-1.000), an increase of 3.1 percentage points compared to the simple microbial model (AUC=0.954), indicating that integrating age parameters significantly enhances the model's discriminative ability. Simultaneously, the model achieves excellent sensitivity (96.4%) and specificity (97.1%), realizing an ideal balance between high disease detection rate and low misdiagnosis rate. This improvement in diagnostic performance is mainly reflected in three aspects: 1) improved identification rate of early ALC cases (sensitivity increased by 10.7 percentage points); 2) more accurate exclusion of non-ALC cases (specificity maintained at a high level of 97.1%); 3) enhanced overall diagnostic confidence (both positive and negative predictive values ​​exceeded 95%). These advantages make this model particularly suitable for clinical applications, effectively identifying potential ALC patients while minimizing overdiagnosis of AH. The overall improvement in model performance confirms the value of the strategy of integrating clinical parameters with microbiome biomarkers, and provides new ideas for developing more accurate ALD diagnostic tools.

[0091] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. Biomarker for discriminating alcoholic hepatitis from alcoholic cirrhosis, characterized in that, The biomarkers include Lactobacillus plantarum s Lactiplantibacillus plantarum , Paenibacillus sp. s__ unclassified_g Paenibacillus , Ammonifex degensii s Mediterraneibacter s Mediterraneibacter glycyrrhizinilyticus , Starmerella bombicola s Zygosaccharomyces pseudobailii , and Unclassified fungi s unclassified_k Fungi .

2. The biomarker of claim 1, wherein, The biomarkers also include age data.

3. Use of a detection reagent for the biomarker of claim 1 in the manufacture of a prediction product, characterized in that, The prediction product is used for identifying alcoholic hepatitis and alcoholic cirrhosis.

4. Use according to claim 3, characterized in that, The detection reagent is an abundance detection reagent, including 16S rRNA gene V3-V4 variable region specific amplification primers for detecting bacteria, and ITS region specific amplification primers for detecting fungi.

5. Use according to claim 4, characterized in that, The nucleotide sequence of the primer for detecting bacteria is: upstream primer: 5'-AGRGTTYGATYMTGGCTCAG-3', downstream primer: 5'-RGYTACCTTGTTACGACTT-3'; And / or, the nucleotide sequence of the primer for detecting fungi is: upstream primer: 5'-CTTGGTCATTTAGAGGAAGTAA-3'; downstream primer: 5'-TCCTCCGCTTATTGATATGC-3'.

6. Use according to claim 3, characterized in that, The prediction product is a kit, a detection system or an instrument.

7. A kit for discriminating alcoholic hepatitis from alcoholic cirrhosis, characterized by, The kit comprises a microbial abundance detection reagent, a prediction formula and a prediction standard; the microbial abundance detection reagent is used for detecting the abundance of the biomarker in claim 1; the prediction formula is: P (ALC) = 1 / (1 + exp (- (-9.234-14478.4 s__ Lactiplantibacillus plantarum +3977114.078 s unclassified_g Paenibacillus +61.995 s Mediterraneibacter glycyrrhizinilyticus -8323.763 s__ Zygosaccharomyces pseudobailii -16.92 s unclassified_k Fungi +0.133 Age )), wherein Age represents the age of the patient; and the prediction standard is that when P (ALC) is greater than 0.369, the patient is determined to be an alcoholic cirrhosis patient.

8. The kit of claim 7, wherein The kit also includes reagents for extracting the biomarkers from the sample.

9. A device for discriminating alcoholic hepatitis from alcoholic cirrhosis, characterized by, It comprises a detection unit and a data analysis unit, wherein: The detection unit is used to obtain a sample, detect and determine the microbial abundance detection result of the sample; the microorganism is the biomarker of claim 1; The data analysis unit is used to analyze the detection result of the detection unit combined with case information to make an identification result.

10. The apparatus of claim 9, wherein, The detection unit detects and obtains a detection result according to the following operations: Detecting a fecal sample to obtain DNA data; The data analysis unit is used to analyze the detection result of the detection unit, which includes: sequencing the DNA data of the fecal sample to obtain intestinal microbial genomic data; Analyzing the intestinal microbial genomic data to obtain relative abundance information of each biomarker; According to the relative abundance information, intestinal flora characteristic data is determined.

Images