Lung adenocarcinoma-derived meningeal carcinoma prognosis prediction method and device, electronic equipment and storage medium

Abstract

The invention discloses a lung adenocarcinoma source meningeal carcinoma prognosis prediction method and device, electronic equipment and a storage medium, and belongs to the field of intelligent prediction methods.The method comprises the steps that genetic information of a cerebrospinal fluid sample of a lung adenocarcinoma source meningeal carcinoma patient is obtained; performing feature extraction on the genetic information to obtain genetic features and content data thereof, the genetic features being IL-6 and / or TNF-alpha; and performing classification prediction based on the content data of the genetic characteristics to obtain a classification result of whether the sample is a sample with good prognosis or poor prognosis, thereby realizing a more accurate prognosis prediction effect of the lung adenocarcinoma source meningeal carcinoma disease, and providing a basis for subsequent clinical diagnosis and treatment.

Description

Technical Field

[0001] The present invention belongs to the field of intelligent prediction methods, and particularly relates to a prognosis prediction method, device, electronic device and storage medium for meningeal carcinomatosis derived from lung adenocarcinoma. Background Art

[0002] Meningeal carcinomatosis (MC) derived from lung adenocarcinoma refers to a central nervous system metastatic tumor in which lung adenocarcinoma cells metastasize into the pia mater of the brain and spinal cord, showing a diffuse, multifocal or localized distribution, with or without metastatic tumor nodules in the brain and spinal cord parenchyma. Lung adenocarcinoma cells are invasive and can break away from the primary tumor and enter the blood circulation or lymphatic system, migrate to the meninges along the blood or lymph fluid, and implant, grow and reproduce on the meninges. Its metastatic pathways include hematogenous metastasis to the choroidal vessels or leptomeningeal vessels and then to the subarachnoid space, retrograde dissemination along the perineural lymphatics and sheaths, metastasis to Batson's veins and then to the subarachnoid cavity, and centripetal expansion along the perivascular lymphatics, etc.

[0003] Meningeal carcinomatosis derived from lung adenocarcinoma is a serious cancer complication, and its clinical manifestations include: increased intracranial pressure symptoms (headache is persistent and progressive; nausea and vomiting are often projectile; visual disturbances such as blurred vision and diplopia will also occur), meningeal irritation signs (manifested as nuchal rigidity, positive Kernig sign and Brudzinski sign), symptoms of brain parenchyma involvement (consciousness disorders such as lethargy and coma, cognitive disorders such as memory loss and disorientation, etc.), symptoms of cranial nerve and spinal nerve damage (diplopia, impaired eye movement, facial numbness, hearing loss, dysphagia, hoarseness, etc. may occur, and spinal nerve involvement will have radicular pain and segmental sensory deficits).

[0004] Currently, the methods commonly used for prognosis prediction of meningeal carcinomatosis derived from lung adenocarcinoma include: clinical factor assessment, imaging assessment, cerebrospinal fluid cytology examination assessment, etc. Among them, it is difficult to accurately quantify the influence degree of each factor in the clinical factor assessment on the prognosis, and there may be differences in the subjective judgments of different doctors; the effects of some treatment means are also affected by many other factors, such as the patient's tolerance and compliance to drugs. Tiny meningeal metastases may be difficult to be accurately detected by imaging examinations, resulting in an underestimation of the severity of the condition; imaging examinations can only provide morphological information of the tumor and cannot reflect the biological characteristics and functional status of tumor cells. There may be false negatives in cerebrospinal fluid cytology examination, which is affected by factors such as sampling time and sampling volume.

[0005] Although these methods commonly used for prognosis prediction of meningeal carcinomatosis derived from lung adenocarcinoma can achieve certain effects at present, the prediction accuracy of these methods for the prognosis of meningeal carcinomatosis derived from lung adenocarcinoma is not high and it is difficult to meet the requirements of clinical applications. Summary of the Invention

[0006] In view of this, aiming at the problems existing in the above-mentioned prior art, the purpose of the present invention is to provide a prognosis prediction method, device, electronic device and storage medium for leptomeningeal carcinomatosis derived from lung adenocarcinoma.

[0007] The present invention adopts the following technical solutions to achieve the above-mentioned invention purpose:

[0008] In the first aspect, the present invention provides a prognosis prediction method for leptomeningeal carcinomatosis derived from lung adenocarcinoma based on IL-6 and / or TNF-α, and the method includes:

[0009] Obtain the genetic information of cerebrospinal fluid samples of patients with leptomeningeal carcinomatosis derived from lung adenocarcinoma;

[0010] Extract features from the genetic information to obtain genetic features and their content data, and the genetic features are IL-6 and / or TNF-α;

[0011] Based on the content data of the genetic features, perform classification prediction to obtain the classification result of whether the sample is a sample with good prognosis or poor prognosis.

[0012] Furthermore, the classification result is obtained based on a prediction model, and the construction method of the prediction model includes:

[0013] Obtain the content data of the genetic features in the training set cerebrospinal fluid samples and the clinical features corresponding to the cerebrospinal fluid samples, where the clinical features are good prognosis or poor prognosis. Extract the content data of the genetic features in the training set and input them into a machine learning model to construct a prediction model, and obtain the constructed prediction model and threshold;

[0014] If the content of IL-6 is higher than the threshold and / or the content of TNF-α is lower than the threshold, obtain the classification result that the sample is a sample with good prognosis;

[0015] If the content of IL-6 is lower than the threshold and / or the content of TNF-α is higher than the threshold, obtain the classification result that the sample is a sample with poor prognosis.

[0016] Furthermore, the machine learning model is a Lasso regression model, linear regression model, logistic regression model, Ridge regression model, linear discriminant analysis model, random forest model, nearest neighbor model, decision tree model, support vector machine model, naive Bayes model and / or perceptron model.

[0017] Furthermore, the content data of the genetic features is obtained by any one or more of the following methods: enzyme-linked immunosorbent assay, immunoblotting, real-time fluorescence quantitative PCR, flow cytometry, liquid chip technology.

[0018] Second aspect, the present invention provides a prognostic prediction system for leptomeningeal carcinomatosis derived from lung adenocarcinoma based on IL-6 and / or TNF-α, the system comprising:

[0019] A genetic information acquisition unit: for acquiring the genetic information of the cerebrospinal fluid sample of a patient with leptomeningeal carcinomatosis derived from lung adenocarcinoma;

[0020] A genetic feature extraction unit: for extracting features from the genetic information to obtain genetic features and their content data, the genetic features being IL-6 and / or TNF-α;

[0021] A disease prognosis prediction unit: for performing classification prediction based on the content data of the genetic features to obtain a classification result of whether the sample is a sample with good prognosis or poor prognosis;

[0022] The classification result is obtained based on a prediction model, and the construction method of the prediction model includes:

[0023] Obtaining the content data of the genetic features in the training set of cerebrospinal fluid samples and the clinical features corresponding to the cerebrospinal fluid samples, the clinical features being good prognosis or poor prognosis, extracting the content data of the genetic features in the training set and inputting them into a machine learning model to construct a prediction model, obtaining the constructed prediction model and a threshold value;

[0024] If the content of IL-6 is higher than the threshold value and / or the content of TNF-α is lower than the threshold value, obtaining a classification result that the sample is a sample with good prognosis;

[0025] If the content of IL-6 is lower than the threshold value and / or the content of TNF-α is higher than the threshold value, obtaining a classification result that the sample is a sample with poor prognosis.

[0026] Further, the machine learning model is a Lasso regression model, a linear regression model, a logistic regression model, a Ridge regression model, a linear discriminant analysis model, a random forest model, a nearest neighbor model, a decision tree model, a support vector machine model, a naive Bayes model and / or a perceptron model.

[0027] Third aspect, the present invention provides a device or electronic device for prognostic prediction of leptomeningeal carcinomatosis derived from lung adenocarcinoma, the device or electronic device comprising a memory and a processor;

[0028] The memory is used for storing program instructions;

[0029] The processor is used for calling the program instructions, and when the program instructions are executed, implementing the prognostic prediction method for leptomeningeal carcinomatosis derived from lung adenocarcinoma based on IL-6 and / or TNF-α described in the first aspect of the present invention.

[0030] Fourth aspect, the present invention provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, it implements the method for predicting the prognosis of leptomeningeal carcinomatosis derived from lung adenocarcinoma according to the first aspect of the present invention.

[0031] Fifth aspect, the present invention provides the use of a reagent for detecting the levels of IL-6 and / or TNF-α in a cerebrospinal fluid sample in the preparation of a product for predicting the prognosis of leptomeningeal carcinomatosis derived from lung adenocarcinoma.

[0032] Furthermore, the reagent includes a reagent for detecting the expression levels of IL-6 and / or TNF-α in a cerebrospinal fluid sample by using one or more of enzyme-linked immunosorbent assay, immunoblotting, real-time fluorescence quantitative PCR, flow cytometry, and liquid chip technology.

[0033] In some embodiments, the reagent includes a reagent for detecting the expression level of the mRNA of IL-6 and / or TNF-α in a cerebrospinal fluid sample, and / or a reagent for detecting the expression level of the protein and / or polypeptide encoded by IL-6 and / or TNF-α in a cerebrospinal fluid sample.

[0034] In some embodiments, the reagent for detecting the expression level of the mRNA of IL-6 and / or TNF-α in a cerebrospinal fluid sample includes a probe specifically recognizing IL-6 and / or TNF-α, and / or a primer specifically amplifying IL-6 and / or TNF-α.

[0035] In some embodiments, the reagent for detecting the expression level of the protein and / or polypeptide encoded by IL-6 and / or TNF-α in a cerebrospinal fluid sample includes an antibody, antibody fragment, and / or affinity protein specifically binding to IL-6 and / or TNF-α.

[0036] Sixth aspect, the present invention provides a product for predicting the prognosis of leptomeningeal carcinomatosis derived from lung adenocarcinoma, and the product contains the reagent for detecting the levels of IL-6 and / or TNF-α in a cerebrospinal fluid sample as described above.

[0037] In some embodiments, the product includes a detection kit, a detection chip, or a test strip.

[0038] In some embodiments, the detection kit further includes an instruction manual or label, a positive control, a negative control, a buffer, an adjuvant, or a solvent; the instruction manual or label details how to use the detection kit provided by the present invention to detect a cerebrospinal fluid sample and the use of the detection kit for predicting the prognosis of a patient with leptomeningeal carcinomatosis derived from lung adenocarcinoma.

[0039] In some embodiments, the detection kit may further comprise a variety of different reagents suitable for practical use (such as for different detection methods), not limited to the reagents listed in the present invention currently. As long as the reagent is based on the detection of IL-6 and / or TNF-α in cerebrospinal fluid samples for prognostic prediction of leptomeningeal carcinomatosis of lung adenocarcinoma origin, it is included within the scope of protection of the present invention.

[0040] In some embodiments, the detection chip can be prepared by using conventional preparation methods of biochips known to those skilled in the art, including but not limited to: using a solid-phase carrier modified with a glass slide or a silicon wafer, with an amino-modified poly dT string at the 5' end of the probe. The oligonucleotide probe is formulated into a solution, and then it is spotted on the modified glass slide or silicon wafer by a spotter, arranged in a predetermined sequence or array, and then fixed by standing overnight to obtain the detection chip of the present invention.

[0041] In some embodiments, the primers included in the product of the present invention can be prepared by chemical synthesis, appropriately designed by referring to known information using methods well-known to those skilled in the art, and prepared by chemical synthesis. In some embodiments, the antibodies included in the product of the present invention can be antibodies or fragments thereof of any structure, size, immunoglobulin class, origin, etc., as long as it binds to the target protein. The antibodies or fragments thereof included in the product of the present invention can be monoclonal or polyclonal. An antibody fragment refers to a part of an antibody that retains the binding activity of the antibody to the antigen or a peptide containing a part of the antibody. Antibody fragments can include F(ab′)2, Fab′, Fab, single-chain Fv (scFv), disulfide-bonded Fv (dsFv) or its polymer, dimerized V regions (bispecific antibody), or a peptide containing CDRs. Antibodies can be obtained by methods well-known to those skilled in the art. For example, a mammalian cell expression vector encoding a polypeptide that retains the whole or part of the target protein or integrating the polynucleotides encoding them is prepared as an antigen. After immunizing an animal with the antigen, immune cells are obtained from the immunized animal and fused with cancer cells to obtain hybridomas. Then the antibodies are collected from the hybridoma cultures. Finally, monoclonal antibodies against IL-6 and / or TNF-α can be obtained by performing antigen-specific purification on the obtained antibodies using IL-6 and / or TNF-α or a part thereof used as the antigen.

[0042] Compared with the prior art, the advantages and beneficial effects of the present invention are as follows:

[0043] (1) The present invention provides a novel prognostic prediction method for leptomeningeal carcinomatosis derived from lung adenocarcinoma in the art. The prediction method includes: obtaining the genetic information of cerebrospinal fluid samples from patients with leptomeningeal carcinomatosis derived from lung adenocarcinoma; extracting features from the genetic information to obtain genetic features and their content data, where the genetic features are IL-6 and / or TNF-α; and performing classification prediction based on the content data of the genetic features to obtain the classification result of whether the sample has a good prognosis or a poor prognosis. The method provided by the present invention can be used for early prediction of disease progression, guiding individualized treatment, or monitoring treatment response, etc.

[0044] (2) The present invention creatively discovers for the first time that there is a close correlation between the expression levels of IL-6 and / or TNF-α in cerebrospinal fluid samples from patients with leptomeningeal carcinomatosis derived from lung adenocarcinoma and the prognostic prediction of patients with leptomeningeal carcinomatosis derived from lung adenocarcinoma. Using the expression levels of IL-6 and / or TNF-α in the cerebrospinal fluid samples as genetic information for prognostic prediction of leptomeningeal carcinomatosis derived from lung adenocarcinoma not only has high prediction accuracy but also high prediction efficiency, meeting the requirements of clinical applications. The present invention provides a novel idea and strategy for prognostic prediction of leptomeningeal carcinomatosis derived from lung adenocarcinoma, provides a more reliable basis for clinical treatment, can be used to help clinicians formulate more precise treatment plans, and has broad application prospects. Description of the Drawings

[0045] Figure 1 : Schematic flowchart of a prognostic prediction method for leptomeningeal carcinomatosis derived from lung adenocarcinoma based on IL-6 and / or TNF-α provided by an embodiment of the present invention;

[0046] Figure 2 : Schematic diagram of a prognostic prediction system for leptomeningeal carcinomatosis derived from lung adenocarcinoma based on IL-6 and / or TNF-α provided by an embodiment of the present invention;

[0047] Figure 3 : Schematic diagram of a device or electronic device for prognostic prediction of leptomeningeal carcinomatosis derived from lung adenocarcinoma provided by an embodiment of the present invention;

[0048] Figure 4 : Result diagram corresponding to the comparison of IL-6, IL-1β, and TNF-α levels in CSF and serum of MC patients;

[0049] Figure 5 : Result diagram corresponding to the significant increase in IL-6, IL-1β, and TNF-α levels in CSF of the MC group compared with CSF of the control group;

[0050] Figure 6 : Result diagram corresponding to the lack of statistical significance in the levels of IL-6, IL-1β, and TNF-α in serum between the MC group and the lung cancer group;

[0051] Figure 7 : The corresponding result graph showing that IL-6 and TNF-α in CSF are positively correlated, and IL-1β and TNF-α in serum are positively correlated. Among them, in Figure A: TNF-α, in Figure B: IL-1β;

[0052] Figure 8 : The corresponding result graph showing that an increase in TNF-α is associated with a poor OS, and an increase in IL-6 is associated with a good OS. Among them, in Figure A: TNF-α, in Figure B: IL-6. Detailed implementation manners

[0053] In order to enable those skilled in the art of the present technology to better understand the solution of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention.

[0054] In some processes described in the specification, claims and the above-mentioned accompanying drawings of the present invention, a plurality of operations appear in a specific order. However, it should be clearly understood that these operations may not be executed in the order in which they appear herein or may be executed in parallel. The serial numbers of the operations, such as S1, S2, etc., are only used to distinguish different operations, and the serial numbers themselves do not represent any execution order. In addition, these processes may include more or fewer operations, and these operations may be executed in sequence or in parallel.

[0055] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without making creative efforts belong to the scope of protection of the present invention.

[0056] Figure 1 This is a schematic flowchart of a method for predicting the prognosis of leptomeningeal carcinomatosis derived from lung adenocarcinoma based on IL-6 and / or TNF-α provided by an embodiment of the present invention. Specifically, the method includes the following steps:

[0057] S1: Obtain the genetic information of the cerebrospinal fluid sample of a patient with leptomeningeal carcinomatosis derived from lung adenocarcinoma;

[0058] In some embodiments, the patient refers to any animal, and also refers to humans and non-human animals. The non-human animals include all vertebrates, for example, mammals, such as non-human primates (especially higher primates), sheep, dogs, rodents (such as mice or rats), guinea pigs, goats, pigs, cats, rabbits, cows, and any domestic animals or pets; and non-mammals, such as chickens, amphibians, reptiles, etc. In the specific implementation of the present invention, the patient is a human.

[0059] In some embodiments, both the IL-6 (Interleukin-6) and TNF-α (Tumor Necrosis Factor-α) are important cytokines that play a key role in processes such as the body's immune response and inflammation regulation.

[0060] Among them, the IL-6 is a glycoprotein, and its gene is located on human chromosome 7. Mature IL-6 consists of 184 amino acids and contains 4 α-helix structures. Its spatial structure is crucial for its binding to the receptor and the exertion of biological functions.

[0061] Among them, the TNF-α is a cytokine, and its gene is located on human chromosome 6. The TNF-α protein encoded by it consists of 233 amino acids. TNF-α can exist in the form of a trimer, and this trimer structure is the key form for its exertion of biological activity.

[0062] S2: Extract features from the genetic information to obtain genetic features and their content data, where the genetic features are IL-6 and / or TNF-α;

[0063] In some embodiments, the content data of the genetic features are obtained by any one or more of the following methods: enzyme-linked immunosorbent assay, immunoblotting, real-time fluorescence quantitative PCR, flow cytometry, liquid phase chip technology, but are not limited to the listed methods. Those skilled in the art understand that any method capable of detecting the content data of the genetic features can be used in the present invention.

[0064] In one embodiment, the enzyme-linked immunosorbent assay (ELISA) utilizes the specific binding of an antigen and an antibody. A known IL-6 or TNF-α antibody is coated on a solid-phase carrier. After adding a cerebrospinal fluid sample, the IL-6 or TNF-α in the sample will bind to the coated antibody. Then, an enzyme-labeled secondary antibody is added to bind to the antigen-antibody complex bound to the solid-phase carrier. Finally, a substrate is added, and the enzyme catalyzes the substrate to develop color. The absorbance value is measured by an enzyme-labeled instrument, and the concentration of IL-6 or TNF-α in the cerebrospinal fluid sample is calculated according to the standard curve.

[0065] In one embodiment, the immunoblotting (Western blot) is to first perform polyacrylamide gel electrophoresis on the proteins in the cerebrospinal fluid sample, separate them according to the molecular weight of the proteins, then transfer the separated proteins to a solid-phase membrane, and then hybridize them with specific IL-6 or TNF-α antibodies. Finally, the protein bands bound to the antibodies are detected by methods such as chemiluminescence or color development, and the expression level of IL-6 or TNF-α in the cerebrospinal fluid sample is semi-quantitatively analyzed according to the intensity of the bands.

[0066] In one embodiment, the real-time fluorescence quantitative PCR (qPCR) uses the total RNA in cerebrospinal fluid as a template, reverse transcribes it into cDNA, and then uses the cDNA as a template for PCR amplification. A fluorescent group is added to the PCR reaction system, and the intensity of the fluorescent signal is proportional to the amount of PCR product. By monitoring the change of the fluorescent signal in real time and using a standard curve to quantitatively analyze the mRNA expression level of IL-6 or TNF-α, the protein expression level can be indirectly reflected.

[0067] In one embodiment, for flow cytometry, the cells in the cerebrospinal fluid sample are incubated with fluorescently labeled IL-6 or TNF-α antibodies. The antibodies specifically bind to IL-6 or TNF-α on the cell surface or inside the cell, and then the fluorescent signal is detected by a flow cytometer. The expression level of IL-6 or TNF-α is analyzed based on the fluorescence intensity and the number of cells.

[0068] In one embodiment, for the liquid chip technology, specific antibodies against various cytokines such as IL-6 and TNF-α are respectively immobilized on different microspheres to form different detection microspheres. These detection microspheres are mixed with the cerebrospinal fluid sample, and the cytokines in the sample bind to the corresponding antibodies. Then, a fluorescently labeled secondary antibody is added, and the fluorescent signal on the microspheres is detected by a liquid chip analyzer to quantitatively analyze multiple cytokines simultaneously.

[0069] S3: Perform classification prediction based on the content data of the genetic characteristics to obtain the classification result of whether the sample is a sample with good prognosis or poor prognosis.

[0070] The classification result is obtained based on a prediction model. The construction method of the prediction model includes:

[0071] Obtain the content data of the genetic characteristics in the training set of cerebrospinal fluid samples and the corresponding clinical characteristics of the cerebrospinal fluid samples. The clinical characteristics are good prognosis or poor prognosis. Extract the content data of the genetic characteristics in the training set and input it into a machine learning model to construct a prediction model, and obtain the constructed prediction model and threshold.

[0072] If the content of IL-6 is higher than the threshold and / or the content of TNF-α is lower than the threshold, obtain the classification result that the sample is a sample with good prognosis.

[0073] If the content of IL-6 is lower than the threshold and / or the content of TNF-α is higher than the threshold, obtain the classification result that the sample is a sample with poor prognosis.

[0074] In some embodiments, the machine learning model is a Lasso regression model, a linear regression model, a logistic regression model, a Ridge regression model, a linear discriminant analysis model, a random forest model, a nearest neighbor model, a decision tree model, a support vector machine model, a naive Bayes model, and / or a perceptron model.

[0075] In one embodiment, the threshold in the machine learning model refers to the critical value used to convert the model output into a classification result. In a binary classification problem, the model usually outputs a real number, which represents the probability of belonging to a certain class. For example, the logistic regression model outputs a probability value between 0 and 1. Usually, the default threshold is 0.5, that is, the samples with a probability greater than or equal to 0.5 are classified as the positive class, and those less than 0.5 are classified as the negative class.

[0076] In one embodiment, taking machine learning models such as the Lasso regression model, the linear regression model, and the logistic regression model as examples, these models are all based on statistical and mathematical principles. By learning and fitting a large amount of training data, they seek the relationship between the input features (i.e., the content data of IL-6 and TNF-α) and the output result (good prognosis or poor prognosis). Taking the logistic regression model as an example, it constructs a logistic function to map the content data of IL-6 and TNF-α to a probability value between 0 and 1, representing the probability that the sample belongs to a good prognosis. By setting a threshold, it is determined whether the sample has a good prognosis or a poor prognosis. Other models such as the random forest model construct multiple decision trees and comprehensively classify and predict based on the results of these decision trees.

[0077] In one embodiment, the following method can be used to determine the threshold: When constructing a prediction model, through the analysis of the training set data and the training of the model, a suitable threshold is found to make the classification effect of the model on the training set the best. This threshold is determined according to the matching degree between the prediction result of the model and the actual clinical features (good prognosis or poor prognosis). Usually, some evaluation indicators such as accuracy, recall rate, and F1 value are used as optimization objectives, and the threshold is adjusted to make these evaluation indicators reach the optimal.

[0078] In one embodiment, the Precision-Recall Curve can be used to select the best threshold. By plotting the Precision-Recall Curve, the impact of different thresholds on the model performance can be intuitively seen. Select the point on the curve to make the precision and recall rate reach the best balance, and this point is the best threshold.

[0079] In one embodiment, the F1 score can be used to select the best threshold. The F1 score is the harmonic mean of the precision and recall rate, and the best threshold is selected by maximizing the F1 score.

[0080] In one embodiment, the ROC curve can be used to select the optimal threshold. The Receiver Operating Characteristic curve (ROC) can also be used to select the optimal threshold. The performance of the model is evaluated by calculating the area under the curve (AUC), and then the optimal threshold is determined.

[0081] In one embodiment, the efficacy of the constructed prediction model can also be predicted. That is, another dataset containing the genetic characteristic content data corresponding to patients with leptomeningeal carcinomatosis from lung adenocarcinoma with good prognosis and patients with leptomeningeal carcinomatosis from lung adenocarcinoma with poor prognosis is taken as the validation set, and the efficacy of the constructed prediction model is further verified in this validation set.

[0082] In one embodiment, when the construction method of the above prediction model is determined, the threshold is included in the obtained prediction model, that is, when the prediction model is determined, the threshold is also determined. Based on the determined threshold, the classification result of whether the sample to be tested is a sample with good prognosis or a sample with poor prognosis can be predicted. The specific judgment result based on the prediction model is: if the content of IL-6 is higher than the threshold and / or the content of TNF-α is lower than the threshold, the classification result that the sample is a sample with good prognosis is obtained; if the content of IL-6 is lower than the threshold and / or the content of TNF-α is higher than the threshold, the classification result that the sample is a sample with poor prognosis is obtained.

[0083] In some embodiments, the prognosis prediction result includes good prognosis or poor prognosis. Among them, good prognosis includes cure, remission or stability, and poor prognosis includes progression or death.

[0084] In some embodiments, the term "cure" means that through existing treatment means and expected treatment processes, the disease can be completely cured, the patient's body returns to the healthy state before the illness, there are no cancer cells or pathogenic factors in the body, the symptoms and signs completely disappear, and there will be no recurrence for a relatively long period of time. For example, for some early-stage malignant tumors, after complete resection of the lesion by surgery and no tumor recurrence is found after long-term follow-up, it can be considered that the prognostic effect of cure has been achieved.

[0085] In some embodiments, the remission can be divided into complete remission and partial remission. Complete remission means that the symptoms and signs of the disease completely disappear, and relevant examination indicators such as imaging examinations and laboratory tests also return to normal, but it does not mean that there are no cancer cells or pathogenic factors in the body at all, but only that they cannot be detected by current detection means. Partial remission means that the symptoms and signs of the disease are significantly reduced, the tumor volume shrinks or relevant abnormal indicators improve, but have not completely returned to normal. For example, in tumor treatment, after chemotherapy, if the tumor volume shrinks by more than a certain proportion and the symptoms are significantly reduced, it belongs to partial remission.

[0086] In some embodiments, the term "stable" means that the progression of the disease is in a relatively static state, with no obvious changes in symptoms and signs, the tumor neither significantly increasing nor decreasing in size, and various examination indicators basically remaining at the original level. This situation may be due to the effective control of the disease progression by treatment, or it may be that the natural course of the disease itself is in a relatively stable stage.

[0087] In some embodiments, the term "progression" indicates that the disease is continuously developing, with symptoms and signs worsening, the tumor volume increasing, new lesions or metastases appearing, and the relevant examination indicators also becoming worse. This may be due to the ineffective existing treatment regimen that fails to effectively inhibit the development of the disease, or the disease itself has developed drug resistance to the treatment, etc.

[0088] In some embodiments, "death" is the worst prognosis outcome, that is, due to the severity of the disease, ineffective treatment, or other uncontrollable factors, the patient ultimately loses their life.

[0089] In one embodiment, the present invention studied the comparison of the levels of IL-6, IL-1β, and TNF-α in CSF and serum. The study found that the PI3K / Akt signaling pathway plays an important regulatory role in processes such as inflammatory responses, cell activation, and apoptosis. It is generally believed that the PI3K / Akt pathway is an upstream activator of the NF-κB signaling cascade, and inflammatory cytokines can participate in the regulation of tumors by activating the NF-κB pathway. Therefore, this study analyzed the levels of IL-6, IL-1β, and TNF-α in cerebrospinal fluid (CSF) and serum of tumor and non-tumor patients. A total of 132 samples were analyzed in this study, including 52 CSF samples and 30 serum samples from patients with meningeal carcinomatosis (MC patients) originating from lung adenocarcinoma, 25 serum samples from patients with stage IV lung cancer, and 25 CSF samples from the control group (non-tumor central nervous system diseases). To analyze the cytokine levels in lung adenocarcinoma and lung adenocarcinoma MC patients, we compared the differences in IL-6, IL-1β, and TNF-α among different groups (as shown in Table 1).

[0090] Table 1 Levels of IL-6, IL-1β, and TNF-α in CSF and serum

[0091]

[0092] The results showed that in the MC group, the levels of IL-6 and TNF-α in CSF were significantly higher than those in serum, with statistically significant differences (P < 0.001, Figure 4 ). However, the level of IL-1β was slightly higher than that in serum, but the difference was not statistically significant (P = 0.07, Figure 4) The levels of IL-6, IL-1β and TNF-α in the CSF of the MC group were significantly higher than those in the CSF of the control group, and the difference was statistically significant (P<0.0001, Figure 5 ) There was no statistical significance in the levels of IL-6, IL-1β and TNF-α in the serum between the MC group and the lung cancer group (P>0.05, Figure 6 ).

[0093] In one embodiment, the present invention studied the correlation analysis of CSF and serum IL-6, IL-1β, TNF-α. Specifically, the correlation of different cytokines was further analyzed ( Figure 7 ) We found that IL-6 and TNF-α in the CSF were positively correlated (r = 0.38, P<0.001, Figure 7 A), while there was no correlation between IL-6 and IL-1β, and between IL-1β and TNF-α (P>0.05). IL-1β and TNF-α in the serum were positively correlated (r = 0.95, P<0.0001, Figure 7 B), while there was no correlation between IL-6 and IL-1β, and between IL-6 and TNF-α (P>0.05).

[0094] In one embodiment, the present invention studied the levels of IL-6, IL-1β, TNF-α in the CSF and the analysis of survival status and survival period. Specifically, we compared the relationship between the cytokine levels in the CSF of MC patients and the KPS and ECOG PS scores, and found that there was no correlation between the levels of IL-6, IL-1β, TNF-α and the KPS and ECOGPS scores (P>0.05). It shows that different levels of cytokines are not related to the survival status score. To clarify the relationship between the levels of IL-6, IL-1β, TNF-α and the survival period (Overall Survival, OS) ( Figure 8 ) We found that TNF-α (40.8 months vs 14.4 months, P = 0.018, Figure 8 A), IL-6 (14.4 months vs not reached, P = 0.034, Figure 8 B) were significantly correlated with the survival period of MC patients, and IL-1β was not related to the survival period (61.5 months vs 14.4 months, P = 0.058).

[0095] The above results indicate that TNF-α and IL-1β are significantly correlated with the prognosis of MC patients. An increase in TNF-α is associated with a poor OS, while an increase in IL-6 is associated with a good OS. That is, the present invention has demonstrated through collecting clinical real samples and a large number of experimental verifications that TNF-α and / or IL-1β in cerebrospinal fluid samples can be used for effective and accurate prediction of the prognosis of patients with meningeal carcinomatosis originating from lung adenocarcinoma. Based on these pioneering research results, the present invention has developed a brand-new method for predicting the prognosis of meningeal carcinomatosis originating from lung adenocarcinoma based on the content of IL-6 and / or TNF-α in cerebrospinal fluid samples of patients with meningeal carcinomatosis originating from lung adenocarcinoma as described above.

[0096] Figure 2 The figure is a schematic diagram of a prognostic prediction system for meningeal carcinomatosis originating from lung adenocarcinoma provided by an embodiment of the present invention. Specifically, the system includes:

[0097] A genetic information acquisition unit: used to acquire the genetic information of cerebrospinal fluid samples of patients with meningeal carcinomatosis originating from lung adenocarcinoma;

[0098] A genetic feature extraction unit: used to extract features from the genetic information to obtain genetic features and their content data, and the genetic features are IL-6 and / or TNF-α;

[0099] A disease prognosis prediction unit: used to perform classification prediction based on the content data of the genetic features to obtain a classification result of whether the sample is a sample with good prognosis or poor prognosis;

[0100] The classification result is obtained based on a prediction model, and the construction method of the prediction model includes:

[0101] Acquire the content data of the genetic features in the training set cerebrospinal fluid samples and the clinical features corresponding to the cerebrospinal fluid samples. The clinical features are good prognosis or poor prognosis. Extract the content data of the genetic features in the training set and input them into a machine learning model to construct a prediction model, and obtain the constructed prediction model and threshold;

[0102] If the content of IL-6 is higher than the threshold and / or the content of TNF-α is lower than the threshold, obtain the classification result that the sample is a sample with good prognosis;

[0103] If the content of IL-6 is lower than the threshold and / or the content of TNF-α is higher than the threshold, obtain the classification result that the sample is a sample with poor prognosis.

[0104] In some embodiments, the machine learning model is a Lasso regression model, a linear regression model, a logistic regression model, a Ridge regression model, a linear discriminant analysis model, a random forest model, a nearest neighbor model, a decision tree model, a support vector machine model, a naive Bayes model, and / or a perceptron model.

[0105] In some embodiments, the content data of the genetic characteristics is obtained by any one or more of the following methods: enzyme-linked immunosorbent assay, immunoblotting, real-time fluorescence quantitative PCR, flow cytometry, liquid chip technology.

[0106] Figure 3 The figure is a schematic diagram of a device or electronic device for predicting the prognosis of leptomeningeal carcinomatosis derived from lung adenocarcinoma provided by an embodiment of the present invention. Specifically, the device or electronic device includes: a memory and a processor;

[0107] The memory is used to store program instructions;

[0108] The processor is used to call the program instructions, and when the program instructions are executed, it implements the method for predicting the prognosis of leptomeningeal carcinomatosis derived from lung adenocarcinoma based on IL-6 and / or TNF-α as described above.

[0109] An embodiment of the present invention also provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, it implements the method for predicting the prognosis of leptomeningeal carcinomatosis derived from lung adenocarcinoma based on IL-6 and / or TNF-α as described above.

[0110] An embodiment of the present invention also provides the use of a reagent for detecting the levels of IL-6 and / or TNF-α in a cerebrospinal fluid sample in the preparation of a product for predicting the prognosis of leptomeningeal carcinomatosis derived from lung adenocarcinoma.

[0111] In some embodiments, the reagent includes a reagent for detecting the expression levels of IL-6 and / or TNF-α in a cerebrospinal fluid sample by one or more of enzyme-linked immunosorbent assay, immunoblotting, real-time fluorescence quantitative PCR, flow cytometry, and liquid chip technology.

[0112] In one embodiment, the cerebrospinal fluid sample is derived from a subject in need, and the subject is preferably a patient with leptomeningeal carcinomatosis derived from lung adenocarcinoma who needs to predict the prognosis of the disease. The present invention has proven through detailed experiments that TNF-α and IL-6 are significantly correlated with the survival period of patients with leptomeningeal carcinomatosis derived from lung adenocarcinoma. Among them, an increase in TNF-α is associated with a worse OS, and an increase in IL-6 is associated with a better OS.

[0113] Those skilled in the art can clearly understand that for the convenience and brevity of description, the specific working processes of the systems, devices, and units described above can refer to the corresponding processes in the foregoing method embodiments, and will not be elaborated herein.

[0114] In several embodiments provided in the present application, it should be understood that the disclosed systems, devices, and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of the units is only a logical function division. In actual implementation, there may be other division methods. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed coupling or direct coupling or communication connection between each other can be through some interfaces, and the indirect coupling or communication connection of the device or unit can be in electrical, mechanical, or other forms.

[0115] The units described as separate components may or may not be physically separated. The components displayed as units may or may not be physical units, that is, they may be located in one place or distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0116] In addition, in each embodiment of the present invention, the functional units can be integrated in a processing unit, or each unit can exist physically alone, or two or more units can be integrated in one unit. The above-mentioned integrated units can be implemented in the form of hardware or in the form of software functional units.

[0117] Those of ordinary skill in the art can understand that all or part of the steps in the various methods of the above embodiments can be completed by instructing relevant hardware through a program. The program can be stored in a computer-readable storage medium. The storage medium can include: read-only memory (ROM, Read Only Memory), random access memory (RAM, Random Access Memory), magnetic disk or optical disk, etc.

[0118] Those of ordinary skill in the art can understand that all or part of the steps in implementing the methods of the above embodiments can be completed by instructing relevant hardware through a program. The program can be stored in a computer-readable storage medium. The above-mentioned storage medium can be read-only memory, magnetic disk or optical disk, etc.

[0119] The above has introduced in detail a computer device provided by the present invention. For those of ordinary skill in the art, according to the idea of the embodiments of the present invention, there will be changes in the specific implementation manners and application scopes. In summary, the content of this specification should not be construed as a limitation to the present invention.

Claims

1. A method for predicting the prognosis of lung adenocarcinoma-derived meningeal carcinoma based on IL-6 and / or TNF-α, characterized in that: The method comprises: Obtain genetic information from cerebrospinal fluid samples from patients with meningeal carcinoma derived from lung adenocarcinoma; Extracting features from the genetic information to obtain genetic features and content data thereof, wherein the genetic features are IL-6 and / or TNF-α; Classification prediction is performed based on the content data of the genetic characteristics to obtain a classification result of whether the sample is a good prognosis sample or a poor prognosis sample.

2. The method for predicting the prognosis of lung adenocarcinoma-derived meningeal carcinoma based on IL-6 and / or TNF-α according to claim 1, characterized in that: The classification result is obtained based on a prediction model, and the method for constructing the prediction model includes: Obtaining the content data of the genetic features in the cerebrospinal fluid samples of the training set and the clinical features corresponding to the cerebrospinal fluid samples, wherein the clinical features are good prognosis or poor prognosis, extracting the content data of the genetic features in the training set and inputting them into the machine learning model to construct a prediction model, thereby obtaining a constructed prediction model and threshold; If the IL-6 content is higher than the threshold value and / or the TNF-α content is lower than the threshold value, a classification result is obtained that the sample is a sample with a good prognosis; If the content of IL-6 is lower than the threshold value and / or the content of TNF-α is higher than the threshold value, a classification result is obtained that the sample is a sample with poor prognosis.

3. The method for predicting the prognosis of lung adenocarcinoma-derived meningeal carcinoma based on IL-6 and / or TNF-α according to claim 2, characterized in that: The machine learning model is a Lasso regression model, a linear regression model, a logistic regression model, a Ridge regression model, a linear discriminant analysis model, a random forest model, a nearest neighbor model, a decision tree model, a support vector machine model, a naive Bayes model and / or a perceptron model.

4. The method for predicting the prognosis of lung adenocarcinoma-derived meningeal carcinoma based on IL-6 and / or TNF-α according to claim 1, characterized in that: The content data of the genetic characteristics are obtained by any one or more of the following methods: enzyme-linked immunosorbent assay, immunoblotting, real-time fluorescence quantitative PCR, flow cytometry, and liquid phase chip technology.

5. A prognosis prediction system for lung adenocarcinoma-derived meningeal carcinoma based on IL-6 and / or TNF-α, characterized in that: The system comprises: Genetic information acquisition unit: used to obtain genetic information of cerebrospinal fluid samples from patients with meningeal carcinoma derived from lung adenocarcinoma; Genetic feature extraction unit: used to extract features from the genetic information to obtain genetic features and content data thereof, wherein the genetic features are IL-6 and / or TNF-α; Disease prognosis prediction unit: used to perform classification prediction based on the content data of the genetic characteristics to obtain a classification result of whether the sample has a good prognosis or a poor prognosis; The classification result is obtained based on a prediction model, and the method for constructing the prediction model includes: Obtaining the content data of the genetic features in the cerebrospinal fluid samples of the training set and the clinical features corresponding to the cerebrospinal fluid samples, wherein the clinical features are good prognosis or poor prognosis, extracting the content data of the genetic features in the training set and inputting them into the machine learning model to construct a prediction model, thereby obtaining a constructed prediction model and threshold; If the IL-6 content is higher than the threshold value and / or the TNF-α content is lower than the threshold value, a classification result is obtained that the sample is a sample with a good prognosis; If the content of IL-6 is lower than the threshold value and / or the content of TNF-α is higher than the threshold value, a classification result is obtained that the sample is a sample with poor prognosis.

6. The prognosis prediction system for lung adenocarcinoma-derived meningeal carcinoma based on IL-6 and / or TNF-α according to claim 5, characterized in that: The machine learning model is a Lasso regression model, a linear regression model, a logistic regression model, a Ridge regression model, a linear discriminant analysis model, a random forest model, a nearest neighbor model, a decision tree model, a support vector machine model, a naive Bayes model and / or a perceptron model.

7. A device or electronic device for predicting the prognosis of meningeal cancer from lung adenocarcinoma, characterized in that: The device or electronic device comprises a memory and a processor; The memory is used to store program instructions; The processor is used to call program instructions, and when the program instructions are executed, the method for predicting the prognosis of lung adenocarcinoma-derived meningeal carcinoma based on IL-6 and / or TNF-α described in any one of claims 1 to 4 is implemented.

8. A computer-readable storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, the method for predicting the prognosis of lung adenocarcinoma-derived meningeal carcinoma based on IL-6 and / or TNF-α according to any one of claims 1 to 4 is implemented.

9. Use of reagents for detecting IL-6 and / or TNF-α levels in cerebrospinal fluid samples in the preparation of a product for predicting the prognosis of meningeal carcinoma derived from lung adenocarcinoma.

10. The use according to claim 9, characterized in that: The reagents include reagents for detecting the expression levels of IL-6 and / or TNF-α in cerebrospinal fluid samples using one or more of enzyme-linked immunosorbent assay, immunoblotting, real-time fluorescence quantitative PCR, flow cytometry, and liquid phase chip technology.

Images