Biomarker and kit for HIV / Mtb co-infection immune state evaluation and clinical diagnosis and application of biomarker and kit

Abstract

The invention relates to the field of medical diagnosis, and particularly discloses a biomarker combination and a kit for HIV / Mtb co-infection immune state evaluation and clinical diagnosis and application of the biomarker combination and the kit. Multi-factor detection is carried out on peripheral blood samples of an HIV infected person, an Mtb infected person and an HIV / Mtb co-infected person, Lasso regression screening is adopted to obtain cell factors closely related to HIV / Mtb co-infection, such as FGF-2, IL-7, I-309, TNF-alpha, MMP-1 and MMP-7, and a multi-factor diagnosis model is constructed. The model shows high sensitivity, specificity and AUC value in a training set and a test set, can effectively distinguish HIV individual infection and HIV / Mtb co-infection individuals, and is especially suitable for diagnosis of patients with low CD4 + T cell level. The marker combination provided by the invention and the kit prepared from the marker combination can be used as a supplement of an immunological detection method, and a new technical means is provided for clinical diagnosis and immune state evaluation of HIV / Mtb co-infection.

Description

Technical Field

[0001] This invention relates to the field of medical diagnostics, specifically to a biomarker, kit, and its application for assessing the immune status and clinical diagnosis of HIV / Mtb co-infection. Background Technology

[0002] Human Immunodeficiency Virus (HIV) combined with Mycobacterium tuberculosis ( Mycobacterium tuberculosis, Mtb infection is a priority for tuberculosis outbreak prevention and control. HIV-infected individuals are approximately 14 times more likely to develop tuberculosis than uninfected individuals. According to the 2024 Global Tuberculosis Report, there were 10.8 million new tuberculosis cases globally in 2023, of whom 662,000 were co-infected with HIV. Tuberculosis causes 1.25 million deaths, including 161,000 HIV-positive individuals, accounting for 12.9% of all tuberculosis-related deaths. Therefore, timely, accurate, and feasible screening and diagnostic strategies can ensure timely intervention in individuals co-infected with HIV / Mtb and effectively reduce hospitalization and mortality rates.

[0003] The currently accepted gold standard for diagnosing tuberculosis remains a positive Mtb culture or a histopathological biopsy consistent with Mtb infection pathology. However, due to the complexity of the condition in HIV / Mtb co-infected individuals and the difficulty in obtaining biopsy specimens, traditional detection methods may lead to missed or misdiagnosed cases, delaying diagnosis and treatment. Immunological detection techniques (including the tuberculin skin test (TST), T-SPOT.TB, QFT-GIT, etc.) have developed rapidly in recent years and have been widely used in the clinical diagnosis of Mtb infection. However, in HIV-infected individuals, the positive rate of TST for tuberculosis detection is only 19.1%, and the sensitivity is only 35.96%, and with the decline of CD4+... + A decrease in T lymphocyte count gradually reduces the positivity rate. Similarly, QFT-GIT has low sensitivity in diagnosing HIV / Mtb co-infection, and the test results may be affected by CD4 count. +T-cell count is a significant factor, therefore, newer strategies are needed to improve the diagnostic efficacy of this test. T-SPOT.TB has higher sensitivity than TST in detecting HIV / Mtb co-infection, but it cannot effectively distinguish between LTBI and ATB and is less effective in predicting ATB pathogenesis. Emerging rapid and immediate molecular tests (including Xpert MTB / RIF assays and linear probe assays) can efficiently identify MTB genes and detect drug resistance mutations; however, their high cost limits their application in resource-constrained areas. Cost reduction may be the primary issue that needs to be addressed for large-scale application in the future. Summary of the Invention

[0004] The purpose of this invention is to overcome the shortcomings of existing HIV / Mtb co-infection diagnostic techniques, such as insufficient sensitivity, low specificity, and limitations in CD4 detection. + To address the shortcomings of poor diagnostic accuracy in patients with low T-cell levels, this paper proposes a new approach for immune status assessment and clinical diagnosis based on a combination of multifactor biomarkers.

[0005] To achieve the above objectives, this invention proposes a combination of biomarkers, a reagent kit, and their applications for assessing the immune status and clinical diagnosis of HIV / Mtb co-infection. The biomarker combination includes at least one or more of the following: fibroblast growth factor-2 (FGF-2), interleukin-7 (IL-7), chemokine I-309, tumor necrosis factor-α (TNF-α), matrix metalloproteinase-1 (MMP-1), and matrix metalloproteinase-7 (MMP-7). This invention utilizes Luminex multifactor detection on peripheral blood samples, combined with Lasso regression model for feature selection, and employs machine learning methods such as XGboost to construct a diagnostic model, thereby achieving effective differentiation between HIV-only infection, Mtb-only infection, and HIV / Mtb co-infection. Specifically: In a first aspect, the present invention provides a combination of biomarkers for assessing the immune status and clinical diagnosis of HIV / Mtb co-infection, said biomarkers comprising at least one or more of the following: fibroblast growth factor-2 (FGF-2), interleukin-7 (IL-7), chemokine I-309, tumor necrosis factor-α (TNF-α), matrix metalloproteinase-1 (MMP-1), and matrix metalloproteinase-7 (MMP-7).

[0006] In one embodiment, the combination simultaneously comprises FGF-2, IL-7, I-309, and TNF-α.

[0007] In one embodiment, it further includes MMP-1 and / or MMP-7.

[0008] A second aspect of the present invention provides an in vitro diagnostic kit for detecting HIV / Mtb co-infection, the kit comprising diagnostic reagents for detecting the biomarker.

[0009] In one embodiment, the detection reagent is selected from ELISA reagents, immunochromatographic reagents, flow cytometry reagents, or Luminex multifactor detection reagents.

[0010] A third aspect of the invention provides the use of the above-described biomarker combination in the preparation of products for assessing the immune status of HIV / Mtb co-infection.

[0011] A fourth aspect of the present invention provides the use of the above-described biomarker combination in the preparation of clinical diagnostic products for HIV / Mtb co-infection.

[0012] A fifth aspect of the invention provides the above-described biomarker combination in the preparation of a product for low CD4+ + Application of T-cell count (≤200 cells / µL) in HIV / Mtb co-infection diagnostic products in HIV-infected individuals.

[0013] In one embodiment, the product is a reagent kit.

[0014] Compared with the prior art, the present invention has the following beneficial effects: 1. Multifactorial combined detection significantly improves the diagnostic sensitivity and specificity for HIV / Mtb co-infection, and has superior diagnostic performance compared to traditional TST, QFT-GIT and T-SPOT.TB methods; 2. The XGboost multi-factor diagnostic model established in this invention has an AUC greater than 0.80 in both the training and test sets, demonstrating stable and reliable diagnostic capabilities. 3. Especially in CD4 + In patients with T cell counts ≤200 cells / µL, the multifactor model of this invention has significantly better diagnostic efficacy than the IGRA detection method, solving the problem of difficult diagnosis in patients with low immune status. 4. The biomarker combination of the present invention can be widely used in the development of in vitro diagnostic kits, and has strong scalability and clinical application prospects.

[0015] Therefore, this invention not only provides a new technical approach for the clinical diagnosis of HIV / Mtb co-infection, but also lays a solid foundation for the improvement and promotion of immunological detection methods, and has important public health value and industrial application potential. Attached Figure Description

[0016] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings: Figure 1 This is a plot of variable selection based on a Lasso regression model, where Figure 1 A: The association between the logarithm (λ) of the variables included in the Lasso analysis and the regression coefficients; Figure 1 B: The process of selecting the optimal λ value in a Lasso regression model using the 10-fold cross-validation method; Figure 2 ROC plot of cytokines for the diagnosis of HIV / Mtb co-infection; where Figure 2 A: ROC plot of FGF-2 for the diagnosis of HIV / Mtb co-infection; Figure 2 B: ROC plot of IL-7 for the diagnosis of HIV / Mtb co-infection; Figure 2 C: ROC plot of I-309 for the diagnosis of HIV / Mtb co-infection; Figure 2 D: ROC plot of TNF-α for the diagnosis of HIV / Mtb co-infection; Figure 2 E: ROC plot of MMP-1 for the diagnosis of HIV / Mtb co-infection; Figure 2 F: ROC plot of MMP-7 for the diagnosis of HIV / Mtb co-infection; Figure 3 To evaluate the performance of machine learning models on the training and test sets; among which Figure 3 A: Accuracy, sensitivity, specificity, precision, and F1 score of the six machine learning models in the training set data; Figure 3 B: Accuracy, sensitivity, specificity, precision, and F1 score of the 6 machine learning models in the test set data; Figure 4 The ROC curves of the diagnostic model in the training and validation sets; Figure 5 Rank the variables of the XGboost model by importance; Figure 6 For multifactor models and IGRA in CD4 + ROC curves for patients with T cells ≤200 cells / µL; among which Figure 6 A: Multifactor model in CD4 + ROC curves and AUC in patients with T cells ≤200 / µL; Figure 6 B: IGRA in CD4 + ROC curves and AUCs in patients with ≤200 T cells / µL. Detailed Implementation

[0017] The preferred embodiments of the present invention will be described below in conjunction with the accompanying drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention and are not used to limit the present invention.

[0018] Example 1 Assessment of the immune status of HIV / Mtb co-infection and screening of clinical diagnostic markers 1. Research subjects This study was approved by the Ethics Committee of Beijing You'an Hospital, Capital Medical University (Ethical Approval No.: Beijing You'an Ke Lun Zi

[2023] No. 020). HIV-infected individuals, HIV / Mtb co-infected individuals who visited the Department of Infection and Immunology of Beijing You'an Hospital, Capital Medical University from January 2023 to June 2024, Mtb-infected individuals who visited the Department of Respiratory and Critical Care Medicine, and healthy individuals of matching age and gender who underwent physical examinations at Beijing You'an Hospital, Capital Medical University during the same period were recruited. Subjects who met the inclusion and exclusion criteria were included in this study. Finally, a total of 90 HIV-infected individuals and 81 HIV / Mtb co-infected individuals were included. The subjects were randomly divided into a training set and a test set at a ratio of 7:3. The training set was used to construct the Lasso regression model and subsequent diagnostic models, and the test set was used to test the accuracy of the constructed models.

[0019] After fully informing the research procedures and the risks and benefits to the subjects, the researchers obtained the informed consent forms from the subjects and collected relevant clinical data (including age, gender, HIV diagnosis time, HIV viral load, HBV, HCV and syphilis co-infection status, blood routine, liver function, ART initiation time and medication status, anti-tuberculosis treatment status, etc.) and collected peripheral blood specimens. The specific inclusion and exclusion criteria are as follows: Inclusion criteria: (1) Age 18 years old ≤ age ≤ 65 years old, both men and women are eligible; (2) The diagnosis of HIV infection meets any one of the following conditions: ① Positive results of HIV-1 / 2 antibody screening test and HIV supplementary tests (including antibody confirmation test and nucleic acid test); ② Epidemiological history, both nucleic acid test results are positive; ③ Positive results of HIV isolation test.

[0020] (3) The enrolled HIV-infected individuals simultaneously have respiratory or systemic symptoms similar to Mtb infection (including symptoms such as fever, cough, expectoration, etc.), and Mtb infection is excluded (IGRA negative); (4) Diagnosis of AIDS stage: Meeting the HIV infection diagnosis, and CD4 +T-cell count less than 200 / µL. Or meeting the diagnostic criteria for HIV infection plus any of the following: ① Unexplained persistent irregular fever above 38°C for more than 1 month; ② Diarrhea for more than 1 month; ③ Weight loss of more than 10% within 6 months; ④ Recurrent oral fungal infections, herpes simplex virus infections, or herpes zoster virus infections; ⑤ Opportunistic infections such as Pneumocystis pneumonia, ATB, or nontuberculous mycobacterial infections; ⑥ Recurrent bacterial pneumonia; ⑦ Deep fungal infections; ⑧ Space-occupying lesions of the central nervous system; ⑨ Dementia in young or middle-aged adults; ⑩ Active cytomegalovirus infection; Toxoplasmosis encephalopathy; Marneffei basket disease; Recurrent sepsis; Kaposi's sarcoma, lymphoma.

[0021] (5) The diagnosis of latent tuberculosis is mainly based on clinical manifestations and IGRA results. Patients who do not have tuberculosis-related clinical manifestations (fever, cough, hemoptysis, night sweats, weight loss, chest pain, fatigue, dyspnea) and whose IGRA results are positive are diagnosed with latent tuberculosis.

[0022] (6) ATB is mainly judged based on clinical manifestations, imaging examination results and etiological test results. The diagnosis must meet at least one of the following conditions: ① Etiological evidence: positive acid-fast staining of sputum; positive culture of Mycobacterium tuberculosis in sputum, bronchoalveolar lavage fluid and pleural effusion; positive Mtb-DNA test in sputum, bronchoalveolar lavage fluid and pleural effusion; positive Mycobacterium tuberculosis test / rifampicin resistance test (Xpert MTB / RIF); ② Pathological evidence: tuberculous granuloma or caseous necrosis found in biopsy tissue, and positive acid-fast staining; (7) The diagnosis of HIV / Mtb co-infection mainly includes the following two situations: ① Diagnosis of AIDS patients with ATB: meeting the above diagnostic criteria for AIDS and ATB, and meeting the diagnosis of ATB; ② HIV infection with latent tuberculosis infection: meeting the diagnostic criteria for HIV infection, without tuberculosis-related clinical manifestations, and with a positive IGRA result.

[0023] (8) Healthy controls: ① Negative HIV-1 / 2 antibody screening test; ② No clinical manifestations of Mtb infection, negative IGRA results; (9) Anti-tuberculosis treatment has not been initiated or the duration of anti-tuberculosis treatment is less than 1 month; (10) Fully understand the purpose and requirements of this experiment and voluntarily sign a written informed consent form.

[0024] Exclusion criteria: (1) The researchers judged that the individual was unsuitable to participate in this study if he or she had any unstable or serious cardiovascular, renal, hepatic, blood, tumor, endocrine, metabolic, mental or rheumatic diseases.

[0025] 2. Peripheral blood collection and plasma separation 20 mL of peripheral venous blood was collected from the enrolled subjects according to standard methods in EDTA anticoagulant vacuum blood collection tubes without separating gel. After blood collection, the blood collection tubes were placed in a 4 ℃ refrigerator for short-term storage, and the peripheral blood was separated and stored within 4 hours.

[0026] 3. Luminex multifactor detection (1) Equilibrate all reagents to room temperature (20 ~ 25℃) before testing; (2) Plasma sample preparation: Take out the frozen plasma sample from the -80 ℃ freezer. After the sample is completely thawed, vortex it on a vortex mixer. Then place the sample in a micro-centrifuge and centrifuge at 13000×g / min for 10min. Carefully aspirate 50 µL of the supernatant into a new sterile EP tube, avoiding aspirating particulate matter or the lower lipid layer. (3) Preparation of pre-coupled antibody magnetic beads: Vortex mix the pre-coupled antibody magnetic bead reagent for 1 min; (4) Preparation of washing buffer: Take 10×Wash Buffer and bring it to room temperature. Vortex mix well. Take 60 mL of 10×Wash Buffer and add it to 540 mL of pure water. Mix well and set aside (the unused portion can be stored in a refrigerator at 4 ℃ for 1 month). (5) Reconstitution of Standards: Before testing, reconstitute the PLEX pedition standards according to the kit instructions. ① Take out the 7 PLEX pedition standards from the kit, gently invert several times to mix the samples in the bottle, then briefly centrifuge the standards in a centrifuge, and let the standards stand at room temperature after centrifugation. After 10 min, add 25 µL of pure water to the standards and vortex to mix; ② Take 20 µL of each reconstituted standard and transfer it to a sterile EP tube, and add Assaybuffer to make the final volume 200 µL (Standard 7); (6) Preparation of standards: Prepare 6 sterile EP tubes and label them Standard 1 to Standard 6. Add 150 µL of Assay buffer to each of the 6 EP tubes. Add 50 µL of Standard 7 solution to the EP tube labeled Standard 6. After thorough mixing, aspirate 50 µL into the EP tube labeled Standard 5. After thorough mixing, aspirate 50 µL into the EP tube labeled Standard 4. Perform serial dilutions of the subsequent standards by 4-fold, for a total of 7 standards. The specific dilution procedure is shown in the table below: Table 1. Standard preparation process (7) Take out the 96-well plate from the kit, arrange the samples according to the kit instructions, then add 200 µL of washing buffer to the 96-well plate, seal the 96-well plate and incubate it at room temperature on a flat shaker for 10 min, then discard the washing buffer and invert the 96-well plate onto absorbent paper to remove any residual liquid in the wells. (8) Add 25 µL of Assay buffer to the 96-well plate. According to the pre-designed sample loading arrangement, add 25 µL of Assay buffer, standard and sample to the corresponding blank well, standard well and sample well respectively. Then add 25 µL of pre-coupled antibody magnetic beads to the 96-well plate. (9) Seal the 96-well plate with a sealing film, and then place it on a flat plate shaker and incubate it at 2-8 ℃ for 16-18 h; (10) Place the 96-well plate on a handheld magnetic rack and let it stand for 60 seconds to allow the magnetic beads to completely sink to the bottom of the 96-well plate. Then gently discard the contents of the 96-well plate (during this process, the 96-well plate and the handheld magnetic rack remain in contact) and invert the 96-well plate onto absorbent paper to remove any remaining liquid in the wells. (11) Remove the 96-well plate from the handheld magnetic rack, add 200 µL of washing buffer to each well, and shake on a plate shaker for 30 seconds to rinse the magnetic beads. (12) Repeat steps (10) and (11) twice; (13) Add 25 µL of detection antibody to each well of a 96-well plate, seal the plate with foil and place it on a plate shaker, and incubate at room temperature (20~25 ℃) for 1 h. (14) Repeat the washing of the 96-well plate three times according to the methods in steps (10) and (11); (15) Add 25 µL of phycoerythrin-labeled streptavidin to each well of a 96-well plate, seal the plate with aluminum foil and place it on a plate shaker, and incubate at room temperature (20~25 ℃) for 30 min. (16) Repeat the washing of the 96-well plate three times according to the methods in steps (10) and (11); (17) Add 150 µL of sheath fluid to each well of a 96-well plate, then place the 96-well plate on a plate shaker and shake for 5 min to resuspend the magnetic beads. Then, use the instrument to detect the cytokine levels in the plasma sample. (18) The instrument reads the fluorescence value of the wells of the standard sample and obtains the fitting curve using the 5-parameters logistic method recommended in the instruction manual. The instrument reads the fluorescence value of the submitted sample and substitutes it into the 5-parameters logistic fitting curve to calculate the concentration of each marker. (19) Data quality control standards: ① R2 of the standard curve > 0.99; ② Intra-batch coefficient of variation of all tested samples < 15%, and inter-batch variation effect < 20%.

[0027] 4. Construction and validation of cytokine immune status assessment and clinical diagnostic models. Based on cytokine detection data from HIV / Mtb co-infected individuals and HIV-infected individuals with corresponding respiratory and systemic symptoms, we explored the value of cytokines in assessing the immune status and clinical diagnosis of HIV / Mtb co-infection.

[0028] 4.1 Data Partitioning Based on the "sample.split" function in the "Caret" package, we divided the subjects into training and test sets in a 7:3 ratio. The data in the training set was used to build the machine learning model, while the data in the test set was used to validate the model. 4.2 Constructing a Least Absolute Shrinkage and Selection Operator (Lasso) Regression Model for Feature Variable Selection (1) Use the “scale” function to standardize the training set data; (2) Use the "glmnet" function to construct the Lasso model, and select "binomal" for the family parameter; (3) Use the "cv.glmnet" function to perform 10-fold cross-validation to select the most suitable lambda value, and select lambda.1se in the output result as the optimal lambda for the final model construction; (4) The variables with non-zero coefficients in the final model output are the characteristic variables most relevant to the diagnosis of HIV / Mtb co-infection in this study.

[0029] The results are as follows Figure 1 As shown, Figure 1A shows the variation of regression coefficients for cytokine variables included in the Lasso analysis with the Lambda value. As the parameter Lambda increases, the number of variables with a regression coefficient of 0 gradually increases. 10-fold cross-validation was used to select the most suitable Lambda value, and the results are as follows... Figure 1 As shown in B. Choosing Lambda = 1se as the optimal Lambda value yields the best Lasso regression model. The model output correlates with HIV / Mtb The most relevant characteristic variables for co-infection include: FGF-2, IL-7, I-309, TNF-α, MMP-1, and MMP-7.

[0030] Example 2: Construction and Evaluation of Immune Status Assessment and Clinical Diagnostic Model Based on the training set data, we constructed six models: Support Vector Machine (SVM), Logistic Regression (LR), Extreme Gradient Boosting (XGboost), Random Forest (RF), Naive Bayes (NB), and Neural Network (NNET), to explore the value of selected variable combinations in assessing the immune status and clinical diagnosis of HIV / Mtb co-infection. Subsequently, we evaluated the models using the test set data. (1) The “svm” function in the “e1071” package is used to build the SVM model, the “lrm” function in the “rms” package is used to build the LR model, the “xgboost” package is used to build the XGboost model, the “randomForest” package is used to build the RF model, the “NaïveBayes” function in the “e1071” package is used to build the NB model, and the “nnet” package is used to build the NNET model; (2) After the initial model construction is completed, the model parameters are tuned using methods such as 10-fold cross-validation and grid search, and then the final model is constructed based on the optimized parameters; (3) The model is evaluated using indicators such as accuracy, sensitivity, specificity, precision, F1 score and area under curve (AUC), and then the optimal diagnostic model is selected based on the results of the evaluation indicators.

[0031] The results are as follows Figure 2As shown, FGF-2, IL-7, I-309, and TNF-α are potential biomarkers for assessing the immune status and clinical diagnosis of HIV / Mtb co-infection, and can be used to differentiate between HIV-infected individuals and HIV / Mtb co-infected individuals, with AUCs of 0.748, 0.731, 0.797, and 0.726, respectively. However, cytokines such as MMP-1 and MMP-7, when used alone, are not effective in assessing the immune status and clinical diagnosis of HIV / Mtb co-infection, with AUCs all less than 0.70. Figure 2 (AF). Its effectiveness as a standalone immunologic status assessment and clinical diagnosis may be poor.

[0032] Based on the above results, we consider that the combination of multiple cytokines may, to some extent, improve HIV / Mtb To assess the sensitivity and specificity of co-infection immune status assessment and clinical diagnosis, we constructed various machine learning models (including SVM, LR, XGboost, RF, NB, and NNET models) using feature variables selected by the Lasso regression model. We then compared the accuracy, sensitivity, specificity, precision, F1 score, and AUC of the constructed models on the training and validation sets to select the best model for immune status assessment and clinical diagnosis.

[0033] XGBoost is an ensemble learning model based on Gradient Boosting Trees (GBDT), where each decision tree... ft The model's mission is not to directly output a class label, but rather a score (or "log odds"). For a given sample, the model sums its output scores across all trees to obtain the sample's total score. The core of this model is to achieve prediction through a weighted ensemble of multiple decision trees. For binary or multi-class classification tasks (such as distinguishing between healthy controls, HIV-only infection, and HIV / Mtb co-infection), its final prediction formula can be expressed as: In this formula, S Represents the total score of each decision tree. f t These represent the functions used in different applications, with Σ representing the mathematical summation sign. m The index categories are represented by 0 (0 for "healthy control", 1 for "HIV-only infection", and 2 for "HIV / Mtb co-infection"). f t m This represents the t-th decision tree under the m-th class. f(x) The feature vector is the patient's test value set.

[0034] According to this formula, assuming there are K decision trees, then for a given sample, the input feature vector of that sample... f ( x ), through the corresponding function f t (Binary classification uses the Sigmoid function, multi-class classification uses the Softmax function) to calculate the final total score. In actual computation, the model will adjust the total score according to different task types. S The probability of a label is converted into the probability of belonging to each category using the Softmax function. Finally, the category with the highest probability is selected as the final predicted label, and the prediction result is output.

[0035] In short, XGBoost integrates multiple weak learners through an additive model (adding the predictions from multiple trees), then converts the aggregated scores into probabilities using a Sigmoid (binary classification) or Softmax (multi-class classification) function, and finally makes classification predictions based on the probabilities.

[0036] The XGBoost model takes the concentration (x) of key immune factors associated with HIV / Mtb infection as input and outputs classification results or probabilities, thereby assisting in the assessment of immune status and clinical diagnosis of HIV infection and HIV / Mtb co-infection. The specific process is as follows: 1. Selection and standardization of input features.

[0037] The model's input consists of key immune factors (FGF2, IL7, I309, TNF-α, MMP-1, MMP-7) identified in the prior screening as being associated with HIV / Mtb infection. Before input, the features need to be standardized (e.g., z-score processing) to ensure fair weighting of factors of different magnitudes in the model.

[0038] 2. The model's prediction process is related to immune status.

[0039] The model uses multiple decision trees to "analyze" the input features layer by layer: each decision tree splits the sample according to the feature threshold, and finally falls into a certain leaf node, outputting the weight of that node. The outputs of all decision trees are weighted and summed, and then converted into probability values ​​by an activation function.

[0040] Specifically, after inputting the results of the selected meaningful cytokines (FGF2, IL7, I309, TNF-α, MMP-1, MMP-7), the model outputs a continuous probability value (e.g., 0.23, 0.78). The prediction model converts this probability value into a final classification decision (0 or 1, i.e., "not infected" or "infected"). The prediction rule is based on a preset threshold. In this embodiment, a preset threshold >70% is considered "high risk," 30%~70% is "medium risk," and <30% is "low risk."

[0041] The results are as follows Figure 3 As shown, in the training set, the XGboost model has an accuracy of 0.875, a sensitivity of 0.842, a specificity of 0.905, a precision of 0.889, and an F1 score of 0.865. Figure 3 A). On the test set, the XGboost model achieved an accuracy of 0.804, a sensitivity of 0.792, a specificity of 0.815, a precision of 0.792, and an F1 score of 0.792. Figure 3 B). Furthermore, the AUC of the XGboost model on the training and test sets were 0.873 and 0.803, respectively. Figure 4 A and Figure 4 B). Compared with other models, the XGboost model performed well on both the training and test sets. Therefore, we chose the XGboost model constructed using the feature variables selected by the Lasso regression model as the HIV / Mtb The optimal model for assessing co-infection immune status and clinical diagnosis. Subsequently, we evaluated the importance of the variables included in the XGboost model based on SHAP values. The results showed that I-309, IL-7, and FGF-2 were the top three variables contributing most to the assessment of HIV / Mtb co-infection immune status and clinical diagnosis. Figure 5 ).

[0042] IGRA is a reliable indicator for diagnosing HIV / Mtb co-infection; however, its role in CD4... + The sensitivity was low in patients with T cell counts below 200 cells / µL. Therefore, we compared the multi-cytokine model developed in this study with IGRA in CD4. + Diagnostic efficacy in patients with fewer than 200 T cells / µL. (Example) Figure 6 As shown, in CD4 + In HIV-infected individuals with T cells below 200 cells / µL, the AUC for diagnosing Mtb infection using the multi-cytokine model was 0.815, while the AUC for diagnosing Mtb infection using IGRA was 0.692. Our diagnostic model can be used for HIV / Mtb co-infected individuals (especially those with CD4+). +This provides a more reliable diagnosis for patients with low T-cell counts.

[0043] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

Claims

1. A combination of biomarkers for assessing immune status and clinical diagnosis of HIV / Mtb co-infection, characterized in that, The biomarkers include at least one or more of the following: FGF-2, IL-7, chemokine I-309, TNF-α, MMP-1, and MMP-7.

2. The biomarker combination according to claim 1, characterized in that, The combination contains FGF-2, IL-7, I-309, and TNF-α.

3. The biomarker combination according to claim 1 or 2, characterized in that, It further includes MMP-1 and / or MMP-7.

4. An in vitro diagnostic kit for detecting HIV / Mtb co-infection, characterized in that, The kit contains detection reagents for detecting the biomarkers of any one of claims 1-3.

5. The reagent kit according to claim 4, characterized in that, The detection reagents are selected from ELISA reagents, immunochromatographic reagents, flow cytometry reagents, or Luminex multifactor detection reagents.

6. The use of the combination of biomarkers according to any one of claims 1-3 in the preparation of products for assessing the immune status of HIV / Mtb co-infection.

7. The use of the combination of biomarkers according to any one of claims 1-3 in the preparation of clinical diagnostic products for HIV / Mtb co-infection.

8. The biomarker combination according to any one of claims 1-3 in the preparation of a product for low CD4 + Application of T-cell counting in diagnostic products for HIV / Mtb co-infection in HIV-infected individuals.

9. The application according to any one of claims 6-8, characterized in that, The product in question is a reagent kit.

Images