Application of biomarker in diagnosis of lung cancer soft meningeal metastasis

Abstract

The invention discloses application of a biomarker in diagnosis of lung cancer soft meningeal metastasis. According to the present invention, it is found for the first time that the metabolites such as Indolastic acid, L-Glutic acid, N-Lactoylphenylalanine, cis-Acetic acid, Formiglumic acid and N-Acetylneomic acid in the cerebrospinal fluid have good diagnosis efficiency on the lung cancer soft meningeal metastasis, have high accuracy, high sensitivity and high specificity, can be used as the biological marker of the lung cancer soft meningeal metastasis, and can be used in the effective diagnosis of the lung cancer soft meningeal metastasis. The invention provides a brand-new thought and strategy for research and development of related diagnostic products for the lung cancer soft meningeal metastasis, and has wide application prospects and important transformation significance in the technical field of diagnosis of the lung cancer soft meningeal metastasis.

Description

Technical Field

[0001] This invention belongs to the field of biomedical technology, specifically, it relates to the application of a biomarker in the diagnosis of leptomeningeal metastasis in lung cancer. Background Technology

[0002] Leptomeningeal metastasis (LM) refers to the spread of malignant tumor cells to the subarachnoid space, pia mater, and arachnoid mater, and / or accompanied by cerebrospinal fluid (CSF) dissemination. In LM, tumor cells affect the pia mater and spread into the CSF. Due to the presence of the blood-brain barrier and the blood-cerebrospinal fluid barrier, traditional chemotherapy drugs struggle to achieve effective therapeutic concentrations in the CSF. This often results in LM having a devastating impact on patient survival. LM has an extremely poor prognosis and a very short survival time. In clinical practice, early diagnosis of LM and appropriate treatment of LM patients can significantly prolong their survival.

[0003] Clinically, the diagnosis of lumbar leukemia (LM) is relatively complex and difficult, currently relying on clinical symptoms, cerebrospinal fluid cytology, and neuroimaging findings. The main symptoms of LM are those of increased intracranial pressure and cranial nerve involvement. However, these changes are not specific and can occur in other situations. Imaging results are confirmed by magnetic resonance imaging (MRI), characterized by nodular or diffuse linear enhancement on the surface of the pia mater. Cerebrospinal fluid cytology results are categorized as positive, negative, or indeterminate; indeterminate cytology results require repeat lumbar puncture to obtain definitive cytological findings.

[0004] Given the low diagnostic accuracy of current methods for liver lesions (LM), and the low sensitivity of the gold standard cerebrospinal fluid cytology examination and the lack of specificity of MRI, developing new diagnostic methods is a pressing issue in clinical practice. Diagnostic prediction models are tools that estimate the probability of a specific disease or condition occurring by assigning relative weights to each diagnostic predictor, combining multiple predictors, and obtaining risk probabilities. Based on these probabilities, they determine whether the patient should undergo further examination or begin treatment directly. Therefore, identifying effective biomarkers to construct predictive models for LM may be an effective means of diagnosing LM. Summary of the Invention

[0005] To overcome the shortcomings of existing technologies, this invention provides an application of biomarkers in the diagnosis of leptomeningeal metastases in lung cancer. These biomarkers include one or more of indolelactic acid, L-Glutamic acid, N-Lactoylphenylalanine, cis-Aconitic acid, Forminoglutamic acid, and N-Acetylneuraminic acid. The research results of this invention provide compelling evidence that the aforementioned biomarkers can effectively diagnose leptomeningeal metastases in lung cancer, providing an important tool for early intervention and treatment of leptomeningeal metastases in lung cancer.

[0006] To achieve the above objectives, the present invention adopts the following technical solution:

[0007] The first aspect of this invention provides the application of reagents for detecting the level of biomarkers in samples in the preparation of diagnostic or auxiliary diagnostic products for leptomeningeal metastases, wherein the biomarkers include one or more of indolelactic acid, L-Glutamic acid, N-Lactoylphenylalanine, cis-Aconitic acid, Forminoglutamic acid, and N-Acetylneuraminic acid.

[0008] In this invention, diagnosis refers to determining the health status of a subject, encompassing aspects such as detecting the presence or absence of disease, response to treatment, recurrence risk assessment, assessment of disease risk and severity, and prognosis. In some cases, the term "diagnosis" refers to a single factor used to determine, verify, or confirm a patient's clinical status.

[0009] In this invention, the biomarker refers to a substance (including metabolites, genes, proteins, microorganisms, etc.) that is differentially expressed between subjects with a second phenotype (e.g., no disease) and subjects with a first phenotype (e.g., disease). Specifically, it is significantly present (i.e., increased or decreased) in biological samples from subjects with a first phenotype (e.g., disease) compared to biological samples from subjects with a second phenotype (e.g., no disease). The biomarker can be differentially present at any level. In a specific embodiment of this invention, the biomarker refers to a metabolite, specifically including one or more of indolelactic acid, L-Glutamic acid, N-Lactoylphenylalanine, cis-Aconitic acid, Forminoglutamic acid, and N-Acetylneuraminic acid.

[0010] Furthermore, the leptomeningeal metastasis is cancerous leptomeningeal metastasis.

[0011] Furthermore, the cancer leptomeningeal metastases include lung cancer leptomeningeal metastases, breast cancer leptomeningeal metastases, gastric cancer leptomeningeal metastases, or melanoma leptomeningeal metastases.

[0012] Furthermore, the leptomeningeal metastasis is a lung cancer leptomeningeal metastasis.

[0013] Furthermore, the reagents include those used to detect the levels of biomarkers in samples using targeted or non-targeted nuclear magnetic resonance, chromatography, spectroscopy, mass spectrometry, or chromatography-mass spectrometry.

[0014] Furthermore, the products include reagent kits, chips, test strips, high-throughput sequencing platforms, systems, equipment, and devices.

[0015] In some implementations, the kit further includes a container, instructions for use, a positive control, a negative control, a buffer solution, and auxiliaries or solvents. The instructions describe how to use the kit for testing and how to use the test results to determine whether a subject has or is suspected of having leptomeningeal metastases.

[0016] In some implementations, the chip also includes a solid-phase carrier, i.e., various commonly used materials in the chip industry, including but not limited to: nylon film, glass or silicon wafers modified with active groups (such as aldehydes, aminos, etc.), unmodified glass wafers, plastic sheets, microparticles, etc.

[0017] In some embodiments, the subject may be human or non-human and may include, for example, animal strains or species used as a "model system" for research purposes. Similarly, the subject may include adults or adolescents (e.g., children). Furthermore, the subject may refer to any living organism, preferably a mammal (e.g., human or non-human). Examples of mammals include, but are not limited to, any member of the mammalian class: humans, non-human primates (e.g., chimpanzees) and other apes and monkeys; livestock, such as cattle, horses, sheep, goats, pigs; domestic animals, such as rabbits, dogs, and cats; laboratory animals include rodents, such as rats, mice, and guinea pigs. Examples of non-mammals include, but are not limited to, birds, fish, etc. More specifically, the subject described in this invention is a human.

[0018] In some implementations, the sample includes any sample obtained from cell, tissue, or body fluid collection (subject-derived), wherein the sample includes, but is not limited to: tissue or cell samples that may be derived from solid samples of fresh, frozen, and / or preserved organ or tissue samples, or biopsies or aspirates, blood or any blood component; body fluids such as cerebrospinal fluid, amniotic fluid, peritoneal fluid, or interstitial fluid. Tissue samples may be primary or in vitro cultured cells or cell lines. Optionally, tissue or cell samples are obtained from diseased tissue / organ. Tissue samples may contain compounds naturally mixed with the tissue, such as preservatives, anticoagulants, buffers, fixatives, nutrients, antibiotics, or similar compounds.

[0019] Furthermore, the sample is a cerebrospinal fluid sample.

[0020] In this invention, verification using real clinical samples collected by this invention revealed that the biomarker has good diagnostic efficacy for the diagnosis of leptomeningeal metastasis in lung cancer. The diagnostic efficacy is verified by the receiver operating characteristic (ROC) curve, and the determination of the area under the curve (AUC) helps to compare the accuracy of the classifier across the overall data range.

[0021] In some implementations, the ROC curve is a coordinate graph composed of the false positive probability (1-specificity) on the horizontal axis and the true positive probability (sensitivity) on the vertical axis, plotting the different results obtained by subjects under specific stimulus conditions due to the use of different judgment criteria. The optimal diagnostic threshold can be selected based on the ROC curve: the closer the ROC curve is to the upper left corner, the higher the accuracy of the test. The point of the ROC curve closest to the upper left corner represents the lowest error threshold, with the fewest total false positives and false negatives. The disease identification ability of two or more different diagnostic tests can also be compared using ROC curves: when comparing two or more diagnostic methods for the same disease, the ROC curves of each test can be plotted on the same coordinate system to visually distinguish their superiority or inferiority; the ROC curve closer to the upper left corner represents the most accurate test results. Alternatively, the area under the ROC curve (AUC) of each test can be calculated and compared; the test with the largest AUC has the best diagnostic value.

[0022] A second aspect of the present invention provides a product for diagnosing or assisting in the diagnosis of leptomeningeal metastasis, the product comprising a reagent for detecting the level of the biomarkers described in the first aspect of the present invention in a sample.

[0023] Furthermore, the reagents include those used to detect the levels of biomarkers in samples using targeted or non-targeted nuclear magnetic resonance, chromatography, spectroscopy, mass spectrometry, or chromatography-mass spectrometry.

[0024] Furthermore, the products include reagent kits, chips, test strips, high-throughput sequencing platforms, systems, equipment, and devices.

[0025] Furthermore, the sample is a cerebrospinal fluid sample.

[0026] The third aspect of this invention provides the application of the biomarkers described in the first aspect of this invention in constructing a diagnostic model for leptomeningeal metastasis.

[0027] The fourth aspect of the present invention provides a method for constructing a diagnostic model for leptomeningeal metastasis, the method comprising the steps of acquiring data on the biomarker levels described in the first aspect of the present invention in a sample, and inputting the data into a machine learning algorithm to construct a diagnostic model.

[0028] Furthermore, the diagnostic model obtains results using the following criteria:

[0029] When the expression of any one or more of the biomarkers Indolelactic acid, L-Glutamic acid, N-Lactoylphenylalanine, and cis-Aconitic acid is above a threshold, or the expression of any one or more of Formiminoglutamic acid and N-Acetylneuraminic acid is below a threshold, a classification result is obtained indicating that the subject has or is suspected of having leptomeningeal metastasis; when the expression of any one or more of the biomarkers Indolelactic acid, L-Glutamic acid, N-Lactoylphenylalanine, and cis-Aconitic acid is below a threshold, or the expression of any one or more of Formiminoglutamic acid and N-Acetylneuraminic acid is above a threshold, a classification result is obtained indicating that the subject does not have leptomeningeal metastasis.

[0030] In some embodiments, the threshold is a representative value of a population sample of untreated leptomeningeal metastases, including but not limited to the maximum, third quartile, and mean. In some preferred embodiments of the invention, the population sample includes 20 or more samples, such as 30, 50, 80, 100, 150, 200, 300, 500, or more.

[0031] Furthermore, the machine learning algorithm includes algorithmic models developed using various development tools.

[0032] Furthermore, the development tools include TensorFlow, Scikit-Learn, PyTorch, OpenNN, RapidMiner, Azure Machine Learning, Apache Mahout, Shogun, KNIME, Vertex AI, H2Oai, Anaconda, Keras, Tableau, Fast.ai, Catalyst, Amazon ML, MLJAR, and Spell.

[0033] Furthermore, the algorithm models include linear regression models, logistic regression models, Lasso regression models, Ridge regression models, linear discriminant analysis models, nearest neighbor models, decision tree models, perceptron models, neural network models, support vector machine models, Naive Bayes models, AdaBoost models, GBDT models, XGBoost models, LightGBM models, CatBoost models, and random forest models.

[0034] The fifth aspect of the present invention provides a system for diagnosing or assisting in the diagnosis of pia mater metastasis, the system comprising a processor, an input module, and an output module.

[0035] The input module is used to input the biomarker level data of the subject sample as described in the first aspect of the present invention; the processor is used to perform logical operations on the input information using bioinformatics methods, and the processor includes a computer-readable medium storing instructions, which, when executed by the processor, execute the method for constructing the diagnostic model as described in the fourth aspect of the present invention on the input biomarker level; the output module is used to output the result of whether the subject has or is suspected of having leptomeningeal metastasis.

[0036] As used in this invention, the term "module" refers to a combination of software and / or hardware that can perform a predetermined function. Implementations of hardware, software, or a combination of hardware and software are conceivable.

[0037] The sixth aspect of the present invention provides a device / apparatus for diagnosing or assisting in the diagnosis of pia mater metastasis, the device / apparatus including a memory and a processor; the memory is used to store program instructions, and the processor is used to call the program instructions, and when the program instructions are executed, the unit functions of the diagnostic system for diagnosing or assisting in the diagnosis of pia mater metastasis as described in the fifth aspect of the present invention are realized.

[0038] In this invention, the device and / or apparatus is a method for distinguishing different components, elements, parts, sections, or assemblies at different levels. However, if other terms can achieve the same purpose, they can be replaced by other expressions. Those skilled in the art will recognize that this invention can be implemented as a device, method, or computer program product. Therefore, the disclosure of this invention can be specifically implemented in the following forms: entirely hardware, entirely software (including firmware, resident software, microcode, etc.), or a combination of software and hardware. Furthermore, in some specific embodiments, this invention can also be implemented as a computer program product in one or more computer-readable storage media containing computer-readable program code.

[0039] The seventh aspect of the present invention provides a computer-readable storage medium for diagnosing or assisting in the diagnosis of pia mater metastasis, wherein the computer-readable storage medium stores a computer program, which, when executed by a processor, implements the unit functions of the diagnostic system for diagnosing or assisting in the diagnosis of pia mater metastasis as described in the fifth aspect of the present invention.

[0040] In some implementations, the present invention can be used in a variety of general-purpose or special-purpose computer system environments. Examples include: personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor devices, microprocessor-based systems, set-top boxes, programmable consumer electronics devices, network PCs, minicomputers, mainframe computers, and distributed computing environments including any of the above systems or devices.

[0041] In some embodiments, the above-described modules or steps can be implemented using a computing device. In some embodiments, the modules or steps can be centralized on a single computing device. In some embodiments, the modules or steps can be distributed across multiple computing devices linked by a network or other means. In some embodiments, the computing device stores program code for executing the above-described modules or steps in a storage device. In some embodiments, the modules or steps can be fabricated as integrated circuits individually or combined into a single integrated circuit. No particular implementation method, whether hardware or software, is limited in the specific embodiments of this invention.

[0042] Advantages and beneficial effects of the present invention:

[0043] This invention is the first to discover that the metabolites in cerebrospinal fluid—indolelactic acid, L-Glutamic acid, N-Lactoylphenylalanine, cis-Aconitic acid, Formicinoglutamic acid, and N-Acetylneuraminic acid—possess good diagnostic efficacy for lung cancer leptomeningeal metastasis. Their diagnostic accuracy, sensitivity, and specificity are all high, and they can serve as biomarkers for the effective diagnosis of lung cancer leptomeningeal metastasis. This invention provides a novel approach and strategy for the research and development of diagnostic products related to lung cancer leptomeningeal metastasis, and has broad application prospects and significant translational value in this technical field. Attached Figure Description

[0044] Figure 1 ROC curves for clinical samples of indolelactic acid for diagnosing leptomeningeal metastasis; where A is the lung cancer leptomeningeal metastasis detection set; B is the lung cancer leptomeningeal metastasis validation set; C is breast cancer leptomeningeal metastasis; and D is gastric cancer or melanoma leptomeningeal metastasis.

[0045] Figure 2The abundance variation of indolelactic acid in clinical samples of leptomeningeal metastases is shown; where A is the lung cancer leptomeningeal metastasis discovery set; B is the lung cancer leptomeningeal metastasis validation set; C is breast cancer leptomeningeal metastases; and D is gastric cancer or melanoma leptomeningeal metastases.

[0046] Figure 3 ROC curves for clinical samples of L-Glutamic acid in diagnosing leptomeningeal metastasis; where A is the lung cancer leptomeningeal metastasis detection set; B is the lung cancer leptomeningeal metastasis validation set; C is breast cancer leptomeningeal metastasis; and D is gastric cancer or melanoma leptomeningeal metastasis.

[0047] Figure 4 The abundance variation of L-Glutamic acid in clinical samples of leptomeningeal metastases is shown; where A is the lung cancer leptomeningeal metastasis discovery set; B is the lung cancer leptomeningeal metastasis validation set; C is breast cancer leptomeningeal metastases; and D is gastric cancer or melanoma leptomeningeal metastases.

[0048] Figure 5 ROC curves for clinical samples of N-Lactoylphenylalanine in diagnosing leptomeningeal metastasis; where A is the lung cancer leptomeningeal metastasis detection set; B is the lung cancer leptomeningeal metastasis validation set; C is breast cancer leptomeningeal metastasis; and D is gastric cancer or melanoma leptomeningeal metastasis.

[0049] Figure 6 The abundance variation of N-Lactoylphenylalanine in clinical samples of leptomeningeal metastases is shown; where A is the lung cancer leptomeningeal metastasis discovery set; B is the lung cancer leptomeningeal metastasis validation set; C is breast cancer leptomeningeal metastases; and D is gastric cancer or melanoma leptomeningeal metastases.

[0050] Figure 7 ROC curves for clinical samples of cis-Aconitic acid in diagnosing leptomeningeal metastasis; where A is the lung cancer leptomeningeal metastasis detection set; B is the lung cancer leptomeningeal metastasis validation set; C is breast cancer leptomeningeal metastasis; and D is gastric cancer or melanoma leptomeningeal metastasis.

[0051] Figure 8 The abundance variation of cis-Aconitic acid in clinical samples of leptomeningeal metastases is shown; where A is the lung cancer leptomeningeal metastasis discovery set; B is the lung cancer leptomeningeal metastasis validation set; C is breast cancer leptomeningeal metastases; and D is gastric cancer or melanoma leptomeningeal metastases.

[0052] Figure 9ROC curves for clinical samples of Formiminoglutamic acid in diagnosing leptomeningeal metastasis; where A is the lung cancer leptomeningeal metastasis detection set; B is the lung cancer leptomeningeal metastasis validation set; C is breast cancer leptomeningeal metastasis; and D is gastric cancer or melanoma leptomeningeal metastasis.

[0053] Figure 10 The abundance variation of formiminoglutamic acid in clinical samples of leptomeningeal metastases is shown; where A is the lung cancer leptomeningeal metastasis discovery set; B is the lung cancer leptomeningeal metastasis validation set; C is breast cancer leptomeningeal metastases; and D is gastric cancer or melanoma leptomeningeal metastases.

[0054] Figure 11 ROC curves for clinical samples of N-Acetylneuraminic acid in diagnosing leptomeningeal metastasis; where A is the lung cancer leptomeningeal metastasis detection set; B is the lung cancer leptomeningeal metastasis validation set; C is breast cancer leptomeningeal metastasis; and D is gastric cancer or melanoma leptomeningeal metastasis.

[0055] Figure 12 The abundance variation of N-Acetylneuraminic acid in clinical samples of leptomeningeal metastases is shown; where A is the lung cancer leptomeningeal metastasis discovery set; B is the lung cancer leptomeningeal metastasis validation set; C is breast cancer leptomeningeal metastases; and D is gastric cancer or melanoma leptomeningeal metastases. Detailed Implementation

[0056] As used in this invention, the terms “having,” “comprising,” or “including,” or any grammatical variations thereof, are used in a non-exclusive manner. Thus, these terms can all refer to a situation where no other features exist in the entity described in this context besides the features introduced by these terms, and can also refer to a situation where one or more other features are present.

[0057] Furthermore, as used in this invention, the terms “preferred,” “more preferred,” “most preferred,” “particularly,” “more particularly,” “specifically,” “more specifically,” or similar terms are used in combination with optional features without limiting other possibilities.

[0058] Unless otherwise stated, all figures used in this specification and claims to represent volume, weight, temperature, time, density, parts by weight, technical effect, etc., should in any case be understood to be modified by the terms "about" or "approximately". Therefore, unless indicated to the contrary, the numerical parameters listed in the specification and appended claims are approximate values. Unless otherwise stated, the terminology used herein has its common meaning as understood by one of ordinary skill in the art, and each numerical parameter should be interpreted according to the number of significant figures and conventional rounding methods or in a manner understood by one of those skilled in the art.

[0059] Example: Study on the effectiveness of biomarkers in diagnosing leptomeningeal metastasis in lung cancer

[0060] I. Experimental Methods

[0061] 1. Clinical Samples

[0062] This study included 210 clinical cerebrospinal fluid (CSF) samples. Control group patients needed to meet the following criteria: no signs of leptomeningeal involvement on MRI and three consecutive negative CSF cytology results. While this criterion significantly reduced the risk of misclassification, we acknowledge that the possibility of false negatives cannot be completely ruled out due to the limitations of CSF cytology and MRI sensitivity. All lumbar punctures were performed strictly according to relevant clinical guidelines, primarily targeting patients with significant meningeal irritation signs and a high suspicion of meningeal leptomeningeal tumour (LM). The relatively large sample size reflects a case accumulation spanning many years and does not necessarily indicate an abnormally high clinical suspicion of LM at a single institution or time point.

[0063] The cohort consisted of: 1) a discovery set cohort: a control group of 11 patients and 39 patients with lung cancer leptomeningeal metastases (LC-LM) (tumor cells detected in cerebrospinal fluid). 2) a validation set cohort: a control group of 16 patients and 60 patients with lung cancer leptomeningeal metastases (LC-LM) (tumor cells detected in cerebrospinal fluid). In addition, we included 12 patients with lung cancer brain parenchymal metastases (no tumor cells detected in cerebrospinal fluid), 21 patients with breast cancer LM, 15 patients with LM from other tumors such as melanoma and gastric cancer, and 36 patients with other diseases (including meningeal infections). After completing routine clinical examinations, residual cerebrospinal fluid was collected for metabolomics analysis. All cerebrospinal fluid samples were processed within 2 hours of collection: centrifuged at 2500 rpm at 4°C for 5 minutes; immediately after centrifugation, the supernatant was aliquoted into multiple small volumetric tubes, avoiding repeated freeze-thaw cycles, and stored at -80°C until analysis.

[0064] 2. Analytical Methods

[0065] Diagnostic model construction

[0066] We leveraged machine learning techniques to construct a data-driven predictive model, transforming complex, high-dimensional metabolomics data into a robust set of diagnostic indicators that can distinguish between LC-LM. First, we focused on significantly differentially expressed metabolites identified through univariate analysis. Then, we implemented an ensemble feature selection strategy by employing four machine learning algorithms—Lasso Minimal Shrinkage and Selection Operator (LASSO), LightGBM, Random Forest (RF), and Extreme Gradient Boosting (XGBoost). Each algorithm ranked metabolite importance based on its discriminative power: LASSO produced regression coefficients, while tree-based algorithms (LightGBM, RF, and XGBoost) calculated feature importance scores based on Gini impurity or SHAP values. Through comprehensive screening using cross-validation, we identified a set of metabolite features repeatedly selected by multiple algorithms, thereby increasing confidence in the selected biomarkers.

[0067] Based on an optimized feature set, we constructed a diagnostic classification model for LC-LM. We systematically evaluated nine commonly used machine learning classifiers, including ensemble methods (CatBoost, LightGBM, RF, XGBoost, Gradient Boosting GBoost) as well as logistic regression (LR), support vector machines (SVM), k-nearest neighbors (KNN), and decision trees (DT). In total, we generated and tested 198 different predictive models, covering a wide range of algorithms and feature combinations. This exhaustive approach allowed us to explore the complete model space, avoiding bias towards a single modeling technique. Model training was primarily performed on the discovery set, with performance evaluated using metrics such as AUC, sensitivity, specificity, and accuracy. Model performance was then validated on an independent validation set. Finally, we optimized the number of features incorporated into the model and evaluated its stability and generalization ability to improve its reliability and clinical applicability.

[0068] II. Experimental Results

[0069] 1. Indolelactic acid

[0070] The results showed that the abundance of indolelactic acid was significantly upregulated in both the discovery and validation sets of patients with lung cancer leptomeningeal metastases compared to the control group. Figure 2 A, Figure 2 B). The receiver operating characteristic (ROC) curve of indolelactic acid was calculated to distinguish between LC-LM patients and healthy controls. The results showed that the area under the curve (AUC) of indolelactic acid was 0.858 (…). Figure 1 A).

[0071] To verify the diagnostic efficacy of indolelactic acid, we plotted ROC curves again using validation set cohort data. The results showed that in this independent validation cohort, indolelactic acid also exhibited high diagnostic performance, with an AUC of 0.977 (…). Figure 1 B).

[0072] Furthermore, we also found that indolelactic acid showed a significant upregulation trend in abundance in other sources, including breast cancer leptomeningeal metastases, gastric cancer, or melanoma leptomeningeal metastases. Figure 2 C Figure 2 D), and it also demonstrates good diagnostic efficacy in diagnosing leptomeningeal metastases from other sources, including breast cancer ( Figure 1 C) AUC was 0.873; gastric cancer and melanoma ( Figure 1 D) The AUC is 0.7259.

[0073] 2. L-Glutamic acid

[0074] The results showed that the abundance of L-Glutamic acid was significantly upregulated in both the discovery and validation sets of patients with lung cancer leptomeningeal metastases compared to the control group. Figure 4 A, Figure 4 B). The receiver operating characteristic (ROC) curve of L-Glutamic acid in the discovery set was used to distinguish between LC-LM patients and healthy controls. The results showed that the area under the curve (AUC) of L-Glutamic acid was 0.876 (…). Figure 3 A).

[0075] To verify the diagnostic efficacy of L-Glutamic acid, we plotted ROC curves again using validation set cohort data. The results showed that L-Glutamic acid also exhibited high diagnostic performance in this independent validation cohort, with an AUC of 0.981 (…). Figure 3 B).

[0076] Furthermore, we also found that L-Glutamic acid showed a significant upregulation trend in abundance in other sources, including breast cancer leptomeningeal metastases, gastric cancer, or melanoma leptomeningeal metastases. Figure 4 C Figure 4 D), and it also demonstrates good diagnostic efficacy in diagnosing leptomeningeal metastases from other sources, including breast cancer ( Figure 3 C) AUC was 0.8466; gastric cancer and melanoma ( Figure 3 D) The AUC is 0.7926.

[0077] 3. N-Lactoylphenylalanine

[0078] The results showed that the abundance of N-Lactoylphenylalanine was significantly upregulated in both the discovery and validation sets of patients with lung cancer leptomeningeal metastases compared to the control group. Figure 6 A, Figure 6 B). The receiver operating characteristic (ROC) curve of N-Lactoylphenylalanine in the discovery set was used to distinguish between LC-LM patients and healthy controls. The results showed that the area under the curve (AUC) of N-Lactoylphenylalanine was 0.86 (…). Figure 5 A).

[0079] To verify the diagnostic efficacy of N-Lactoylphenylalanine, we plotted ROC curves again using validation set cohort data. The results showed that N-Lactoylphenylalanine also exhibited high diagnostic performance in this independent validation cohort, with an AUC of 0.96 (…). Figure 5 B).

[0080] Furthermore, we also found that N-Lactoylphenylalanine showed a significant upregulation trend in abundance in other sources, including breast cancer leptomeningeal metastases, gastric cancer, or melanoma leptomeningeal metastases. Figure 6 C Figure 6 D), and it also demonstrates good diagnostic efficacy in diagnosing leptomeningeal metastases from other sources, including breast cancer ( Figure 5 C) AUC was 0.8818; gastric cancer and melanoma ( Figure 5 D) The AUC is 0.7432.

[0081] 4. cis-Aconitic acid

[0082] The results showed that the abundance of cis-aconitic acid was significantly upregulated in both the discovery and validation sets of patients with lung cancer leptomeningeal metastases compared to the control group. Figure 8 A, Figure 8 B). The receiver operating characteristic (ROC) curve of cis-aconitic acid in the discovery set was used to distinguish between LC-LM patients and healthy controls. The results showed that the area under the curve (AUC) of cis-aconitic acid was 0.926 (…). Figure 7 A).

[0083] To verify the diagnostic efficacy of cis-aconitic acid, we plotted ROC curves again using validation set cohort data. The results showed that cis-aconitic acid also exhibited high diagnostic performance in this independent validation cohort, with an AUC of 0.899 (…). Figure 7 B).

[0084] Furthermore, we also found that cis-Aconitic acid showed a significant upregulation trend in abundance in other sources, including breast cancer leptomeningeal metastases, gastric cancer, or melanoma leptomeningeal metastases. Figure 8 C Figure 8 D), and it also demonstrates good diagnostic efficacy in diagnosing leptomeningeal metastases from other sources, including breast cancer ( Figure 7 C) AUC was 0.8307; gastric cancer and melanoma ( Figure 7 D) The AUC is 0.8198.

[0085] 5. Formicoglutamic acid

[0086] The results showed that the abundance of formiminoglutamic acid was significantly downregulated in both the discovery and validation sets of patients with lung cancer leptomeningeal metastases compared to the control group. Figure 10 A, Figure 10 B). The receiver operating characteristic (ROC) curve of formalinoglutamic acid was calculated to distinguish between LC-LM patients and healthy controls. The results showed that the area under the curve (AUC) of formalinoglutamic acid was 0.925 (…). Figure 9 A).

[0087] To verify the diagnostic efficacy of formalinoglutamic acid, we plotted ROC curves again using validation set cohort data. The results showed that formalinoglutamic acid also demonstrated high diagnostic performance in this independent validation cohort, with an AUC of 0.87 (…). Figure 9 B).

[0088] Furthermore, we also found that formininoglutamic acid showed a significant downregulation trend in abundance from other sources, including breast cancer leptomeningeal metastases, gastric cancer, or melanoma leptomeningeal metastases. Figure 10 C Figure 10 D), and it also demonstrates good diagnostic efficacy in diagnosing leptomeningeal metastases from other sources, including breast cancer ( Figure 9 C) AUC was 0.8748; gastric cancer and melanoma ( Figure 9D) The AUC is 0.8815.

[0089] 6. N-Acetylneuraminic acid

[0090] The results showed that the abundance of N-Acetylneuraminic acid was significantly downregulated in both the discovery and validation sets of patients with lung cancer leptomeningeal metastases compared to the control group. Figure 12 A, Figure 12 B). The receiver operating characteristic (ROC) curve of N-Acetylneuraminic acid was calculated to distinguish between LC-LM patients and healthy controls. The results showed that the area under the curve (AUC) of N-Acetylneuraminic acid was 0.921 (…). Figure 11 A).

[0091] To verify the diagnostic efficacy of N-Acetylneuraminic acid, we plotted ROC curves again using validation set cohort data. The results showed that N-Acetylneuraminic acid also exhibited high diagnostic performance in this independent validation cohort, with an AUC of 0.886 (…). Figure 11 B).

[0092] Furthermore, we also found that N-Acetylneuraminic acid showed a significant downregulation trend in abundance in other sources, including breast cancer leptomeningeal metastases, gastric cancer, or melanoma leptomeningeal metastases. Figure 12 C Figure 12 D), and it also demonstrates good diagnostic efficacy in diagnosing leptomeningeal metastases from other sources, including breast cancer ( Figure 11 C) AUC was 0.836; gastric cancer and melanoma ( Figure 11 D) The AUC is 0.9062.

[0093] Based on the above experimental results, we believe that indolelactic acid, L-Glutamic acid, N-Lactoylphenylalanine, cis-Aconitic acid, Formiminoglutamic acid, and N-Acetylneuraminic acid can all serve as reliable biomarkers for diagnosing LM, especially LC-LM.

[0094] The above description of the embodiments is only for understanding the method and core ideas of the present invention. It should be noted that those skilled in the art can make various improvements and modifications to the present invention without departing from the principles of the invention, and these improvements and modifications will also fall within the protection scope of the claims of the present invention.

Claims

1. Use of a reagent for detecting the level of a biomarker in a sample for the manufacture of a diagnostic or an aid-diagnostic product for leptomeningeal metastasis, characterized in that, The biomarkers include one or more of Indolelactic acid, L-Glutamic acid, N-Lactoylphenylalanine, cis-Aconitic acid, Formiminoglutamic acid, N-Acetylneuraminic acid.

2. Use according to claim 1, characterized in that, The leptomeningeal metastasis is cancer leptomeningeal metastasis. Preferably, the cancer leptomeningeal metastasis includes lung cancer leptomeningeal metastasis, breast cancer leptomeningeal metastasis, gastric cancer leptomeningeal metastasis or melanoma leptomeningeal metastasis. Preferably, the leptomeningeal metastasis is lung cancer leptomeningeal metastasis.

3. Use according to claim 1, characterized in that, The product includes a kit, a chip, a test strip, a high-throughput sequencing platform, a system, an equipment, a device.

4. Use according to claim 1, characterized in that, The sample is a cerebrospinal fluid sample.

5. A product for the diagnosis or aiding in the diagnosis of leptomeningeal metastasis, characterized in that, The product includes a reagent for detecting the level of the biomarker of claim 1 in a sample; Preferably, the product includes a kit, a chip, a test strip, a high-throughput sequencing platform, a system, an equipment, a device. Preferably, the sample is a cerebrospinal fluid sample.

6. Use of the biomarker of claim 1 in constructing a leptomeningeal metastasis diagnosis model.

7. A method of constructing a dura mater transfer diagnosis model, characterized by, The steps of the method include obtaining the level data of the biomarker of claim 1 in a sample, and inputting the data into a machine learning algorithm to construct a diagnosis model; Preferably, the diagnosis model obtains a result through the following criteria: When any one or more of the biomarkers Indolelactic acid, L-Glutamic acid, N-Lactoylphenylalanine, cis-Aconitic acid is expressed higher than a threshold value or any one or more of Formiminoglutamic acid, N-Acetylneuraminic acid is expressed lower than a threshold value, a classification result that the subject has or is suspected to have leptomeningeal metastasis is obtained; when any one or more of the biomarkers Indolelactic acid, L-Glutamic acid, N-Lactoylphenylalanine, cis-Aconitic acid is expressed lower than a threshold value or any one or more of Formiminoglutamic acid, N-Acetylneuraminic acid is expressed higher than a threshold value, a classification result that the subject does not have leptomeningeal metastasis is obtained; Preferably, the machine learning algorithm includes an algorithm model developed by using various development tools; Preferably, the machine learning algorithm includes an algorithm model developed by using various development tools; Preferably, the development tool comprises TensorFlow, Scikit Learn, PyTorch, OpenNN, RapidMiner, Azure Machine Learning, Apache Mahout, Shogun, KNIME, Vertex AI, H2Oai, Anaconda, Keras, Tableau, Fast.ai, Catalyst, Amazon ML, MLJAR, Spell; Preferably, the algorithm model comprises a linear regression model, a logistic regression model, a Lasso regression model, a Ridge regression model, a linear discriminant analysis model, a neighbor model, a decision tree model, a perception model, a neural network model, a support vector machine model, a naive Bayes model, an AdaBoost model, a GBDT model, an XGBoost model, a LightGBM model, a CatBoost model, and a random forest model.

8. A system for diagnosing or aiding in the diagnosis of leptomeningeal metastasis, characterized in that The system comprises a processor, an input module, and an output module; The input module is used to input the biomarker level data in the sample of the subject; the processor is used to perform logical operation on the input information by using bioinformatics method; the processor comprises a computer readable medium storing instructions, which, when executed by the processor, perform the construction method of the diagnostic model of claim 7 on the input biomarker level; and the output module is used to output the result of whether the subject has or is suspected to have leptomeningeal metastasis.

9. An apparatus / device for diagnosing or aiding in the diagnosis of leptomeningeal metastasis, characterized in that, The device / apparatus comprises a memory and a processor; the memory is used to store program instructions, and the processor is used to call the program instructions; when the program instructions are executed, the unit functions of the diagnostic system for diagnosing or assisting in diagnosing leptomeningeal metastasis of claim 8 are realized.

10. A computer readable storage medium for diagnosing or aiding in the diagnosis of leptomeningeal metastasis, characterized in that, The computer readable storage medium stores a computer program, and the computer program realizes the unit functions of the diagnostic system for diagnosing or assisting in diagnosing leptomeningeal metastasis of claim 8 when executed by the processor.

Images