WO2024014580A1

WO2024014580A1 - Method for providing information for artificial intelligence-based cancer diagnosis using biomarker expressed in exosome

Info

Publication number: WO2024014580A1
Application number: PCT/KR2022/010342
Authority: WO
Inventors: 이왕재; 이희원
Original assignee: 이왕재바이오연구소 주식회사
Priority date: 2022-07-13
Filing date: 2022-07-15
Publication date: 2024-01-18

Abstract

The present invention relates to a method for providing information for cancer diagnosis, the method comprising the steps of: isolating an exosome fraction from a biological sample isolated from an individual of interest; and measuring the expression level of a biomarker in the exosome fraction.

Description

Artificial intelligence-based cancer diagnosis information provision method using biomarkers expressed in exosomes

The present invention relates to a method for diagnosing cancer by analyzing circulating tumor cells, and more specifically, to a composition, diagnostic kit, and diagnostic method for diagnosing cancer using the gene methylation level of circulating tumor cells.

Cancer is one of the most common causes of death worldwide. Approximately 10 million new cases occur each year, accounting for approximately 12% of all deaths, making it the third most common cause of death.

One way to diagnose such cancer early is to use cancer markers such as antibodies. The diagnostic market using antibodies has been rapidly increasing since 1980, and is used in the development of highly efficient diagnostic kits and diagnostic techniques due to its excellent sensitivity as it can detect proteins specifically expressed according to diseases or symptoms even in very small amounts. . To achieve this, it is essential to develop antibodies that have both specificity and sensitivity to proteins (antigens) expressed by diseases and symptoms. In order to generate an antibody with specificity and sensitivity, several considerations can be taken to maximize the efficiency of the antibody. First, comparing the primary sequence of the antigen to be generated and selecting one with strong heterogeneity between the animal species of the antigen and the immune animal can maximize the immune response. For example, when using the sequence of a human protein, high-titer antibodies can be obtained by selecting an animal with high heterogeneity from the sequence and inducing an immune response. Second, the quality of the antigen may vary depending on the characteristics and three-dimensional structure of the antigen due to post-translational modifications in the sequence, so select the antigen in consideration. In particular, when using a partial peptide as an antigen rather than using the entire protein as an antigen, special consideration must be taken when selecting it.

Additionally, recently, methods for diagnosing cancer by measuring DNA methylation have been proposed. When a specific gene is hypermethylated, the expression of that gene is blocked (gene silencing). This is the main mechanism by which the function of a gene is lost in vivo without a mutation in the protein coding sequence of the gene, and the function of many tumor suppressor genes is lost in human cancer. It is interpreted as the cause. Therefore, methods that can be used in various fields such as cancer diagnosis and early diagnosis, prediction of carcinogenesis risk, prediction of cancer prognosis, follow-up after treatment, and prediction of response to anticancer therapy are being studied through methylation of tumor-related genes. There is controversy about the exact mechanism of carcinogenesis.

Meanwhile, exosomes, one of the extracellular vesicles, are nano-sized vesicles with the same composition as the outer cell wall. Exosomes were discovered about 30 years ago and were initially recognized as a mechanism for removing waste from cells. According to research over the past 10 years, it is known to play a functional role in mediating biologically important cell-cell communication and cellular immunity.

Accordingly, the present inventors developed a diagnostic method that can diagnose cancer with high accuracy by analyzing the methylation of genes in exosomes and completed the present invention.

Therefore, the present invention includes the steps of isolating an exosome fraction from a biological sample isolated from a subject of interest; The object of the present invention is to provide a method and device for providing information for diagnosing cancer, further comprising calculating the probability of cancer occurring in the subject using a learned deep neural network.

The problems to be solved by the present invention are not limited to the problems mentioned above, and other problems not mentioned can be clearly understood by those skilled in the art from the description below.

In order to achieve the above object, according to an embodiment of the present invention, a method of providing information for cancer diagnosis is provided.

Additionally, in the present invention, the method includes the steps of isolating an exosome fraction from a biological sample isolated from a subject of interest; And measuring the expression level of the biomarker in the exosome fraction.

In addition, in the present invention, the biomarkers include phospholipid transport ATPase IB, beta-1,3-galactosyltransferase 6, CUGBP Elav-like family member 6, leptin, glutathione S-transferase P1, and neural pentraxin. 2, P16, and high-affinity choline transporter 1.

Additionally, in the present invention, the method may further include calculating the likelihood of cancer occurring in the subject using a learned deep neural network.

Additionally, in the present invention, the deep neural network includes an input layer into which data is input; first hidden layer (hidden layer 1); second hidden layer (hidden layer 1); third hidden layer (hidden layer 1); fourth hidden layer (hidden layer 1); and an output layer.

According to another embodiment of the present invention, an information provision device for cancer diagnosis can be provided.

Additionally, in the present invention, the device includes a sample processing unit for isolating an exosome fraction from a biological sample isolated from a target individual; And it may include a measuring unit that measures methylation of nucleotide molecules in the exosome fraction.

Additionally, in the present invention, the device may further include an arithmetic unit that calculates the likelihood of cancer occurring in the individual using a learned deep neural network.

Cancer can be diagnosed through the method and device for providing information for cancer diagnosis according to the present invention, and whether a candidate substance can be used as a cancer treatment can be confirmed through a screening method.

The effects of the present invention are not limited to the effects mentioned above, and other effects not mentioned will be clearly understood by those skilled in the art from the description below.

Figure 1 shows a deep neural network with three hidden layers according to an embodiment of the present invention.

Figure 2 shows a deep neural network with four hidden layers according to an embodiment of the present invention.

Hereinafter, the present invention will be described in detail. Prior to this, the terms or words used in this specification and patent claims should not be construed as limited to their usual or dictionary meanings, and the inventor must appropriately use the concept of the term to explain his or her invention in the best way. It must be interpreted as meaning and concept consistent with the technical idea of the present invention based on the principle that it can be defined clearly. Accordingly, the configuration described in the embodiments described in this specification is only one of the most preferred embodiments of the present invention and does not represent the entire technical idea of the present invention, so at the time of filing this application, various equivalents and It should be understood that variations may exist.

Meanwhile, each description and embodiment disclosed in this specification may also be applied to each other description and embodiment. That is, all combinations of the various elements disclosed herein fall within the scope of the present application. Additionally, the scope of the present invention cannot be considered limited by the specific description described below.

Additionally, the terms used in this specification are merely used to describe specific embodiments and are not intended to limit the present invention. Singular expressions include plural expressions unless the context clearly dictates otherwise. In this application, terms such as “comprise” or “have” are intended to designate the presence of features, numbers, steps, operations, components, parts, or combinations thereof described in the specification, but are not intended to indicate the presence of one or more other features. It should be understood that this does not exclude in advance the possibility of the existence or addition of elements, numbers, steps, operations, components, parts, or combinations thereof.

When a component is said to be 'connected' or 'connected' to another component, it is understood that it may be directly connected or connected to the other component, but that other components may exist in between. It should be. On the other hand, when it is mentioned that a component is 'directly connected' or 'directly connected' to another component, it should be understood that there are no other components in between.

Additionally, in this specification, a device may be a general device (or object) connected to a gateway and applied to IoT (Internet of Things). Additionally, in this specification, device may be used interchangeably with 'equipment' or 'apparatus', and 'device', 'equipment', and 'apparatus' may be described with the same expression.

According to one embodiment of the present invention, a method for providing information for cancer diagnosis is provided.

In the present invention, “cancer” to “tumor” is a disease characterized by rapid and uncontrolled growth of mutant cells, and the cancer includes breast cancer, ovarian cancer, colon cancer, stomach cancer, liver cancer, pancreatic cancer, cervical cancer, thyroid cancer, Parathyroid cancer, lung cancer, non-small cell lung cancer, prostate cancer, gallbladder cancer, biliary tract cancer, non-Hodgkin lymphoma, Hodgkin lymphoma, blood cancer, bladder cancer, kidney cancer, melanoma, colon cancer, bone cancer, skin cancer, head and neck cancer, uterine cancer, rectal cancer, Brain tumor, perianal cancer, fallopian tube carcinoma, endometrial carcinoma, vaginal cancer, vulvar carcinoma, esophageal cancer, small intestine cancer, endocrine cancer, adrenal cancer, soft tissue sarcoma, urethral cancer, penile cancer, ureteral carcinoma, renal cell carcinoma, renal pelvic carcinoma, central It may be a nervous system tumor, spinal cord tumor, brainstem glioma, or pituitary adenoma.

In the present invention, the terms “marker” or “biomarker” are those that can detect changes in a living organism and objectively measure the normal or pathological state of the living organism, the degree of response to a drug, etc.

In the present invention, the term "diagnosis" refers to determining a subject's susceptibility to a specific disease or condition, determining whether the subject currently has a specific disease or condition, and determining whether the subject currently has a specific disease or condition. Predicting or determining the prognosis of an affected subject, or therametrics (e.g., monitoring the condition of a subject to provide information about treatment efficacy).

In the present invention, isolating an exosome fraction from a biological sample isolated from an individual of interest; And measuring the expression level of the biomarker in the exosome fraction.

In the present invention, the term "individual" refers to any living organism that develops or is likely to develop cancer, and specific examples include mammals including mice, monkeys, cows, pigs, mini-pigs, livestock, humans, etc., and farmed fish. It may include, but is not limited to, etc.

In the present invention, the terms “sample” and “biological sample” refer to any material, biological fluid, tissue, or cell derived from an individual that has developed or is suspected of having cancer. For example, the biological sample is whole blood. (whole blood), leukocytes, peripheral blood mononuclear cells, buffy coat, plasma, serum, sputum, tears, mucus ( mucus, nasal washes, nasal aspirate, breath, urine, semen, saliva, peritoneal washings, ascites, Cystic fluid, meningeal fluid, amniotic fluid, glandular fluid, pancreatic fluid, lymph fluid, pleural fluid, nipple aspirate ), bronchial aspirate, synovial fluid, joint aspirate, organ secretions, or cerebrospinal fluid.

In the present invention, the biomarkers include phospholipid transport ATPase IB, beta-1,3-galactosyltransferase 6, CUGBP Elav-like family member 6, leptin, glutathione S-transferase P1, and neuronal pentraxin 2. , P16, and high-affinity choline transporter 1, and the biomarker may be shown in Table 1 below.

서열번호sequence number	ProteinProtein	GeneGene
1One	인지질 수송 ATPase IB (Phospholipid-transporting ATPase IB)Phospholipid-transporting ATPase IB		ATP8A2ATP8A2
22	베타-1,3-갈락토실트랜스퍼라제 6 (BetaGal beta 1, 3-galactosyltransferase polypeptide 6)Beta-1,3-galactosyltransferase 6 (BetaGal beta 1, 3-galactosyltransferase polypeptide 6)	B3GALT6 B3GALT6
33	CUGBP 엘라브-유사 패밀리 멤버 6 (CUGBP Elav-like family member 6)CUGBP Elav-like family member 6	CELF6 CELF6
44	렙틴 (Leptin)Leptin	LEPLEP
55	글루타티온 S-트랜스퍼라제 P1 (Glutathione S-transferase P1)Glutathione S-transferase P1	GSTP1GSTP1
66	신경 펜트랙신 2(Neuronal Pentraxin 2) Neuronal Pentraxin 2	NPTX2 NPTX2
77	P16P16	CDKN2ACDKN2A
88	고친화성 콜린 수송체 1 (High affinity choline transporter 1)High affinity choline transporter 1	SLC5A7SLC5A7

The expression level of the biomarker may include an agent that measures the expression level of the protein represented by SEQ ID NO: 1 to 8 or the gene encoding the same, and the exosome is produced in the endosomal compartment of most eukaryotic cells. Extracellular vesicles (EV), and the isolation method may be included without limitation.

As an example, the measurement process involves obtaining serum samples from pancreatic cancer patients (n=50) and normal people (n=80) between the ages of 40 and 70, concentrating exosomes (extracellular vesicles) using a centrifugal filter unit, and then , it may be quantitatively detected by performing multiplexed immunoassay on the biomarkers of SEQ ID NOs: 1 to 8 expressed in exosomes.

In the present invention, the term "detection" may be used to measure and compare the presence or absence of a biomarker to be detected, concentration in a biological sample, expression level, etc., and an increase in the level of the biomarker as a result of the detection encodes the biomarker. increasing the number of genes that are activated, increasing gene transcription (e.g., by placing genes under the control of a constitutive promoter), increasing translation of genes, knocking out competing genes, or combinations of these and/or other methods. It can be done by a number of methods, and it can be measuring the expression level of exosomal biomarkers.

Agents for detecting the biomarkers of the present invention are not particularly limited, but include, for example, a group consisting of antibodies, oligopeptides, ligands, PNA (peptide nucleic acids) and aptamers that specifically bind to the biomarkers. It may include one or more selected types.

In the present invention, the “antibody” refers to a substance that specifically binds to an antigen and causes an antigen-antibody reaction. The antibodies of the present invention include polyclonal antibodies, monoclonal antibodies, and recombinant antibodies. The antibody can be easily produced using techniques well known in the art. For example, polyclonal antibodies can be produced by methods well known in the art, which include injecting the protein antigen into an animal and collecting blood from the animal to obtain serum containing the antibody. These polyclonal antibodies can be produced from any animal, such as goats, rabbits, sheep, monkeys, horses, pigs, cows, dogs, etc. Additionally, monoclonal antibodies can be produced using hybridoma methods or phage antibody library technology well known in the art. Antibodies prepared by the above method can be separated and purified using methods such as gel electrophoresis, dialysis, salt precipitation, ion exchange chromatography, and affinity chromatography. Additionally, antibodies of the invention include intact forms with two full-length light chains and two full-length heavy chains, as well as functional fragments of the antibody molecule. A functional fragment of an antibody molecule refers to a fragment that possesses at least an antigen-binding function, and includes Fab, F(ab'), F(ab')2, and Fv.

In the present invention, "Peptide Nucleic Acid (PNA)" refers to an artificially synthesized polymer similar to DNA or RNA, and was first introduced by Professors Nielsen, Egholm, Berg and Buchardt at the University of Copenhagen, Denmark in 1991. While DNA has a phosphate-ribose sugar backbone, PNA has a repeated N-(2-aminoethyl)-glycine backbone linked by peptide bonds, which greatly increases its binding force and stability to DNA or RNA and is used in molecular biology. , is used in diagnostic analysis and antisense therapy. Additionally, the “aptamer” is an oligonucleic acid or peptide molecule.

In the present invention, the agent for measuring the expression level of genes encoding the biomarkers may include one or more selected from the group consisting of primers, probes, and antisense nucleotides that specifically bind to the genes.

In the present invention, the “primer” is a fragment that recognizes the target gene sequence and includes forward and reverse primer pairs, but is preferably a primer pair that provides analysis results with specificity and sensitivity. High specificity can be granted when the nucleic acid sequence of the primer is a sequence that is inconsistent with the non-target sequence present in the sample, so that the primer amplifies only the target gene sequence containing the complementary primer binding site and does not cause non-specific amplification. .

In the present invention, the “probe” refers to a substance that can specifically bind to a target substance to be detected in a sample, and refers to a substance that can specifically confirm the presence of the target substance in the sample through the binding. The type of probe is not limited as it is a material commonly used in the art, but is preferably PNA (peptide nucleic acid), LNA (locked nucleic acid), peptide, polypeptide, protein, RNA or DNA, and most preferably is PNA. More specifically, the probe is a biomaterial that is derived from or similar to living organisms or includes those manufactured in vitro, such as enzymes, proteins, antibodies, microorganisms, animal and plant cells and organs, nerve cells, DNA, and It may be RNA, DNA includes cDNA, genomic DNA, and oligonucleotides, RNA includes genomic RNA, mRNA, and oligonucleotides, and examples of proteins may include antibodies, antigens, enzymes, peptides, etc.

In the present invention, “LNA (Locked nucleic acids)” refers to nucleic acid analogs containing 2'-O, 4'-C methylene bridges. LNA nucleosides contain the common nucleic acid bases of DNA and RNA and can form base pairs according to the Watson-Crick base pairing rules. However, due to the 'locking' of the molecule due to the methylene bridge, LNA does not form the ideal shape in Watson-Crick bonding. When LNA is included in a DNA or RNA oligonucleotide, the LNA can pair with the complementary nucleotide chain more quickly and increase the stability of the double helix. In the present invention, the "antisense" refers to a sequence of nucleotide bases in which an antisense oligomer hybridizes with a target sequence in RNA by Watson-Crick base pairing, typically allowing the formation of an mRNA and RNA:oligomer heteroduplex within the target sequence. and oligomers having an intersubunit backbone. Oligomers may have exact or approximate sequence complementarity to the target sequence.

In the present invention, the method may further include calculating the likelihood of cancer occurring in the individual using a learned deep neural network.

In the present invention, the calculation may be to divide the individual into a risk group with a high possibility of developing cancer and a control group. The “control group” may refer to a general population that does not develop cancer, an entire population of subjects that develop cancer, or a population that shows a good prognosis among subjects that develop cancer.

In the present invention, the deep neural network is a type of artificial neural network and includes a convolution layer, a pooling layer, a ReLu layer, and a transpose layer that allows deep learning learning. Convolutional layer, Unpooling layer, 1x1 convolutional layer, Skip connection, Global Average Pooling (GAP) layer, Fully Connected layer, SVM (support Vector Machine), LSTM (long short term memory) , It may be a neural network including one or more layers or elements selected from the group consisting of Atrous Convolution, Atrous Spatial Pyramid Pooling, Separable Convolution, and Bilinear Upsampling, It may further include a batch normalization operation at the front end. For example, as shown in FIG. 1, the deep neural network includes an input layer into which data is input; first hidden layer (hidden layer 1); second hidden layer (hidden layer 1); third hidden layer (hidden layer 1); fourth hidden layer (hidden layer 1); and an output layer. At this time, the expression level of the biomarker of the present invention can be input into the input layer.

In the present invention, the deep neural network may be learned by a backpropagation learning algorithm, and the learning data may include data on the expression level of the biomarker of the present invention and result data on whether cancer actually occurs. . The learning data is data input for learning of a deep neural network, and for learning, information about the error between the predicted value obtained by processing the information by the deep neural network and the actual result is required. Accordingly, the learning data must necessarily include actual results. In addition, the learning data may require patient-specific data above a certain standard value to ensure that the weight coefficients of the deep neural network are derived at a sufficiently reliable level. Thereafter, the diagnostic device may perform an operation of generating an initial prediction value that predicts the probability of developing cancer using a deep neural network. Thereafter, the training unit may perform an operation to modify the weight coefficient based on the error between the derived initial prediction value and the actual result value input as learning data. The operation may be an operation using a backpropagation algorithm. For example, as shown in FIG. 1, each node of the first to third hidden layers may receive a value that is a result of calculation by all nodes and weights of the input layer. Alternatively, as shown in FIG. 1, each node of the first to fourth hidden layers may receive a result calculated by all nodes and weights of the input layer. Afterwards, each node of the hidden layers can determine whether or not it is activated using an activation function. Each node in the hidden layer is determined to be activated when the corresponding value satisfies a certain standard or deactivated when it does not meet a certain standard, and this is performed based on the activation function. Additionally, in the process of connecting each node of the hidden layer to the positive and negative nodes of the output layer, corresponding weight coefficients may exist. Therefore, if the weights in the deep neural network can be set differently, the weight coefficients applied at the stage between the input layer and the output layer can be set in various ways. For example, a neural network with first to third hidden layers as shown in FIG. 1 may have a maximum of 5,184 weight coefficients, and a neural network with first to fourth hidden layers as shown in FIG. 2 may have a maximum of 46,656 weight coefficients. You can.

Through this process, the training unit modifies the weights starting from the weight coefficients applied at the stage closest to the output layer, and when all inputs of the prepared learning data are completed, the final calculated coefficients are fixed as the weight coefficients of the deep neural network to obtain the learned deep layer. The neural network model can be completed.

In the present invention, the data for learning the deep neural network may further include one or more selected from the group consisting of the individual's weight, gender, and age, which are risk factors other than the biomarker.

According to another embodiment of the present invention, an apparatus for diagnosing cancer can be provided.

In the present invention, the device includes a sample processing unit for isolating an exosome fraction from a biological sample isolated from a subject of interest; And it may include a measuring unit that measures methylation of nucleotide molecules in the exosome fraction.

In the present invention, the device may further include a calculation unit that calculates the likelihood of cancer occurring in the individual using a learned deep neural network.

In the present invention, the device may further include a training unit that receives learning data and performs learning. At this time, the learning data may be individually input by the user, or may be input collectively from patient data separately stored in a server computer of a medical institution.

The training unit according to an embodiment of the present invention may directly receive data corresponding to learning data (training cohort) or test data (test cohort). Additionally, according to various embodiments, the training unit may receive a user manipulation to change an item of data to be input. In addition, patient data according to various embodiments may be input into the training unit, and the data may be in digital form.

In the present invention, the device may further include a storage unit. The storage unit may include, for example, internal memory or external memory. Internal memory includes, for example, volatile memory (e.g., dynamic RAM (DRAM), static RAM (SRAM), or synchronous dynamic RAM (SDRAM), etc.), non-volatile memory (e.g., OTPROM (one time programmable ROM), programmable ROM (PROM), erasable and programmable ROM (EPROM), electrically erasable and programmable ROM (EEPROM), mask ROM, flash ROM, flash memory (such as NAND flash or NOR flash, etc.), hard drive, Or it may include at least one of a solid state drive (SSD).The external memory may be a flash drive, for example, compact flash (CF), secure digital (SD), or Micro-SD. It may further include (micro secure digital), Mini-SD (mini secure digital), xD (extreme digital), MMC (multi-media card), or memory stick, etc. External memory can be stored through various interfaces. May be functionally and/or physically connected to an electronic device.

In the present invention, the patient data may be extracted by a reader from only data corresponding to risk factors from information collected from other electronic devices or external servers (e.g., related medical institution servers). For example, information transmitted through the communication unit may include information that does not correspond to test results, such as the patient's name, date of birth, and smoking status, and result information for a specific test. However, since there may be items unrelated to cancer prediction among these many various items, the patient data can be input to the training unit to extract only the data items required to perform a cancer diagnosis.

In the present invention, the deep neural network can compare the values of positive and negative nodes and calculate the one with the larger value as the final result. For example, the comparison may compare the size of the difference between the values of the positive node and the negative node, or if it is determined that the positive value is greater than or equal to the reference value or the negative value is less than the reference value, the corresponding probability of occurrence of cancer may be calculated.

In the present invention, the calculation unit can calculate the most effective risk factors to reduce the probability of cancer occurrence based on the deep neural network. The calculation unit may generate comparison data by changing the value of a specific item selected from each risk factor data at regular intervals. Thereafter, the electronic device can compare the predicted value generated by inputting the raw data into the deep neural network with the predicted value generated by inputting the comparison data into the deep neural network, and if the predicted value (e.g., positive) generated by the raw data is compared. If the predicted values (e.g., negative) generated by the data are calculated differently, data values for each risk factor item for which the two values begin to be calculated differently can be calculated and provided to the user.

In the present invention, continuous data can be expressed as mean and standard deviation (SD), and cancer incidence factors can be determined from retrospective data using a model of the training cohort.

In the present invention, the function of displaying data input to the device and results calculated as a result of the operation of a deep neural network can be performed. Additionally, the display unit may calculate data regarding the probability of a result processed by a deep neural network according to various embodiments. For example, according to various embodiments, the deep neural network not only predicts whether cancer will occur, but also provides information about the probability of cancer occurring if it will occur, or the probability of cancer not occurring if it is predicted not to occur. can be calculated together.

In the present invention, the display unit may include a panel, a hologram device, or a projector. The panel may be composed of a touch panel and one module. Holographic devices can display three-dimensional images in the air using the interference of light. A projector can display images by projecting light onto a screen. The screen may be located, for example, inside or outside the electronic device. According to one embodiment, the display unit may further include a control circuit for controlling a panel, a hologram device, or a projector.

In the present invention, the device may further include a communication unit. The communication unit can use a network to transmit and receive data with other user electronic devices or other servers, and the type of the network is not particularly limited. For example, the network may be an Internet Protocol (IP) network that provides large data transmission and reception services through the Internet Protocol (IP), or an All IP network that integrates different IP networks. In addition, the network includes a wired network, a Wibro (Wireless Broadband) network, a mobile communication network including WCDMA, a mobile communication network including a HSDPA (High Speed Downlink Packet Access) network and an LTE (Long Term Evolution) network, and LTE advanced (LTE-A). ), a mobile communication network including 5G (Five Generation), a satellite communication network, and a Wi-Fi network, or a combination of at least one of these. Additionally, the communication unit may support a communication function for receiving methylation information to be input into a deep neural network learned from another electronic device or an external server. Additionally, the communication unit may transmit the information processing results of the deep neural network to another electronic device or external server.

In the present invention, the device may further include a control unit, and the control unit may also be called a processor, controller, microcontroller, microprocessor, microcomputer, etc. You can. Meanwhile, the control unit may be implemented by hardware, firmware, software, or a combination thereof.

In the present invention, in the case of implementation by firmware or software, an embodiment of the present invention may be implemented in the form of a module, procedure, function, etc. that performs the functions or operations described above. Software code can be stored in memory and driven by a control unit. The memory may be located inside or outside the user terminal and server, and may exchange data with the control unit through various known means.

In the present invention, the remaining omitted descriptions may be interpreted similarly to the remaining descriptions in the specification.

According to another embodiment of the present invention, a screening method for a cancer treatment drug can be provided.

In the present invention, the steps include treating a sample isolated from a target individual or a cancer disease animal model with a candidate substance; and confirming the expression level of the biomarker in a sample treated with the candidate substance or a cancer disease animal model. It may be to provide a method of screening a drug for the prevention or treatment of cancer, including: " “Candidate substance” may include without limitation substances that can improve or beneficially change the prognosis when applied to cancer patients, and the candidate substance may include synthetic substances as well as natural substances, but is not limited thereto.

In the present invention, the expression level may be the expression level in exosomes of the sample derived from the model.

In the present invention, the measurement of the expression level may be performed multiple times.

In the present invention, the remaining omitted descriptions can be interpreted similarly to the remaining descriptions in the specification.

Although the present invention has been described in detail above, the scope of the present invention is not limited thereto, and it is known in the art that various modifications and variations are possible without departing from the technical spirit of the present invention as set forth in the claims. It will be self-evident to those with knowledge.

Claims

As a method of providing information for cancer diagnosis,

Isolating an exosome fraction from a biological sample isolated from a subject of interest; and

Measuring the expression level of the biomarker in the exosome fraction,

method.
According to paragraph 1,

The biomarkers include phospholipid transport ATPase IB, beta-1,3-galactosyltransferase 6, CUGBP elabe-like family member 6, leptin, glutathione S-transferase P1, neuronal pentraxin 2, P16, and Gochin. Containing at least one selected from the group consisting of Martian choline transporter 1,

method.
According to paragraph 1,

The method further includes calculating the likelihood of cancer occurring in the subject using a learned deep neural network.

method.
According to paragraph 3,

The deep neural network includes an input layer into which data is input; first hidden layer (hidden layer 1); second hidden layer (hidden layer 1); third hidden layer (hidden layer 1); fourth hidden layer (hidden layer 1); and an output layer;

method.
As a device for cancer diagnosis,

A sample processing unit for isolating an exosome fraction from a biological sample isolated from a target individual; and

It includes a measuring unit that measures methylation of nucleotide molecules in the exosome fraction,

Device.
According to clause 5,

The device further includes an arithmetic unit that calculates the likelihood of cancer occurring in the individual using a learned deep neural network.

Device.