Method for guiding accurate medication by using tumor organ-like chip and application thereof

By constructing a tumor organoid chip with vascularized co-culture, simulating the tumor microenvironment in vivo, drug sensitivity detection and omic correlation analysis are carried out, and drug sensitivity model is generated, which solves the problem of insufficient accuracy of drug sensitivity prediction in the existing technology. , improves clinical treatment effect.

CN119061106APending Publication Date: 2024-12-03SOUTHEAST UNIV NANJING INST OF BIOMATERIALS & MEDICAL DEVICES
View PDF 1 Cites 0 Cited by

Patent Information

Application Number
CN202411314963.9
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-09-20
Publication Date
2024-12-03

AI Technical Summary

Technical Problem

The accuracy and prediction effect of existing drug sensitivity prediction models need to be improved in clinical applications. Traditional in vitro drug sensitivity detection cannot fully simulate the tumor microenvironment in vivo, resulting in inconsistent drug screening results with clinical efficacy.

Method used

The tumor organoid chip with vascularized co-culture was used to simulate the tumor microenvironment in vivo, drug sensitivity was detected, and a mathematical model of drug sensitivity was generated through omic correlation analysis to guide individual drug use.

Benefits of technology

More accurate drug sensitivity detection data are provided, which improves the accuracy of drug sensitivity prediction and clinical treatment effect, and overcomes the problem of inconsistent drug screening results and clinical efficacy in the prior art.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119061106A_ABST
    Figure CN119061106A_ABST
Patent Text Reader

Abstract

The invention relates to the technical field of organ chip application, and relates to a method for guiding accurate medication by using a tumor organ chip, and the method comprises the following specific steps: S1, constructing a vascularized co-culture organ chip; s2, performing drug sensitivity detection aiming at an anti-tumor drug combination scheme; s3, for the screened optimal drug combination, selecting a biomarker related to the sensitivity of the drug combination; s4, constructing a drug sensitivity prediction model according to the screened gene mutation and gene abnormal expression significantly related to the drug sensitivity; and S5, performing drug sensitivity evaluation on the patient according to the drug sensitivity prediction model, and taking an evaluation result as a reference basis for individual medication guidance. The constructed method and the drug sensitivity prediction model can have the advantages of tumor microenvironment simulation and high-throughput detection at the same time; and a basis can be provided for individual medication guidance of drugs.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of organ chip application technology, and in particular to a method for guiding precise medication using a tumor organ chip and its application. Background Art

[0002] With the development of molecular biology and precision medicine, personalized treatment has become an important concept in tumor treatment. Through molecular typing, genetic testing and other means, doctors can choose the most appropriate treatment plan based on the patient's specific tumor characteristics. However, although these technologies have improved the treatment effect to a certain extent, in clinical practice, the efficacy of anti-tumor drugs is still affected by many factors, including tumor heterogeneity and drug resistance. In addition, due to the complexity of tumors and individual differences among patients, the current drug sensitivity prediction model still faces great challenges in practical application, and the accuracy and prediction effect need to be improved.

[0003] Existing clinical guidance for drug use mainly relies on in vitro drug sensitivity testing and genomic analysis technology. However, traditional in vitro drug sensitivity testing (such as cell lines, animal models, etc.) has certain limitations and cannot fully simulate the in vivo tumor microenvironment, resulting in inconsistencies between drug screening results and clinical efficacy. In addition, although genomic analysis has certain advantages in guiding drug use, relying solely on genetic information for drug sensitivity prediction often cannot fully reflect the response of tumor cells to drugs. How to more accurately predict the sensitivity of individual patients to anti-tumor drugs and thereby improve clinical treatment effects has become an urgent problem to be solved in the current field of tumor treatment. Summary of the invention

[0004] The purpose of the present invention is to address the problems existing in the background technology and propose a method and application of using tumor organoid chips to guide precision drug use. This method can simulate the in vivo tumor microenvironment by constructing a tumor organoid chip with vascularized co-culture, provide more accurate drug sensitivity detection data, perform omics association analysis based on drug sensitivity differences, and generate a mathematical model for predicting drug sensitivity to guide individual medication.

[0005] The technical solution of the present invention is a method for guiding precise drug use using a tumor organoid chip, comprising the following specific steps:

[0006] S1. Construction of vascularized co-culture organ chips;

[0007] S11, Organoid preparation and seeding;

[0008] S111. Sample collection: Collect multiple different human tumor tissue samples to maintain tissue activity;

[0009] S112, tissue digestion: cut the tumor tissue into small pieces, digest it in a digestive enzyme solution, and filter and collect the cell suspension;

[0010] S113, cell culture: mix the transfected cell suspension with matrix gel, add it dropwise to the culture plate, and add organoid culture medium for culture after solidification;

[0011] S114, initial culture: culture in organoid culture medium for 7-14 days, during which time the culture medium is changed;

[0012] S115, subculture: subculture by mechanical or chemical methods, and collect for later use;

[0013] S116. Organoid inoculation: Take out the prepared blank organ chip, inoculate the organoid sample mixed with matrix gel into the reserved chip hole / channel, and place it in a cell culture incubator to wait for the matrix gel to solidify;

[0014] S12, construction of vascularized microenvironment;

[0015] S121. Cell culture: Take frozen human vascular endothelial cells, resuscitate, resuspend and centrifuge them, transfer them into cell culture bottles, and place them in a cell culture incubator for standby use;

[0016] S122. Cell inoculation: Take out the organ chip inoculated with organoids, inoculate the human vascular endothelial cell suspension into the reserved chip holes / channels, and place it on a rocking perfusion instrument in a cell culture incubator for culture to ensure that the vascular endothelium forms a dense lumen;

[0017] S123. Model verification: Organ chip modules were randomly selected to verify the density of vascularized channels and characterize the barrier function, proving that the constructed vascularized lumen had good density and barrier function and could be used as a drug delivery pathway for drug testing;

[0018] S124. Vascular network: After the vascular lumen is dense, vascular growth factors are added into the channel to stimulate endothelial cells to penetrate into the matrix gel mixed with organoid samples to construct an organoid vascularized microenvironment.

[0019] S2. Conduct drug sensitivity testing for anti-tumor combination drug regimens;

[0020] S21. Group the test drugs;

[0021] In S21, the drug regimen of the experimental group is a drug that inhibits tumor growth. A blank group is set up at the same time. The dosage is designed in a gradient based on the maximum blood drug concentration Cmax value of the reference drug. The number of drug combinations to be tested is allocated based on the single-chip flux number of the organ chip and the number of repetitions in each experimental group.

[0022] S22, using organ chips for drug sensitivity testing;

[0023] S221. Before drug administration, perform preliminary high-content image recognition on the constructed vascularized co-culture organ chip;

[0024] S222. In accordance with the designed dosing sequence and frequency, the drug regimen is injected into the reserved vascularized lumen channel, and after the drug injection is completed, the cell is placed on a rocking perfusion apparatus in a cell culture incubator for drug intervention culture;

[0025] S223. During the drug administration period, high-content image recognition is performed regularly on the organ chip cultured with drug intervention;

[0026] S23. Conduct drug sensitivity assessment on the organ chip.

[0027] S231. Organoid area: Use a high-content imaging analyzer to take images of organoids, collect image data and analyze the area change ratio (OAC) of the organoids to determine the sensitivity of the drug;

[0028] Endpoint OAC formula:

[0029] OAC1(%)=(1-ROA T / ROA C )*100%

[0030] ROA T =(T t -T0) / T0

[0031] ROA C =(C t -C0) / C0

[0032] T0: identification area of ​​the experimental / control group organoids when the drug was administered; T t : recognition area of ​​the experimental / control group organoids at time t; C0: recognition area of ​​the blank group organoids when drug administration; C t : The recognition area of ​​the blank group organoids at time t

[0033] When OAC1 ≥ 60%, it is judged as drug sensitivity positive; when OAC1 < 60%, it is judged as drug sensitivity negative;

[0034] Autologous OAC formula:

[0035] OAC2(%)=(A t -A0) / A0*100%

[0036] A0: identification area of ​​each group of organoids when administered, A t : The identification area of ​​each group of organoids at time t;

[0037] When OAC2 is less than 20%, it is judged as drug sensitivity positive; when OAC2 is greater than or equal to 20%, it is judged as drug sensitivity negative;

[0038] Comprehensive evaluation: When OAC1≥60% and OAC2<20%, ​​the drug is judged to be effective; when OAC1<60% and OAC2≥20%, the drug is judged to be ineffective; other situations are judged to be potentially sensitive;

[0039] S232. Compare and analyze the changes in organoid area before and after the administration cycle. When OAC1 ≥ 60% and OAC2 < 20%, it is judged as drug effective P; when OAC1 < 60% and OAC2 ≥ 20%, it is judged as drug ineffective N; other cases are judged as intermediate sensitivity M;

[0040] S3. For the optimal drug combination that has been screened, select biomarkers related to the sensitivity of the drug combination;

[0041] S31. Data collection;

[0042] S311, gene mutation data: obtaining gene mutation and copy number variation data of the organoid sample of the drug sensitivity test result in step S2;

[0043] S312, gene expression data: obtaining gene expression data of the organoid sample of the drug sensitivity test result in step S2;

[0044] S313, drug sensitivity data: obtaining the inhibitory effect judgment information of the organoid sample of the drug sensitivity test result in step S2 through high-throughput detection of the organ chip;

[0045] S32, data preprocessing;

[0046] S321. Gene mutation data: All somatic and germline mutations were detected using general bioinformatics software, and then somatic mutations were obtained through filtering. These mutations included tumor-driving mutations and non-tumor-driving mutations, and the genotype data were standardized.

[0047] S322, Gene expression data: filter low-quality gene expression data and standardize the gene expression data;

[0048] S323, drug sensitivity data: filter low-quality drug sensitivity data. If the judgment results of OAC1 and OAC2 are both positive for drug sensitivity, it is considered that the drug is effective P; if the judgment results of OAC1 and OAC2 are both negative for drug sensitivity, it is considered that the drug is ineffective N;

[0049] S33, correlation analysis;

[0050] S331. Using the chi-square exact test, the distribution of gene mutations and abnormal gene expressions in the drug-susceptibility negative group and the drug-susceptibility positive group was compared, and the p value < 0.05 was used as the threshold to screen out gene mutations and abnormal gene expressions with strong correlation;

[0051] S332, using the Spearman correlation test, with a p value of <0.05 as the threshold, confirm the gene mutations and abnormal gene expressions with strong correlation;

[0052] S333, Functional Annotation: Functional prediction and annotation of genes with potential functional impact screened out by analysis, data from STRING database;

[0053] S334, Pathway enrichment analysis: Use bioinformatics databases (such as KEGG, Reactome) to perform pathway enrichment analysis to identify affected biological pathways and functional modules, further explain the biological connection between genes and drug sensitivity, and provide a basis for guiding personalized drug treatment;

[0054] S335. Gene mutations with significant linkage disequilibrium were excluded using Plink software.

[0055] S4. Construct a drug sensitivity prediction model based on the gene mutations and abnormal gene expressions that are significantly associated with drug sensitivity;

[0056] S41. Construct a drug sensitivity prediction model based on the gene mutations and abnormal gene expressions that are significantly associated with drug sensitivity: establish a multivariate linear regression model using the linear regression method, where each gene mutation and abnormal gene expression can be considered as a "meta";

[0057] S42. Introduce a constraint optimization penalty factor as a weight coefficient of the element to correct the deviation. The specific constraint optimization penalty factor can be confirmed by verifying the positive prediction value and ROC curve of the overall model. When the positive prediction value reaches the maximum or the area under the ROC curve is the largest, the element constraint optimization penalty factor used in the model is established; gene mutations and abnormal gene expressions with correlation coefficients close to zero in the linear model can be excluded;

[0058] S43, selecting the omics information and drug sensitivity data of the samples not used in constructing the drug sensitivity prediction model for secondary verification, so as to finally optimize the parameters of the drug sensitivity prediction model and generate a drug sensitivity detection analysis model;

[0059] S5. Evaluate the drug sensitivity of patients according to the drug sensitivity prediction model, and use the evaluation results as a reference for individual medication guidance;

[0060] S51. Sample processing: Based on the specific site information of the mutant genes and abnormally expressed genes involved in the drug sensitivity prediction model, conventional PCR and qPCR of specific genes are performed on the tumor tissue genome of the subject;

[0061] S52. Sample prediction: determine the mutation or expression of the corresponding gene loci of the subject's sample, and evaluate the specific gene mutation or abnormal expression information of the subject's sample in combination with the drug sensitivity prediction software;

[0062] S53. Drug sensitivity assessment: The comprehensive score is calculated through the drug sensitivity test prediction model, which is divided into high sensitivity and low sensitivity, indicating the sensitivity of the subject to the drug, which will serve as a reference for individual medication guidance.

[0063] An application of a tumor organoid chip uses the above method to construct a drug sensitivity prediction model, calculates a comprehensive score through the drug sensitivity prediction model, evaluates the sensitivity of the subject to the drug, and uses it as a reference for individual medication guidance.

[0064] Compared with the prior art, the present invention has the following beneficial technical effects:

[0065] Most existing drug sensitivity prediction models are based on two-dimensional cell line data, lacking consideration of the tumor microenvironment and individual differences among patients, resulting in poor prediction results of the models. Traditional in vitro cell line experiments cannot fully simulate the heterogeneity and microenvironment of tumors, resulting in significant differences between drug screening results and clinical efficacy. The present invention, by constructing a tumor organoid chip with vascularized co-culture, can simulate the in vivo tumor microenvironment, provide more accurate drug sensitivity detection data, perform omics association analysis based on drug sensitivity differences, and generate a mathematical model for predicting drug sensitivity to guide individual medication. BRIEF DESCRIPTION OF THE DRAWINGS

[0066] Figure 1 This is a flow chart of a method for guiding medication using an organ chip in an embodiment of the present invention;

[0067] Figure 2 This is a schematic diagram of the process of individual guided medication in an embodiment of the present invention;

[0068] Figure 3 Schematic diagram of the structure of the organ chip in an embodiment of the present invention. DETAILED DESCRIPTION

[0069] Example 1

[0070] The present invention proposes a method for guiding precise medication using a tumor organoid chip:

[0071] Step 1: Construction of vascularized co-culture organ-on-a-chip

[0072] The organ chip based on the present invention is a membrane-free barrier chip constructed using Phaseguide technology. The overall appearance is a 384-well plate. A chip plate contains 40 separate chips, each of which consists of three channels: upper, middle and lower. Its structure is as follows: Figure 3 As shown: the specific structure has been disclosed in the Chinese patent publication number CN118421563A, and will not be repeated here.

[0073] (1) Organoid preparation and seeding

[0074] a) Sample collection: Collect multiple different human tumor tissue samples to maintain tissue activity;

[0075] b) Tissue digestion: Cut the tumor tissue into small pieces, digest it in a digestive enzyme solution, and filter and collect the cell suspension;

[0076] c) Cell culture: The transfected cell suspension is mixed with matrix gel and added dropwise to the culture plate. After solidification, organoid culture medium is added for culture.

[0077] d) Initial culture: Culture in organoid culture medium for 7-14 days, during which the culture medium is changed;

[0078] e) Subculture: subculture by mechanical or chemical methods and collect for future use;

[0079] f) Organoid inoculation: Take out the prepared blank organ chip, inoculate the organoid sample mixed with matrix gel into the reserved chip hole / channel, and place it in a cell culture incubator to wait for the matrix gel to solidify.

[0080] Specific operations:

[0081] 1) Take out the organoids from the CO2 incubator and move them to the operating table.

[0082] 2) Discard the culture medium in each well.

[0083] 3) Add an appropriate amount of 4°C pre-cooled PBS to each well and let stand for 1 min.

[0084] 4) Blow away the matrix gel in each well, then transfer the liquid in the well into a 15 mL centrifuge tube and centrifuge at 300 x g at 4°C for 5 min.

[0085] 5) After centrifugation, the tube is divided into three layers. Use a pipette to discard the middle and upper layers and keep the cell layer;

[0086] 6) Resuspend the cells in 1-2 mL of digestion solution, pipette and mix well, and place the centrifuge tube in a 37°C water bath for oscillation digestion for 1-3 minutes;

[0087] 7) After digestion is completed, determine whether digestion is complete under a microscope.

[0088] 8) Use a pipette to draw PBS three times the volume of the digestion solution, add it to the centrifuge tube, pipette and mix to stop the digestion, and centrifuge at 300xg, 4℃ for 5 minutes.

[0089] 9) After centrifugation, discard the supernatant, resuspend with 1 mL of PBS, and centrifuge at 300 x g, 4°C for 5 min.

[0090] 10) After centrifugation, discard the supernatant, resuspend the precipitate in matrix gel in an ice bath to an appropriate density, aspirate 1-2 μL, add to the B1 or B3 inoculation well, and then place in a carbon dioxide incubator and let stand for 20 minutes to allow the gel to solidify.

[0091] (2) Construction of vascularized microenvironment

[0092] a) Cell culture: Take frozen human vascular endothelial cells, resuspension, centrifugation, transfer into cell culture bottles, and place in a cell culture incubator for standby use;

[0093] b) Cell inoculation: Take out the organ chip inoculated with organoids, inoculate the human vascular endothelial cell suspension (mixed cell culture medium) into the reserved chip holes / channels, and place it on a rocking perfusion instrument in a cell culture incubator for culture to ensure that the vascular endothelium forms a dense lumen;

[0094] Specific operations:

[0095] 1) While waiting for the gel to solidify, collect endothelial cells (T25): discard the cell culture medium, wash each bottle of cells twice with PBS, then add 2 mL of digestion solution for digestion, and quickly add 2 mL of the corresponding culture medium of equal volume to terminate the digestion.

[0096] 2) After terminating the digestion, blow the cells with a pipette tip with appropriate force to blow the cells off the surface as completely as possible, then collect the liquid into a 15 mL centrifuge tube and centrifuge at 300xg for 5 minutes.

[0097] 3) After centrifugation, discard the supernatant and resuspend the endothelial cells in co-culture medium to a density of approximately 1x107 to 2x107 cells / mL.

[0098] 4) Take out the solidified chip plate, pipette 2 μL of endothelial cell suspension into well A1 or A3, and then tilt the chip 90° from A to C with channel B as the axis for about 1 hour to allow the cells to adhere to the matrix gel side, and then add culture medium (no more than 40 μL per well), and add an equal amount of culture medium to channel C.

[0099] 5) Place the chip on a rocking perfusion instrument in an incubator for culture. The rocking direction is up and down (A to C and C to A), the angle is set to 7° to 10°, and the rocking period is 15 to 30 minutes.

[0100] 6) After the vascular endothelium of pathway A forms a dense lumen, it can be used for drug testing.

[0101] c) Model validation: Organ chip modules were randomly selected to verify the density of the vascularized channel and characterize the barrier function, proving that the constructed vascularized lumen had good density and barrier function and could be used as a drug delivery pathway for drug testing.

[0102] d) Vascular network: After the vascular lumen is dense, vascular growth factors are added into the channel to stimulate endothelial cells to penetrate into the matrix gel mixed with organoid samples to construct an organoid vascularized microenvironment.

[0103] Step 2: Drug sensitivity test for anti-tumor combination therapy

[0104] (1) Drug regimen grouping

[0105] a) The drug regimen of the experimental group is a drug that inhibits tumor growth. A blank group is also set up. The dosage is designed in a gradient based on the maximum blood drug concentration (Cmax value) of the drug. The number of drug combinations to be tested is allocated based on the single-chip flux number of the organ chip and the number of repetitions in each experimental group.

[0106] b) The dosage is designed in a gradient based on the maximum blood drug concentration (Cmax value) of the drug, and the number of drug combinations to be tested is allocated based on the single-chip flux number of the organ chip and the number of repetitions in each test group.

[0107] Concentration gradient: 0.1Cmax, 0.3Cmax, Cmax, 3Cmax, 10Cmax

[0108] (2) Organ Chip Drug Sensitivity Test

[0109] a) Before drug administration, the constructed vascularized co-culture organ chip was preliminarily imaged with high content images, and a random portion was selected for initial fluorescence quantitative staining identification, including the seeded organoids and vascular networks;

[0110] The details are as follows:

[0111] 1) Use a high-content imaging analyzer to locate, photograph, and scan the inoculated organoids and vascular networks in all groups to collect initial imaging data; after photography, spray the surface with 75% alcohol to disinfect it and return it to the carbon dioxide incubator for culture.

[0112] 2) Randomly select some vascularized co-cultured organ chips, discard the culture medium, and wash with PBS three times.

[0113] b) According to the designed dosing sequence and frequency, the drug regimen is injected into the reserved vascularized lumen channel, and after the drug addition is completed, it is placed on a rocking perfusion instrument in a cell culture incubator for drug intervention culture;

[0114] Specific operations:

[0115] 1) Add 20 μL of drug-containing culture medium to the middle hole of the chip, and add 60 μL of drug-containing culture medium to channel A or channel C.

[0116] 2) Place the chip plate on a rocking perfusion instrument in a carbon dioxide incubator (angle set to 25°, rocking interval 15 min) for incubation, and replenish the drug on day 2, day 4, and day 6.

[0117] c) During the drug administration period, high-content image recognition is performed regularly on the organ chip cultured with drug intervention.

[0118] Specific operations:

[0119] 1) During the drug administration period, the inoculated organoids and vascular networks of all groups were positioned, photographed and scanned, and the area changes of the organoids were analyzed by imaging software.

[0120] 2) After reaching the dosing endpoint, the culture medium was discarded and the cells were washed 3 times with PBS.

[0121] (3) Organ Chip Drug Sensitivity Assessment

[0122] Before administration and after the completion of the administration cycle, the changes in organoid area (OAC) and cell viability (CAC / CNC) were compared and analyzed. When OAC1≥60% and OAC2<20%, ​​the drug was judged to be effective (P); when OAC1<60% and OAC2≥20%, the drug was judged to be ineffective (N); other cases were judged to be intermediately sensitive (M).

[0123] Take the drug sensitivity test of organoid 1 sample as an example:

[0124] <![CDATA[OAC1]]> <![CDATA[OAC2]]> Efficacy evaluation Drug 1 35% 46% N Drug 2 83% -33% P Drug 3 69% 38% M Drug 4 74% -28% P …… …… …… ……

[0125] Evaluation results: Drugs 2 and 4 were confirmed to be effective, drug 3 was intermediately sensitive, and drug 1 was ineffective. Some drugs with significant effects were selected for repeated experiments to verify the reliability of the results.

[0126] Step 3: Discovery of markers associated with drug sensitivity

[0127] For the optimal drug combination that has been screened, biomarkers (gene mutations or gene expressions) associated with the sensitivity of the drug combination are discovered.

[0128] (1) Data Collection

[0129] a) Gene mutation data: Obtain gene mutation and copy number variation data of the organoid samples in the second step. These data can come from whole exome sequencing (WES) and other sources;

[0130] b) Gene expression data: Obtain gene expression data of the organoid samples in the second step. These data can come from transcriptome sequencing (RNA-seq), etc.

[0131] c) Drug sensitivity data: The drug sensitivity data of the optimal drug regimen (the drug regimen with the highest absolute sensitivity rate in the test group that is superior to the control group) is screened out by obtaining the inhibitory effect judgment information on the organoid samples in the second step through high-throughput detection of the organ chip.

[0132] (2) Data preprocessing

[0133] a) Gene mutation data: Use general bioinformatics software such as Mutect 2.0 to detect all somatic and germline mutations, and then obtain somatic mutations through filtering. These mutations include tumor-driving mutations and non-tumor-driving mutations, and standardize the genotype data;

[0134] b) Gene expression data: filter low-quality gene expression data (such as genes with no abnormal expression levels and high noise), and standardize the gene expression data (such as log transformation, z-score standardization, TPM value, etc.);

[0135] c) Drug sensitivity data: Filter low-quality drug sensitivity data. If the judgment results of OAC1 and OAC2 are both positive for drug sensitivity, the drug is considered effective (P); if the judgment results of OAC1 and OAC2 are both negative for drug sensitivity, the drug is considered ineffective (N).

[0136] (3) Correlation analysis

[0137] a) Using the chi-squared exact test, the distribution of gene mutations and abnormal gene expression in the drug-susceptibility negative group and the drug-susceptibility positive group was compared, and the p value < 0.05 was used as the threshold to screen out gene mutations and abnormal gene expression with strong correlation;

[0138] b) Take the drug sensitivity correlation analysis of drug 2 as an example:

[0139]

[0140] c) In the data queue, sensitive and resistant are marked as S and R, respectively, gene mutant and wild type are marked as M and W, respectively, and gene high expression and low expression are marked as H and L, respectively.

[0141] d) Spearman correlation test, the value range is between -1 and 1, 0 means no correlation, 1 means perfect positive correlation, -1 means perfect negative correlation. Among them, 0-0.1: means no correlation; 0.1-0.3: means weak correlation; 0.3-0.5: means moderate correlation; >0.5: means strong correlation, confirming gene mutations and abnormal gene expression with strong correlation (positive or negative correlation).

[0142] e) Take the drug sensitivity correlation analysis of drug 2 as an example:

[0143]

[0144] f) Functional annotation: Functional prediction and annotation were performed on genes with potential functional impacts screened out by the analysis. The data were derived from the STRING database (https: / / cn.string-db.org / ).

[0145] g) Pathway enrichment analysis: Use bioinformatics databases (such as KEGG, Reactome) to perform pathway enrichment analysis to identify affected biological pathways and functional modules, further explain the biological connection between genes and drug sensitivity, and provide a basis for guiding personalized drug treatment.

[0146] h) Gene mutations with significant linkage disequilibrium were excluded through Plink software, and pathway sites closely related to drug absorption, distribution, metabolism, excretion and toxicology were selected through biological pathway analysis.

[0147] Step 4: Build a drug sensitivity prediction model

[0148] a) Based on the gene mutations and abnormal gene expressions that are significantly associated with drug sensitivity, a drug sensitivity prediction model is constructed: a multivariate linear regression model is suggested according to the linear regression method, and each gene mutation and abnormal gene expression can be regarded as a "meta".

[0149] b) Introduce a constraint optimization penalty factor (as a weight coefficient of the element) to correct the deviation. The specific constraint optimization penalty factor can be confirmed by verifying the positive prediction value and ROC curve of the overall model. When the positive prediction value reaches the maximum or the area under the ROC curve is the largest, the constraint optimization penalty factor of the element (gene mutation and abnormal gene expression) used in the model is established. Gene mutations and abnormal gene expression with correlation coefficients close to zero in the linear model can be excluded.

[0150] c) Setting a reasonable threshold can maximize the accuracy of distinguishing drug-susceptibility positive and negative samples, and selecting the omics information and drug sensitivity data of samples not used in constructing the drug sensitivity prediction model for secondary verification, so as to ultimately optimize the parameters of the drug sensitivity prediction model.

[0151] d) Take the drug sensitivity prediction model of drug 2 as an example:

[0152] Actual positive Actual negative total Predict positive 21 3 24 Predict negative 4 12 16 total 25 15 40

[0153] The score interval was set to 0-1, and the threshold was set to 0.5, where 0-0.5 indicated a negative prediction and 0.5-1 indicated a positive prediction. The results showed: positive prediction 87.5%; negative prediction 75%; sensitivity 84%; specificity 80%.

[0154] e) The prediction performance was greater than 70%, indicating that the screened gene mutations and abnormal gene expressions were indeed significantly correlated with drug sensitivity.

[0155] Step 5: Personal Medication Guidance

[0156] a) Sample processing: Based on the specific site information of the mutant genes and abnormally expressed genes involved in the drug sensitivity prediction model, conventional PCR (gene mutation) and qPCR (gene expression) of specific genes are performed on the tumor tissue genome of the subject.

[0157] 1) Sample DNA analysis: Use a DNA extraction kit to extract whole blood genomic DNA, complete sample loading in the sample preparation area, and complete PCR amplification and analysis in the amplification and product analysis area.

[0158] 2) Sample RNA analysis: Use an RNA extraction kit to extract whole blood genomic RNA, complete sample loading in the sample preparation area, and complete qPCR amplification and analysis in the amplification and product analysis area.

[0159] b) Sample prediction: Based on the results of conventional PCR (gene mutation) and qPCR (gene expression) amplification analysis, the mutation (yes or no) or expression (low or high) of the corresponding gene locus of the subject's sample is judged, and the specific gene mutation or abnormal expression information of the subject's sample is evaluated in combination with the drug sensitivity prediction software.

[0160] c) Drug sensitivity assessment: The comprehensive score is calculated by drug sensitivity prediction analysis software. A score greater than 0.5 indicates high sensitivity, suggesting that the subject may be sensitive to the drug regimen; a score less than 0.5 indicates low sensitivity, suggesting that the subject may not be sensitive to the drug regimen.

[0161] In this example, a drug sensitivity screening method based on a vascularized organ chip simulating the tumor microenvironment is constructed; and a drug sensitivity prediction model based on the drug sensitivity detection data of the tumor organoid chip and omics association analysis is constructed. The constructed vascularized organ chip model can have the advantages of simulating the tumor microenvironment and high-throughput detection at the same time; discovering drug sensitivity-related markers and constructing a drug sensitivity prediction model can provide a basis for individual drug use guidance.

[0162] Example 2

[0163] The present invention proposes an application of a tumor organoid chip, using the steps of the method described in Example 1 to first screen out drugs and drug sensitivity-related markers, and construct a drug sensitivity prediction model based on this; the comprehensive score is calculated by the drug sensitivity prediction model to evaluate the sensitivity of the subject to the drug, and it is used as a reference for individual medication guidance to facilitate accurate medication.

[0164] The embodiments of the present invention are described in detail above with reference to the accompanying drawings, but the present invention is not limited thereto, and various changes can be made within the knowledge scope of technicians in the relevant technical field without departing from the purpose of the present invention.

Claims

1. A method for guiding precision medication using a tumor organoid chip, characterized in that: The specific steps include: S1. Construction of vascularized co-culture organ chips; S2. Conduct drug sensitivity testing for anti-tumor combination drug regimens; S3. For the optimal drug combination that has been screened, select biomarkers related to the sensitivity of the drug combination; S4. Construct a drug sensitivity prediction model based on the gene mutations and abnormal gene expressions that are significantly associated with drug sensitivity; S5. Conduct drug sensitivity assessment on patients based on the drug sensitivity prediction model, and use the assessment results as a reference for individual medication guidance.

2. The method for guiding precise medication using a tumor organoid chip according to claim 1, characterized in that: S1 also includes: S11, Organoid preparation and seeding; S111. Sample collection: Collect multiple different human tumor tissue samples to maintain tissue activity; S112, tissue digestion: cut the tumor tissue into small pieces, digest it in a digestive enzyme solution, and filter and collect the cell suspension; S113, cell culture: mix the transfected cell suspension with matrix gel, add it dropwise to the culture plate, and add organoid culture medium for culture after solidification; S114, initial culture: culture in organoid culture medium for 7-14 days, during which time the culture medium is changed; S115, subculture: subculture by mechanical or chemical methods, and collect for later use; S116. Organoid inoculation: Take out the prepared blank organ chip, inoculate the organoid sample mixed with matrix gel into the reserved chip hole / channel, and place it in a cell culture incubator to wait for the matrix gel to solidify; S12, construction of vascularized microenvironment; S121. Cell culture: Take frozen human vascular endothelial cells, resuscitate, resuspend and centrifuge them, transfer them into cell culture bottles, and place them in a cell culture incubator for standby use; S122. Cell inoculation: Take out the organ chip inoculated with organoids, inoculate the human vascular endothelial cell suspension into the reserved chip holes / channels, and place it on a rocking perfusion instrument in a cell culture incubator for culture to ensure that the vascular endothelium forms a dense lumen; S123. Model verification: Organ chip modules were randomly selected to verify the density of vascularized channels and characterize the barrier function, proving that the constructed vascularized lumen had good density and barrier function and could be used as a drug delivery pathway for drug testing; S124. Vascular network: After the vascular lumen is dense, vascular growth factors are added into the channel to stimulate endothelial cells to penetrate into the matrix gel mixed with organoid samples to construct an organoid vascularized microenvironment.

3. The method for guiding precise medication using a tumor organoid chip according to claim 2, characterized in that: S2 also includes: S21. Group the test drugs; S22, using organ chips for drug sensitivity testing; S23. Conduct drug sensitivity assessment on the organ chip.

4. The method of using tumor organoid chips to guide precision medication according to claim 3, characterized in that: The drug regimen of the experimental group in S21 is a drug that inhibits tumor growth. A blank group is also set up. The dosage is designed in a gradient based on the maximum blood drug concentration Cmax value of the reference drug. The number of drug combinations to be tested is allocated based on the single-chip flux number of the organ chip and the number of repetitions in each experimental group.

5. The method of using tumor organoid chips to guide precision medication according to claim 3, characterized in that: The S22 also includes: S221. Before drug administration, perform preliminary high-content image recognition on the constructed vascularized co-culture organ chip; S222. In accordance with the designed dosing sequence and frequency, the drug regimen is injected into the reserved vascularized lumen channel, and after the drug injection is completed, the cell is placed on a rocking perfusion apparatus in a cell culture incubator for drug intervention culture; S223. During the drug administration period, high-content image recognition is performed regularly on the organ chip cultured with drug intervention.

6. The method of using tumor organoid chips to guide precision medication according to claim 3, characterized in that: S23 also includes: S231. Organoid area: Use a high-content imaging analyzer to take images of organoids, collect image data and analyze the area change ratio (OAC) of the organoids to determine drug sensitivity; Endpoint OAC formula: OAC1(%)=(1-LOA) T / LONG C )*100% LONG T =(T t -T0) / T0 ROA C =(C t -C0) / C0 T0: identification area of ​​the experimental / control group organoids when the drug was administered; T t : recognition area of ​​the organoids in the experimental group / control group at time t; C0: recognition area of ​​the organoids in the blank group when the drug was administered; C t : When the recognition area of ​​the blank group organoid at time t is OAC1 ≥ 60%, it is judged as drug sensitivity positive; when OAC1 < 60%, it is judged as drug sensitivity negative; Autologous OAC formula: OAC2(%)=(A t -A0) / A0*100% A0: identification area of ​​each group of organoids when administered, A t : The identification area of ​​each group of organoids at time t; When OAC2 is less than 20%, it is judged as drug sensitivity positive; when OAC2 is greater than or equal to 20%, it is judged as drug sensitivity negative; Comprehensive evaluation: When OAC1 ≥ 60% and OAC2 < 20%, the drug is judged to be effective; when OAC1 < 60% and OAC2 ≥ 20%, the drug is judged to be ineffective; other situations are judged as potential sensitivity; S232. Before dosing and after the completion of the dosing cycle, compare and analyze the changes in organoid area. When OAC1 ≥ 60% and OAC2 < 20%, it is judged as drug effective P; when OAC1 < 60% and OAC2 ≥ 20%, it is judged as drug ineffective N; other cases are judged as intermediate sensitivity M.

7. The method of using a tumor organoid chip to guide precision medication according to claim 6, characterized in that: S3 also includes: S31. Data collection; S311, gene mutation data: obtaining gene mutation and copy number variation data of the organoid sample of the drug sensitivity test result in step S2; S312, gene expression data: obtaining gene expression data of the organoid sample of the drug sensitivity test result in step S2; S313, drug sensitivity data: obtaining the inhibitory effect judgment information of the organoid sample of the drug sensitivity test result in step S2 through high-throughput detection of the organ chip; S32, data preprocessing; S321. Gene mutation data: All somatic and germline mutations were detected using general bioinformatics software, and then somatic mutations were obtained through filtering. These mutations included tumor-driving mutations and non-tumor-driving mutations, and the genotype data were standardized. S322, Gene expression data: filter low-quality gene expression data and standardize the gene expression data; S323, drug sensitivity data: filter low-quality drug sensitivity data. If the judgment results of OAC1 and OAC2 are both positive for drug sensitivity, it is considered that the drug is effective P; if the judgment results of OAC1 and OAC2 are both negative for drug sensitivity, it is considered that the drug is ineffective N; S33, correlation analysis; S331. Using the chi-square exact test, the distribution of gene mutations and abnormal gene expressions in the drug-susceptibility negative group and the drug-susceptibility positive group was compared, and the p value < 0.05 was used as the threshold to screen out gene mutations and abnormal gene expressions with strong correlation; S332, using the Spearman correlation test, with a p value of <0.05 as the threshold, confirm the gene mutations and abnormal gene expressions with strong correlation; S333, Functional Annotation: Functional prediction and annotation of genes with potential functional impact screened out by analysis, data from STRING database; S334, Pathway enrichment analysis: Use bioinformatics databases (such as KEGG, Reactome) to perform pathway enrichment analysis to identify affected biological pathways and functional modules, further explain the biological connection between genes and drug sensitivity, and provide a basis for guiding personalized drug treatment; S335. Gene mutations with significant linkage disequilibrium were excluded using Plink software.

8. The method for guiding precise medication using a tumor organoid chip according to claim 1 or 6, characterized in that: S4 The steps of constructing a drug sensitivity prediction model include: S41. Construct a drug sensitivity prediction model based on the gene mutations and abnormal gene expressions that are significantly associated with drug sensitivity: establish a multivariate linear regression model using the linear regression method, where each gene mutation and abnormal gene expression can be considered as a "meta"; S42. Introduce a constraint optimization penalty factor as a weight coefficient of the element to correct the deviation. The specific constraint optimization penalty factor can be confirmed by verifying the positive prediction value and ROC curve of the overall model. When the positive prediction value reaches the maximum or the area under the ROC curve is the largest, the element constraint optimization penalty factor used in the model is established; gene mutations and abnormal gene expressions with correlation coefficients close to zero in the linear model can be excluded; S43. Select the omics information and drug sensitivity data of the samples not used in constructing the drug sensitivity prediction model for secondary verification, so as to finally optimize the parameters of the drug sensitivity prediction model and generate a drug sensitivity detection and analysis model.

9. The method of using a tumor organoid chip to guide precise medication according to claim 1, characterized in that: The S5 also includes: S51. Sample processing: Based on the specific site information of the mutant genes and abnormally expressed genes involved in the drug sensitivity prediction model, conventional PCR and qPCR of specific genes are performed on the tumor tissue genome of the subject; S52. Sample prediction: determine the mutation or expression of the corresponding gene loci of the subject's sample, and evaluate the specific gene mutation or abnormal expression information of the subject's sample in combination with the drug sensitivity prediction software; S53. Drug sensitivity assessment: The comprehensive score is calculated through the drug sensitivity test prediction model, which is divided into high sensitivity and low sensitivity, indicating the sensitivity of the subject to the drug, which will serve as a reference for individual medication guidance.

10. An application of a tumor organoid chip, characterized in that: A drug sensitivity prediction model is constructed using the method described in any one of claims 1 to 9, and a comprehensive score is calculated using the drug sensitivity prediction model to evaluate the sensitivity of the subject to the drug, and the score is used as a reference for individual medication guidance.

Citation Information

Patent Citations

  • Method for constructing lung cancer organ vascularization co-culture model based on membrane-free organ chip

    CN118421563A