Methods and applications for constructing models to predict the protection probability of type O foot-and-mouth disease vaccines

By constructing a liquid-phase blocking ELISA method and a binary logistic regression model based on foot-and-mouth disease neutralizing nanobodies, the biosafety and long testing cycle issues in the efficacy evaluation of type O foot-and-mouth disease vaccines were solved, and safe and rapid vaccine efficacy testing was achieved.

CN123120271APending Publication Date: 2026-11-17LANZHOU VETERINARY RESEARCH INSTITUTE CHINESE ACADEMY OF AGRICULTURAL SCIENCES(LANZHOU BRANCH CENTER OF CHINA ANIMAL HEALTH & EPIDEMIOLOGY CENTER)
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202611386816.1
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-09-08
Publication Date
2026-11-17

AI Technical Summary

Technical Problem

Current methods for evaluating the efficacy of type O foot-and-mouth disease vaccines use high-titer live virus challenge experiments, which poses risks to biosafety, causes suffering to laboratory animals, and results in long testing cycles.

Method used

A liquid-phase blocking ELISA method based on foot-and-mouth disease neutralizing nanobodies was constructed. A predictive model for vaccine protection probability was established through binary logistic regression analysis. The neutralizing antibody titer was detected by liquid-phase blocking ELISA, and the vaccine efficacy was calculated by the binary logistic regression model. This method avoids the use of live virus, reduces the number of experimental animals used, and shortens the detection cycle.

Benefits of technology

It has achieved the elimination of biosafety risks, the protection of laboratory animal welfare, and the shortening of the testing cycle, reducing testing costs and yielding significant social and economic benefits.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN123120271A_ABST
    Figure CN123120271A_ABST
Patent Text Reader

Abstract

This invention relates to the field of vaccine efficacy testing technology, and discloses a method for constructing a model to predict the protection probability of a type O foot-and-mouth disease vaccine and its application. The method for constructing a model to predict the protection probability of a type O foot-and-mouth disease vaccine includes: S10, selecting at least three batches of type O foot-and-mouth disease vaccine, immunizing experimental animals according to at least three graded dilution dose groups, collecting serum after immunization, and simultaneously conducting virulent challenge experiments to record the challenge protection results of each experimental animal; S20, using a liquid-phase blocking ELISA method based on foot-and-mouth disease neutralizing nanobodies, detecting the neutralizing antibody titer of each serum sample, and logarithmic the results. 10 Transformation; S30, perform binary logistic regression analysis, and obtain the binary logistic regression model logit(P)=ln[P / (1-P)]= β 0 + β 1 ×log 10 ( X In ) β 0 and β 1 The numerical value; S40, for model parameters β 0 and β 1 The model was validated and a unified model was obtained for predicting the protection probability of type O foot-and-mouth disease vaccine. Pigs were used as experimental animals, and the model parameters were... β 0 = -6.240, β 1 =4.238.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention relates to the field of vaccine efficacy testing technology, and in particular to a method for constructing a model for predicting the protection probability of a type O foot-and-mouth disease vaccine, a method for testing the in vitro efficacy of a type O foot-and-mouth disease vaccine, a system for testing the in vitro efficacy of a type O foot-and-mouth disease vaccine, and a product for testing the in vitro efficacy of a type O foot-and-mouth disease vaccine. Background Technology

[0002] Foot-and-mouth disease (FMD) is an acute, febrile, and highly contagious disease caused by the foot-and-mouth disease virus. It primarily infects cloven-hoofed animals such as pigs, cattle, and sheep. With a high morbidity rate and rapid spread, it is listed as a notifiable animal disease by the World Organisation for Animal Health (WOAH) and is also classified as a Class A animal disease in my country. Vaccination is the most economical and effective means of preventing and controlling FMD. According to the Chinese Veterinary Pharmacopoeia and related regulations, FMD vaccines must undergo potency testing before being marketed and before each batch is released to ensure that each dose meets the prescribed potency standards.

[0003] Currently, the gold standard method for evaluating the efficacy of type O foot-and-mouth disease vaccines is the virulent virus challenge protection test. This method involves immunizing experimental animals with serially diluted vaccine, followed by administration of 1000 IDs at a certain post-immunization time. 50 The virus was challenged with a dose of virulent type O foot-and-mouth disease virus to determine its efficacy.

[0004] However, the aforementioned traditional methods of attacking viruses have the following insurmountable drawbacks:

[0005] First, challenge experiments must use high-titer live O-type foot-and-mouth disease virus (a virulent virus). During the operation, there is a risk of virus leakage, spread, and infection of experimental personnel and the environment. This places stringent requirements on the biosafety laboratory level, posing significant safety hazards to the quality control of vaccine manufacturers and the preclinical research of new veterinary drugs.

[0006] Second, unprotected experimental animals will become infected and develop symptoms after being challenged with the virus, exhibiting obvious clinical symptoms and suffering, and will eventually need to be euthanized. This method causes severe suffering and stress to the experimental animals.

[0007] Third, traditional challenge tests typically take nearly two months from animal immunization, feeding, blood collection, challenge to observation and judgment, which significantly prolongs the vaccine development and batch release cycle and indirectly shortens the vaccine's shelf life. Summary of the Invention

[0008] To address the problems existing in the background art, this application provides a method for constructing a model for predicting the protection probability of a type O foot-and-mouth disease vaccine, a method for testing the in vitro efficacy of a type O foot-and-mouth disease vaccine, a system for testing the in vitro efficacy of a type O foot-and-mouth disease vaccine, and a product for testing the in vitro efficacy of a type O foot-and-mouth disease vaccine.

[0009] According to a first aspect of this application, some embodiments provide a method for constructing a model for predicting the protection probability of a type O foot-and-mouth disease vaccine, comprising the following steps:

[0010] S10. Select at least three batches of type O foot-and-mouth disease vaccine and immunize experimental animals according to at least three graded dilution dose groups. Immunize at least five experimental animals in each dose group. Collect serum 28 days after immunization and conduct virulent challenge experiments simultaneously to record the challenge protection results of each experimental animal.

[0011] S20, using a liquid-phase blocking ELISA method based on foot-and-mouth disease neutralizing nanobodies, the neutralizing antibody titer of each serum sample was detected, and the neutralizing antibody titer was logarithmic. 10 Conversion;

[0012] S30, in log 10 The converted neutralizing antibody titer is used as the independent variable X, and the challenge protection result is used as the dependent variable Y. A binary logistic regression analysis is performed, yielding the binary logistic regression model logit(P) = ln[P / (1-P)] = β 0 + β 1 ×log 10 ( X In ) β 0 and β 1 The value of ; where P is the protection probability, X For antibody titer, β 0 The intercept is... β 1 The slope;

[0013] S40, Perform significance tests, coefficient significance tests, and goodness-of-fit tests on the binary logistic regression model to verify the model parameters. β 0 and β 1 This yields a model for predicting the protection probability of type O foot-and-mouth disease vaccine, known as the unified model.

[0014] The experimental animal is a pig, and the model parameters are... β 0 =-6.240, β1 =4.238.

[0015] According to a second aspect of this application, some embodiments provide a method for testing the in vitro efficacy of a type O foot-and-mouth disease vaccine, comprising the following steps:

[0016] S10, the O-type foot-and-mouth disease vaccine to be tested was immunized in experimental animals according to at least three graded dilution dose groups, with at least five experimental animals immunized in each dose group, and serum was collected 28 days after immunization;

[0017] S20, using a liquid-phase blocking ELISA method based on foot-and-mouth disease neutralizing nanobodies, the titer of type O foot-and-mouth disease neutralizing antibodies in the serum of each experimental animal was detected;

[0018] S30, the neutralizing antibody titer is log 10 After conversion, the model is substituted into the unified model constructed by the method described in the first aspect to calculate the protection probability P of each experimental animal;

[0019] S40, Calculate the average protection probability P for each dose group. 平均 And with the logarithm of dose as the independent variable and P as the variable. 平均 The logit value is used as the dependent variable for linear regression to fit the dose-response curve;

[0020] S50, based on the dose-response curve, calculate the PD according to the dose corresponding to the mean protection probability of 0.5. 50 Value, and based on the PD content of each dose of vaccine. 50 Quantity determines vaccine efficacy.

[0021] According to a third aspect of this application, some embodiments provide an in vitro potency testing system for type O foot-and-mouth disease vaccine, comprising:

[0022] The input module is used to receive neutralizing antibody titer data from experimental animals immunized with different dose groups of the same vaccine;

[0023] The probability calculation module is used to substitute the neutralizing antibody titer into the unified model constructed as described in the first aspect to calculate the individual protection probability of each experimental animal and the average protection probability of the dose group.

[0024] The curve fitting module is used to perform linear regression with the logarithm of the dose as the independent variable and the logit value of the average protection probability as the dependent variable to fit the dose-response curve.

[0025] The efficacy determination module is used to calculate the PD based on the dose-response curve. 50 The value was calculated and converted into the PD (PD) per dose of vaccine. 50 The quantity is considered sufficient; when the quantity is ≥6, the vaccine is deemed to be of acceptable efficacy.

[0026] According to a fourth aspect of this application, some embodiments provide a product for testing the in vitro efficacy of a type O foot-and-mouth disease vaccine, comprising:

[0027] Memory, used to store computer programs;

[0028] A processor for implementing steps S30-S50 of the method described in the second aspect by executing the computer program stored in the memory.

[0029] The method for constructing a model for predicting the protection probability of a type O foot-and-mouth disease vaccine, the method for testing the in vitro efficacy of a type O foot-and-mouth disease vaccine, the in vitro efficacy testing system for a type O foot-and-mouth disease vaccine, and the product for testing the in vitro efficacy of a type O foot-and-mouth disease vaccine provided by this invention can avoid the use of live virus in challenge experiments, eliminate biosafety risks, reduce the amount of experimental animals used and their suffering, comply with animal welfare and ethical principles, and shorten the testing cycle and reduce testing costs, thus having significant social and economic benefits. Attached Figure Description

[0030] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the structures shown in these drawings without creative effort.

[0031] Figure 1 This is a graph showing the positive and negative serum titer results provided in an embodiment of the present invention;

[0032] Figure 2 This is a diagram showing the cross-reaction results provided in an embodiment of the present invention;

[0033] Figure 3 This is a graph showing the experimental results of the concordance rate between type O NA-ELISA and VNT provided in an embodiment of the present invention;

[0034] Figure 4 This is a curve diagram of the foot-and-mouth disease type O logistic regression model provided in an embodiment of the present invention.

[0035] The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0036] The present invention will now be described in further detail with reference to specific embodiments and accompanying drawings. In the following embodiments, many details are described to facilitate a better understanding of the present application. However, those skilled in the art will readily recognize that some features may be omitted in different situations, or may be replaced by other materials or methods. In some cases, certain operations related to the present application are not shown or described in the specification. This is to avoid obscuring the core parts of the present application with excessive description. For those skilled in the art, detailed description of these related operations is not necessary; the relevant operations can be fully understood based on the description in the specification and general technical knowledge in the art.

[0037] Furthermore, the features, operations, or characteristics described in the specification can be combined in any suitable manner to form various embodiments. At the same time, the steps or actions in the method description can be rearranged or adjusted in a manner obvious to those skilled in the art. Therefore, the various orders in the specification and drawings are only for the clear description of a particular embodiment and do not imply a necessary order, unless otherwise stated that a particular order must be followed.

[0038] According to a first aspect of this application, in some embodiments a method for constructing a model for predicting the protection probability of type O foot-and-mouth disease vaccine includes the following steps:

[0039] S10. Select at least three batches of type O foot-and-mouth disease vaccine and immunize experimental animals according to at least three graded dilution dose groups. Immunize at least five experimental animals in each dose group. Collect serum 28 days after immunization and conduct virulent challenge experiments simultaneously to record the challenge protection results of each experimental animal.

[0040] S20, using a liquid-phase blocking ELISA method based on foot-and-mouth disease neutralizing nanobodies, the neutralizing antibody titers of each serum sample were detected, and the neutralizing antibody titers were logarithmic. 10 Conversion;

[0041] S30, in log 10 The converted neutralizing antibody titer is used as the independent variable X, and the challenge protection result is used as the dependent variable Y. A binary logistic regression analysis is performed, yielding the binary logistic regression model logit(P) = ln[P / (1-P)] = β 0 + β 1 ×log 10 ( X In ) β 0 and β 1 The value of ; where P is the protection probability, X For antibody titer,β 0 The intercept is... β 1 The slope;

[0042] S40. Perform significance tests, coefficient significance tests, and goodness-of-fit tests on the binary logistic regression model to validate the model parameters. β 0 and β 1 This yields a model for predicting the protection probability of type O foot-and-mouth disease vaccine, known as the unified model.

[0043] The experimental animals were pigs, and the model parameters were... β 0 =-6.240, β 1 =4.238. Therefore, the unified model for predicting the protection probability of type O foot-and-mouth disease vaccine is: logit(P) = ln[P / (1-P)] = -6.240 + 4.238 × log 10 ( X P is the protection probability. X This refers to the antibody titer.

[0044] It should be noted that the multiple batches of type O foot-and-mouth disease vaccine used in the operational steps can be entirely monovalent type O foot-and-mouth disease vaccine, or some batches can be monovalent type O foot-and-mouth disease vaccine and some batches can be multivalent or combination foot-and-mouth disease vaccine containing type O. For example, some batches can be bivalent foot-and-mouth disease vaccine containing type O, and the bivalent foot-and-mouth disease vaccine containing type O can be a type A and type O bivalent foot-and-mouth disease vaccine.

[0045] The unified model constructed using the above method for determining the in vitro efficacy of type O foot-and-mouth disease vaccine can avoid the use of live virus in challenge experiments, eliminate biosafety risks, reduce the amount of experimental animals used and their suffering, comply with animal welfare and ethical principles, and shorten the testing cycle and reduce testing costs, thus having significant social and economic benefits.

[0046] In some embodiments, binary logistic regression analysis is performed using Python programming. It is understood that in other embodiments, binary logistic regression analysis can also be performed using languages ​​such as C / C++ and Java.

[0047] In some embodiments, in step S10, experimental animals are immunized with three dose groups: 1 dose, 1 / 3 dose, and 1 / 9 dose.

[0048] In some embodiments, in step S10, at least ten batches of type O foot-and-mouth disease vaccine are selected, and at least five experimental animals are immunized in each dose group. Specifically, batches of type O foot-and-mouth disease vaccine from batches 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, and 36 can be selected. This arrangement is used to improve the accuracy of the uniform model.

[0049] Preferably, in step S10, at least twenty batches of type O foot-and-mouth disease vaccine are selected, and at least five experimental animals are immunized in each dose group.

[0050] In some embodiments, in step S10, a total of no less than 200 experimental animals are immunized. In specific applications, a total of 200, 210, 220, 230, 240, 250, 260, 270, 280, 290, 300, 310, 320, 330, 340, 350, 360, 370, 380, 390, or 400 experimental animals can be immunized.

[0051] Preferably, in step S10, a total of no less than 300 experimental animals are immunized. In this embodiment, a total of 390 experimental animals are immunized. This provides 390 valid data points.

[0052] In some embodiments, in step S10, the criteria for determining the protection status of the virulent challenge experiment are: the presence of typical foot-and-mouth disease symptoms indicates no protection, while the absence of typical foot-and-mouth disease symptoms indicates protection. Typical foot-and-mouth disease symptoms include blisters and ulcers on the nasal surface and hooves. In specific applications, the presence of any one of these symptoms (nasal surface, blisters, or ulcers on the hooves) indicates no protection.

[0053] In some embodiments, in step S40, the significance test is a likelihood ratio test, the coefficient significance test is a Wald test, and the goodness-of-fit test is a Nagelkerke R-squared test. 2 test.

[0054] The null hypothesis for the likelihood ratio significance test is that the model is insignificant and all coefficients are 0. The null hypothesis is rejected if the p-value of the likelihood ratio chi-square is less than 0.05. The Wald coefficient significance test is used to determine whether the regression coefficients of the intercept and slope are significantly different from zero. Nagelkerke R... 2 Goodness-of-fit tests are used to assess a model’s ability to explain the variation in the dependent variable.

[0055] In some embodiments, the construction method further includes step S50 after step S40: converting the PD calculated based on the unified model... 50 The value compared to the PD obtained from traditional challenge experiments 50 Pearson correlation analysis was performed on the values.

[0056] In some embodiments, step S20 of the liquid-phase blocking ELISA method includes:

[0057] S21, the nanobody Ab2 that neutralizes foot-and-mouth disease virus was diluted to the working concentration with carbonate buffer, added to the ELISA plate, coated overnight at 4°C, washed with PBST and patted dry;

[0058] S22, the serum to be tested is serially diluted 2-fold with PBST, mixed with an equal volume of PBST buffer containing VLPs, vortexed and mixed, and reacted overnight at 4°C to obtain serum-antigen mixture;

[0059] S23, Add the serum-antigen mixture obtained in step S22 to the enzyme-labeled plate obtained in step S21, and react at room temperature;

[0060] After washing with S24 and PBST, add biotin-labeled detection antibody and react at room temperature;

[0061] After washing with S25 and PBST, HRP-labeled streptavidin was added and the reaction was carried out at room temperature.

[0062] S26, after washing with PBST, added the substrate TMB for color development at room temperature, and terminated the reaction with a stop solution;

[0063] S27, read the OD of each well using a microplate reader at a wavelength of 450 nm. 450 nm The value is used to calculate the neutralizing antibody titer based on preset judgment criteria.

[0064] In some embodiments, the working concentration in step S21 is 1 µg / mL.

[0065] In some embodiments, in step S22, the final concentration of VLPs is 5 µg / mL.

[0066] In some embodiments, in step S24, the reaction concentration of the biotin-labeled detection antibody is 1.5 µg / mL.

[0067] In some embodiments, in step S25, the reaction concentration of HRP-labeled streptavidin is 0.5 µg / mL.

[0068] In some embodiments, in step S27, the preset judgment criterion is: the OD of the VLPs antigen control well. 450 nm 50% of the average value is used as the critical value.

[0069] According to a second aspect of this application, some embodiments provide a method for testing the in vitro efficacy of a type O foot-and-mouth disease vaccine, comprising the following steps:

[0070] S10, immunize experimental animals with the O-type foot-and-mouth disease vaccine to be tested according to at least three graded dilution dose groups, with at least five experimental animals immunized in each dose group, and collect serum 28 days after immunization; wherein, each dose group can immunize five, six, seven or eight experimental animals, etc.

[0071] S20, using a liquid-phase blocking ELISA method based on foot-and-mouth disease neutralizing nanobodies, the titer of type O foot-and-mouth disease neutralizing antibodies in the serum of each experimental animal was detected;

[0072] S30, the neutralizing antibody titer is log 10 After conversion, the model is substituted into the unified model constructed by the method described in the first aspect to calculate the protection probability P of each experimental animal;

[0073] S40, Calculate the average protection probability P for each dose group. 平均 And with the logarithm of dose as the independent variable and P as the variable. 平均 The logit value is used as the dependent variable for linear regression to fit the dose-response curve;

[0074] S50, based on the dose-response curve, calculate the PD according to the dose corresponding to the mean protection probability of 0.5. 50 Value, and based on the PD content of each dose of vaccine. 50 Quantity determines vaccine efficacy.

[0075] It should be noted that by using the above method to determine the in vitro efficacy of type O foot-and-mouth disease vaccine, the use of live virus in challenge experiments can be avoided, biosafety risks can be eliminated, the amount of experimental animals used and their suffering can be reduced, which is in line with animal welfare and ethical principles. Furthermore, the testing cycle can be shortened and the testing cost can be reduced, which has significant social and economic benefits.

[0076] In some embodiments, in step S50, when each dose of vaccine contains PD 50 A vaccine is considered effective if it contains ≥6 mg / dL of PD. For example, if each dose of vaccine contains ≥6 mg / dL of PD... 50 A score of 6, 7, 8, 9, or 10 indicates that the vaccine is effective.

[0077] In some embodiments, the liquid-phase blocking ELISA method includes:

[0078] S21, the nanobody Ab2 that neutralizes foot-and-mouth disease virus was diluted to the working concentration with carbonate buffer, added to the ELISA plate, coated overnight at 4°C, washed with PBST and patted dry;

[0079] S22, the serum to be tested is serially diluted 2-fold with PBST, mixed with an equal volume of PBST buffer containing VLPs, vortexed and mixed, and reacted overnight at 4°C to obtain serum-antigen mixture;

[0080] S23, Add the serum-antigen mixture obtained in step S22 to the enzyme-labeled plate obtained in step S21, and react at room temperature;

[0081] After washing with S24 and PBST, add biotin-labeled detection antibody and react at room temperature;

[0082] After washing with S25 and PBST, HRP-labeled streptavidin was added and the reaction was carried out at room temperature.

[0083] S26, after washing with PBST, added the substrate TMB for color development at room temperature, and terminated the reaction with a stop solution;

[0084] S27, read the OD of each well using a microplate reader at a wavelength of 450 nm. 450 nm The value is used to calculate the neutralizing antibody titer based on preset judgment criteria.

[0085] In some embodiments, the working concentration in step S21 is 1 µg / mL.

[0086] In some embodiments, in step S22, the final concentration of VLPs is 5 µg / mL.

[0087] In some embodiments, in step S24, the reaction concentration of the biotin-labeled detection antibody is 1.5 µg / mL.

[0088] In some embodiments, in step S25, the reaction concentration of HRP-labeled streptavidin is 0.5 µg / mL.

[0089] In some embodiments, in step S10, the experimental animal is a pig weighing 35-45 kg. It should be noted that all selected pigs are healthy. In specific applications, the pig's weight can be 35 kg, 36 kg, 37 kg, 38 kg, 39 kg, 40 kg, 41 kg, 42 kg, 43 kg, 44 kg, or 45 kg, etc.

[0090] Preferably, in step S10, the experimental animal is a pig weighing 39-41 kg.

[0091] In some embodiments, in step S10, experimental animals are immunized with three dose groups: 1 dose, 1 / 3 dose, and 1 / 9 dose.

[0092] In some embodiments, in step S40, the linear regression model is:

[0093] logit(P 平均) = a + b × log 10 (dose)

[0094] Where a is the intercept and b is the slope.

[0095] In some embodiments, in step S50, the PD 50 The value is based on the formula PD. 50 Calculated using 10^(-a / b), the PD content per dose of vaccine is... 50 Quantity = 1 / PD 50 .

[0096] According to a third aspect of this application, some embodiments provide an in vitro potency testing system for type O foot-and-mouth disease vaccine, comprising:

[0097] The input module is used to receive neutralizing antibody titer data from experimental animals immunized with different dose groups of the same vaccine;

[0098] The probability calculation module is used to substitute the neutralizing antibody titer into the unified model constructed as described in the first aspect to calculate the individual protection probability of each experimental animal and the average protection probability of the dose group.

[0099] The curve fitting module is used to perform linear regression with the logarithm of the dose as the independent variable and the logit value of the average protection probability as the dependent variable to fit the dose-response curve.

[0100] The efficacy assessment module is used to calculate PD based on the dose-response curve. 50 The value was calculated and converted into the PD (PD) per dose of vaccine. 50 The quantity is considered sufficient; when the quantity is ≥6, the vaccine is deemed to be of acceptable efficacy.

[0101] According to a fourth aspect of this application, in some embodiments a product for testing the in vitro efficacy of type O foot-and-mouth disease vaccine includes:

[0102] Memory, used to store computer programs;

[0103] A processor for implementing steps S30-S50 of the method described in the second aspect by executing a computer program stored in memory.

[0104] The present application will be further described in detail below through specific embodiments. These embodiments are only for further illustration and should not be construed as limiting the present application. It should be noted that, unless otherwise specified, the reagents, instruments, etc., used in the embodiments are all commercially available.

[0105] 1. Method

[0106] 1.1 Neutralizing antibody liquid-phase blocking ELISA detection method

[0107] 1.1.1 Reagents, antigens and antibodies, serum

[0108] The O-type VLPs purified and assembled from prokaryotic expression were prepared and preserved in our laboratory; the eukaryotic specific neutralizing antibody Ab2 (amino acid sequence as shown in SEQ ID NO: 1: HVQLVESGGGSVQAGGSLRLSCAASGYTYSSNCMGWFRQAPGKGREGVATIVTGGGSTYYADSVKGRFTISQDNAKNTVYLQMNSLKPEDTAMYYCAADSRSPSYCSGGYIGDADFAYWGQGTQVTVSS) was prepared and preserved in our laboratory; 96-well microplates were purchased from Corning; carbonate powder was purchased from SIGMA; and the substrate TMB was purchased from SURMODICS.

[0109] Serum sources: 400 negative serum samples from pigs and cattle were screened by the diagnostic group of our laboratory, and 600 serum samples were obtained from pigs and cattle immunized with VLPs / inactivated vaccines in our laboratory.

[0110] 1.1.2 Identification of VLPs

[0111] The assembled VLPs protein was quantified by SDS-PAGE, Western blotting (primary antibody was type O FMDV-positive porcine serum, enzyme-labeled secondary antibody was anti-porcine HRP-IgG, primary antibody dilution ratio was 1:1000, enzyme-labeled secondary antibody dilution ratio was 1:2000), DLS identification, and the VLPs intact particle quantification method developed in our laboratory: 1. Coating with 0.5 µg / ml of specific antibody overnight at 4°C using phosphate buffer; 2. After washing with PBST, the standard was serially modified 2-fold at the specified concentration, and the detection antigen was diluted to 2-4 µg / ml, then serially diluted 2-fold, and incubated at room temperature for 45 min; 3. After washing with PBST, 0.25 µg / ml of detection antibody was added, and incubated at 37°C for 30 min; 4. After washing with PBST, 0.0625 µg / ml of enzyme-labeled secondary antibody was added, and incubated at room temperature for 60 min, followed by TMB color development at room temperature for 1.5 minutes. 5. Terminate color development after min; 6. Plot a standard curve based on the OD value and concentration of the standard, and the VLP content of the detected antigen can be calculated.

[0112] 1.1.3 Identification of Ab2

[0113] Ab2 expressed and purified by the mammalian cell Expi-293-F expression system was identified by SDS-PAGE, and its concentration was determined by the BCA protein concentration assay kit. Its neutralizing activity was identified by a virus neutralization assay, and the O-type FMDV epidemic strain was selected for identification.

[0114] 1.1.4 Establishment of the liquid-phase blocking ELISA method

[0115] Liquid-phase blocking ELISA uses the nanobody Ab2, which neutralizes FMDV, as the capture antibody, type O VLPs as the reaction antigen, the same biotin-labeled nanobody as the detection antibody (primary antibody), and HRP-labeled streptavidin as the secondary antibody.

[0116] Neutralizing antibody Ab2 was used for coating. Ab2 was diluted to 1 µg / mL with coating buffer and added to each well of the ELISA plate (100 µL). Coating was carried out overnight at 4°C. After washing, the liquid in the wells was discarded the next day. PBST was added, and the plates were washed 3-4 times, with 300 µL added to each well each time. The plates were then patted dry on absorbent paper. For 96-well serum dilution plates, the serum to be tested was diluted 2-fold with PBS starting at an initial 1:2 gradient (from AH in the 96-well plate), with 50 µL added to each well. After dilution, 50 µL of 5 µg / mL type O VLPs was added to each well to block the reaction. The mixture was manually or with a micro-shaker for 30-60 seconds and incubated overnight at 4°C. The next day, 50 µL of the serum-antigen mixture was transferred to each well of the 96-well plate coated with the specific neutralizing antibody in situ. The plate was incubated at room temperature for 60 min. The plates were washed as described above. 100 µL of 1.5 ppm PBST was added to each well. The same specific neutralizing antibody labeled with µg / mL biotin was used to recognize the remaining neutralizing epitopes on VLPs after the blocking reaction, and the reaction was carried out at room temperature for 60 min. After washing as above, 100 µL of 0.5 µg / mL HRP-labeled streptavidin was added, and the reaction was carried out at room temperature for 60 min. After washing as above, 100 µL of substrate TMB was added to each well, and the reaction was carried out at room temperature for 5 min. 50 µL of stop solution was added to each well and mixed well. The OD value of the sample in the well was read at 450 nm on a microplate reader.

[0117] 1.1.4.1 Preliminary determination of the working concentration of biotin-labeled antibody and the dilution ratio of HRP-labeled streptavidin

[0118] The checkerboard cross-titration method was used. 1. Biotin-labeled antibodies were serially diluted with carbonate buffer to different concentrations: 0.5 µg / mL, 1 µg / mL, 1.5 µg / mL, 2 µg / mL, 2.5 µg / mL, and 3 µg / mL. Each concentration was added to one column of the plate, and the plates were incubated overnight at 4°C. 2. The next day, the liquid in the wells was discarded, and the plates were washed 3-4 times with PBST, adding 300 µL of PBST to each well each time. The plates were then patted dry on absorbent paper. 3. HRP-labeled streptavidin diluted with PBST was added to each well: 0.1 µg / mL, 0.25 µg / mL, 0.5 µg / mL, 0.75 µg / mL, 1 µg / mL, and 1.25 µg / mL. The mixture was gently vortexed manually or with a micro-oscillator for 30-60 seconds, and incubated for 60 minutes at room temperature. 4. After washing as above, 100 µL of PBST was added to each well. 5. After color development, add 50 µL of TMB substrate to each well, gently shake to mix, being careful not to generate bubbles, and read the OD value of the sample at 450 nm on the microplate reader.

[0119] 1.1.4.2 Preliminary determination of specific neutralizing antibody coating and antigen concentration

[0120] The checkerboard cross-titer method was used. 1. Specific neutralizing antibodies were serially diluted with carbonate buffer to different concentrations: 0.25 µg / mL, 0.5 µg / mL, 1 µg / mL, 1.5 µg / mL, 2 µg / mL, 2.5 µg / mL, and 3 µg / mL. Each concentration was added to one column in the ELISA plate, and the plates were incubated overnight at 4°C. 2. The next day, the liquid in the wells was discarded, and the plates were washed 3-4 times with PBST, adding 300 µL of PBST to each well each time, and then patted dry on absorbent paper. 3. Type O VLP antigen diluted with PBST was added to each column: 1 µg / mL, 2 µg / mL, 3 µg / mL, 4 µg / mL, 5 µg / mL, and 6 µg / mL. The reaction was carried out at room temperature for 60 min. 4. After washing as above, 100 µL of biotin-labeled antibody diluted with PBST in the optimal ratio was added to each well, and the reaction was carried out at room temperature for 60 min. 5. Wash as above, add 100 µL of HRP-labeled streptavidin diluted with PBST in the optimal ratio to each well, and incubate at room temperature for 60 min; 6. Wash as above, add 100 µL of substrate TMB to each well, and incubate at room temperature for 1 to 2 min; 7. After incubation, add 50 µL of stop solution to each well, gently vortex to mix, avoiding air bubbles, and read the OD value of the sample at 450 nm on a microplate reader. The highest dilution of the specific neutralizing antibody corresponding to the lowest concentration of VLPs antigen with an OD value of 1.5 or higher is its optimal coating concentration. After determining the concentrations of other reagents, the working concentration of each batch of VLPs antigen must be titrated to achieve the lowest concentration corresponding to the standard positive serum titer, which is its working concentration.

[0121] 1.1.4.3 Determination of the optimal concentration for each reagent

[0122] 1. Dilute specific neutralizing antibodies to the working concentration with carbonate buffer overnight and coat the ELISA plate. Simultaneously, dilute standard sera on the dilution plate: positive serum 1:4, 1:8, 1:16, 1:32, 1:64, 1:128, 1:256, 1:512; negative serum 1:4, 1:8, 1:16, 1:32. Add 50 µL of serum to each well and 50 µL of type O VLPs (preliminary determined working concentration) to each well. Mix manually or with a micro-shaker for 30-60 seconds and incubate overnight at 4°C. 2. The next day, discard the liquid in the wells and wash 3-4 times with PBST, adding 300 µL of PBST to each well each time. Pat dry on absorbent paper. Transfer the antigen-antibody mixture from the dilution plate to the ELISA plate, 50 µL per well. Pure antigen wells contain VLPs diluted to the working concentration. Each group contains three columns of standard samples: one column of strongly positive O-type serum, one column of weakly positive O-type serum, half a column of negative O-type serum, and half a column of pure antigen. React at room temperature for 60 min. 3. After washing, add 100 µL of biotin-labeled specific neutralizing antibody diluted to the working concentration with PBST per well, and react at room temperature for 60 min. 4. Wash as above, then add HRP-labeled streptavidin diluted to the working concentration with PBST, and react at room temperature for 60 min. 5. Wash as above, then add 100 µL of substrate TMB to each well, and incubate at room temperature for 5 min. Add 50 µL of stop solution to each well and mix well. Read the OD value of the sample in the well at 450 nm using a microplate reader. 6. Adjust the final reaction concentrations of each reagent and the dilution of VLPs according to the actual neutralizing titers of the control positive and negative sera.

[0123] 1.1.4.4 Determination of judgment criteria, specificity, and sensitivity

[0124] A liquid-phase blocking ELISA assay was performed using the experimental conditions described above, testing a total of 200 type O and 200 negative serum samples. The titer of each sample was calculated, and the result was determined based on the logarithm of the titer. 10 Logarithmic statistical analysis results were used to plot ROC curves, calculate the thresholds for negative and positive results, and determine the specificity and sensitivity of the method.

[0125] 1.1.4.5 Cross-reactivity test

[0126] The established liquid-phase blocking ELISA method was used to detect three positive sera of type O FMDV and 32 positive sera of common swine diseases (porcine circovirus type 2, porcine parvovirus disease, porcine epidemic diarrhea, porcine reproductive and respiratory syndrome (PRRS), classical swine fever, porcine infectious pleuropneumonia, swine encephalitis B, swine colibacillosis, and Haemophilus parasuis infection). The antibody titers were calculated, and the cross-reactivity of the method was determined based on the results.

[0127] 1.1.4.6 Detection of neutralizing antibody titers in serum

[0128] The established liquid-phase blocking ELISA was used to detect the immune serum of 600 swine and bovine samples from those vaccinated against foot-and-mouth disease type O. The titers of each sample were calculated, and the statistical results were analyzed.

[0129] 1.1.4.7 VNT Compliance Rate Test

[0130] The method established in this study was used to simultaneously detect 600 porcine and bovine type O serum samples with VNT. The statistical results were analyzed, and the concordance rate between this method and VNT was calculated. The VNT procedure involved inactivating the serum samples by placing them in a 56°C metal bath for 30 minutes, using TCID45 with a known toxicity titer. 50 10 -6 The O-type FMDV strain and BHK cells were tested. Serum samples were serially diluted eight times in 96-well plates using serum-free DMEM: 1 / 8; 1 / 16; 1 / 32; 1 / 64; 1 / 128; 1 / 256; 1 / 512; 1 / 1024; each dilution was repeated three times; 50 µl of 100 TCID45 was added to each well. 50 Virus solution: Gently tap the plate to mix the serum and virus solution, and incubate in a cell culture incubator for 1 hour. Digest cells to prepare a cell suspension, add 50 µl of cell suspension to each well, gently tap the plate to mix the cell suspension, virus solution, and serum to be tested. Incubate in a cell culture incubator for 48-72 hours, then count the number of wells with cytopathic effects and the number of wells without cytopathic effects. Calculate the serum neutralizing antibody titer using the Reed-Muench method. Perform a virus regression experiment by serially diluting the virus 10-fold (serum-free DMEM): 10 5 TCID 50 10 4 TCID 50 10 3 TCID 50 10 2 TCID 50 10TCID 50 10 -1 TCID 50 Add 50 µl of diluted virus solution to each well of a 96-well plate, with 8 replicates per dilution. Top up the remaining wells with serum-free DMEM to 50 µl. Digest and resuspend the confluent cells, adding 50 µl of cell suspension to each well. Gently tap the plate to mix the cell suspension and virus solution, and incubate for 48-72 hours. Then, count the number of wells with cytopathic effects and the number of wells without cytopathic effects. Statistical analysis, including Pearson coefficient test and ROC curve analysis, was performed using GraphPad Prism (version 6.0) software.

[0131] 1.2 PD 50 Alternative calculation methods

[0132] 1.2.1 Data Sources and Sample Design

[0133] The data in this study were derived from multiple batches of foot-and-mouth disease virus-like particle (FMD) vaccine and swine serum immunized with FMD type O and type A bivalent mineralized virus-like particle (BMP) vaccines during the research process of this project. The FMD vaccine included: swine FMD type O BMP vaccine [Certificate No.: (2021) New Veterinary Drug Certificate No. 58] and swine FMD type O and type A bivalent BMP vaccine [Certificate No.: (2024) New Veterinary Drug Certificate No. 81]. Each batch of vaccine had three dosage groups: 1 dose, 1 / 3 dose, and 1 / 9 dose. Five animals were immunized in each dosage group, resulting in a total of 390 valid data points. Serum was collected from all animals at specified time points after immunization, and liquid-phase blocking ELISA antibody detection and virulent challenge experiments were performed simultaneously. The virulent challenge experiment was conducted according to the Chinese Veterinary Pharmacopoeia, with each pig receiving an intramuscular injection of 1000 IDs in the neck area. 50 The virus is highly virulent. The criteria for determining protection status are as follows: the presence of typical foot-and-mouth disease symptoms indicates no protection "-", and the absence of symptoms indicates protection "+". Typical symptoms of foot-and-mouth disease include blisters and ulcers on the nasal surface and hooves. The presence of any one of these symptoms indicates no protection "-".

[0134] 1.2.2 Establishment and Evaluation of the Binary Logistic Regression Model

[0135] 1.2.2.1 Establishment of a Binary Logistic Regression Model

[0136] (1) Antibody detection: Serum neutralizing antibody titers were detected using a validated liquid-phase blocking ELISA method, and all titer values ​​were converted to logarithmic values. 10 Statistical analysis was then performed.

[0137] (2) Model construction: A binary logistic regression model was used to analyze antibody titer (independent variable). X log 10 The quantitative relationship between transformation and protection status (dependent variable Y, protection=1, no protection=0).

[0138] The model is in the form of: logit(P) = ln[P / (1-P)] = β 0 + β 1 ×log 10 ( X ), where P is the protection probability, X For antibody titer, β 0 The intercept is... β 1 The slope is denoted as .

[0139] 1.2.2.2 Evaluation of the Binary Logistic Regression Model

[0140] (1) Significance test of binary logistic regression equation (likelihood ratio test) The aim is to determine whether there is a significant linear relationship between the set of all explanatory variables in the model and the logarithm ratio logit(P), i.e., whether linear modeling is suitable. If the explanatory variables have a significant linear effect on logit(P), then the model's fit to the sample data should be significantly improved after introducing these variables.

[0141] Null hypothesis (H0): The model is not significant; all coefficients are 0.

[0142] Alternative hypothesis (H1): At least one coefficient in the model is not zero;

[0143] The improvement in goodness of fit is usually measured by the log-likelihood ratio, and the corresponding test statistic is the likelihood ratio chi-square (r-square), which is calculated as follows:

[0144] χ 2 = -2 ln L 0 - (-2 ln L 1 )

[0145] in:

[0146] L 0 It is the likelihood function value of the zero model (containing only the intercept term);

[0147] L 1 It is the likelihood function value of the current model (including the intercept term and all explanatory variables);

[0148] The likelihood ratio chi-square statistic reflects the degree of change in the log-likelihood value after a new explanatory variable is introduced into the model. The larger the value, the more significant the contribution of the new variable to the model, that is, the greater the impact of the variable on logit(P).

[0149] When performing a significance test, if the observed value of the likelihood ratio chi-squared corresponds to... p If the value is less than the pre-set significance level α, the null hypothesis is rejected, indicating that not all regression coefficients in the model are simultaneously zero, meaning that the linear relationship between all explanatory variables and logit(P) is significant; otherwise, the null hypothesis cannot be rejected, indicating that the linear relationship is not significant.

[0150] (2) Significance test of the coefficients of the binary logistic regression equation (Wald test) This test is used to determine whether the regression coefficient of each explanatory variable in the model is significantly non-zero, i.e., whether the variable has a significant linear relationship with the logarithm logit(P) and makes a significant contribution to explaining logit(P). The Wald statistic is typically used as the test statistic, and its calculation formula is as follows:

[0151] Wald i = ( β i / S βi ) 2

[0152] in:

[0153] β i Indicates the first i The regression coefficients of the explanatory variables;

[0154] S βi express β i The standard error.

[0155] The Wald statistic approximately follows a chi-square distribution. When performing the test, for each explanatory variable, if the observed value of its Wald statistic corresponds to... p If the value is less than the given significance level α, then the null hypothesis should be rejected (H0: β i If the regression coefficient of the explanatory variable is 0, it is considered that the regression coefficient of the explanatory variable is significantly different from zero, and the linear relationship between the explanatory variable and logit(P) is significant, so it should be retained in the equation; otherwise, the null hypothesis is accepted, and the variable is considered not significant and should not be retained in the equation.

[0156] (3) Goodness-of-fit test of the binary logistic regression equation It is used to evaluate the ability of a regression model to explain the variation in the dependent variable, and the degree of agreement between the model's predictions and the actual observed values. A model with a high goodness of fit means that it can explain a large portion of the variation in the dependent variable and has a low overall misclassification rate; conversely, a model with a low goodness of fit indicates a poor fit.

[0157] In binary logistic regression, a commonly used goodness-of-fit metric is the Nagorno-R². 2 (Nagelkerke R) 2 This statistic is based on the Cox & Snell R. 2 The correction aims to bring its maximum value to 1, making it easier to interpret. Its calculation formula is as follows:

[0158] R2 Nagelkerke =[1-( L 0 / L 1 )] 2 / n / [1- L 0 ) 2 / n ]

[0159] in:

[0160] L 0 It is the likelihood function value of the zero model (containing only the intercept term);

[0161] L 1 It is the likelihood function value of the current model (including the intercept term and all explanatory variables);

[0162] n is the sample size;

[0163] Negorko R 2 The value of is between 0 and 1. The closer the value is to 1, the higher the goodness of fit of the equation; the closer the value is to 0, the lower the goodness of fit of the equation.

[0164] 1.2.2.3 Model Prediction Performance

[0165] The predicted protection results and actual protection results are compared statistically. The predicted protection rate and actual protection rate for different antibody titer ranges are calculated to evaluate the accuracy of the model's prediction.

[0166] 1.2.3 PD 50 Calculation method

[0167] First, the individual protection probability of each animal within each dose group is calculated using the logistic regression model described above. Then, the average protection probability of each dose group is calculated. The logarithmic value of the dose is used as the independent variable, and the logarithmic value of the average protection probability is used as the independent variable. 10 The value is the dependent variable, and linear regression is performed: logit(P) 平均 ) = a + b × log 10 (Dosage). When P 平均 When = 0.5, logit(P) 平均 )=ln[P 平均 / (1-P 平均 Substituting 0 into the equation 0 = a + b × log 10 (PD 50 Solving for PD yields... 50 Dosage, and then calculate the PD content per dose of vaccine. 50 quantity.

[0168] 1.2.4 Statistical Analysis and Validation

[0169] To verify the effectiveness of this calculation method (based on a binary logistic regression model), a quantitative evaluation of the consistency between the calculation results and those obtained using traditional challenge methods is necessary. Specifically, the PD (progression point) of multiple batches of vaccines calculated using this method will be used. 50 Values ​​compared to PD obtained through traditional virus challenge experiments 50 Pearson correlation analysis was performed on the values.

[0170] The Pearson correlation coefficient is used to measure the degree of linear correlation between two groups of continuous variables. Its calculation formula is as follows:

[0171]

[0172] in:

[0173] x i With y i Representing the first i The PD of batch vaccines obtained through traditional challenge experiments and this calculation method 50 value.

[0174] and These represent the PD obtained by the two methods respectively. 50 The sample mean of the values.

[0175] n represents the number of vaccine batches participating in the comparison.

[0176] The correlation coefficient r ranges from -1 to 1. If the r value is close to 1 and the statistical test is significant (p-value less than the significance level, such as 0.05), it indicates a strong positive linear correlation between the results obtained from the two methods, thus statistically verifying the reliability and consistency of this calculation method.

[0177] 2. Results

[0178] 2.1 Neutralizing antibody liquid-phase blocking ELISA detection method

[0179] 2.1.1 VLPs protein identification results

[0180] The purified and in vitro assembled target protein (type O VLPs) was identified by SDS-PAGE and Western Blot. SDS-PAGE results showed that the band size of the purified and assembled target protein was consistent with the theoretical value. Western Blot results showed that the VLPs specifically reacted with type O FMDV positive serum, and could be used as the reaction antigen in the liquid-phase blocking ELISA neutralizing antibody detection method for type O FMDV VLPs.

[0181] 2.1.2 Results of screening for optimal experimental conditions for liquid-phase blocking ELISA

[0182] This study employed cross-checkerboard titration and matrix titration methods. Statistical analysis of the results was conducted to screen for optimal experimental conditions for liquid-phase blocking ELISA. The optimal experimental conditions for the liquid-phase blocking ELISA method were determined as follows: optimal coating concentration of specific neutralizing antibody 1 µg / mL; optimal dilution of VLPs 5 µg / mL; optimal reaction concentration of specific neutralizing antibody 1.5 µg / mL; and HRP-labeled streptavidin 0.5 µg / mL.

[0183] 2.1.3 Determination of Judgment Criteria

[0184] Titer determination: Perform a liquid-phase blocking ELISA test using the above steps. Use half of the average OD value of the four wells of the VLPs antigen control as the baseline value. Set 50% of the baseline value as the cutoff value. If the OD value of the tested serum is exactly the same as the cutoff value, the neutralizing antibody titer is the corresponding dilution factor. If the OD value of the tested serum is between two dilution factors, the titer is the average of the antibody titers of the two adjacent wells.

[0185] The established method was used to test 400 samples of type O FMDV negative and positive sera, including 200 negative sera and 200 neutralizing antibody positive sera. The results were statistically analyzed and ROC curves were plotted. For the detection to be effective, the average OD value of the antigen control needed to be greater than 1.5. The titers of strong positive serum controls were all ≥2.7 log10, the titers of weak positive serum controls were ≥1.8 log10, and the titers of negative serum controls were <0.9 log10. Figure 1 A). ROC curve analysis showed that when the cut-off value was 1.65 log10, the AUC had a maximum area of ​​0.998 ( ). Figure 1 B) has the best sensitivity and specificity values, at 100% and 99.21% respectively, therefore the critical value is determined to be 1.65 log10 ( Figure 1 C).

[0186] 2.1.4 Results of the cross-test

[0187] The method established in this study was used to detect 35 positive sera from different swine diseases, including 3 positive sera from type O FMD and 32 positive sera from other swine diseases. After testing, the ratio of each serum sample to the positive sera was less than 1:8, indicating that the detection method has good specificity and no cross-reactivity. Figure 2 As shown.

[0188] 2.1.5 Concordance rate between VNT and liquid-phase blocking ELISA

[0189] A VNT titer of 1.65 was considered the threshold for distinguishing between positive and negative sera. The Pearson coefficient test was used to determine the correlation between NA-ELISA titers and VNT titers, with a P-value of 0.05 considered statistically significant. A total of 600 serum samples from cattle and pigs immunized with FMDV O VLPs were measured using Type O FMDV O VLPs liquid-phase blocking ELISA neutralizing antibody detection (Type O NA-ELISA) and VNT. The concordance rate between the Type O NA-ELISA and VNT results was 90.83% (545 / 600). The Pearson correlation coefficient between NA-ELISA and VNT results was calculated by comparing individual-level data. NA-ELISA results showed a statistically significant correlation with VNT results (r = 0.9722, P < 0.0001), as shown in Figure 3.

[0190] 2.2 PD 50 Alternative calculation methods

[0191] 2.2.1 Statistical Results of Serum Samples Neutralizing antibody titers were detected in 390 serum samples of foot-and-mouth disease type O using the established liquid-phase blocking ELISA method. The logarithmic value of the antibody titer was calculated. 10 The antibody titers were calculated and divided into "protected" and "unprotected" groups based on their challenge protection results. The average antibody titer and standard error were calculated (Table 1).

[0192] Table 1 Statistical results of serum samples

[0193] 2.2.2 Fitting of the Binary Logistic Regression Model

[0194] A binary logistic regression analysis was performed using Python programming to fit a logistic regression model of antibody titer (log10 transformation) and challenge protection rate. Figure 4 The figure shows the logistic regression model curve for foot-and-mouth disease type O.

[0195] Specifically, the model is: logit(P) = ln[P / (1-P)] = -6.240 + 4.238 × log 10 (Antibody titer)

[0196] 2.2.3 Testing the Binary Logistic Regression Model

[0197] (1) Significance test of logistic regression equation (likelihood ratio test) Based on the formula for the likelihood ratio statistic, calculate the zero model bias (-2 ln) respectively. L 0) and current model bias (-2ln L1 The results show that the likelihood ratio statistic... χ 2 The p-value was 154.190, and the p-value for the test was <0.001. Therefore, the null hypothesis was rejected, indicating that the model is highly significant overall. The calculation process is as follows:

[0198] The log-likelihood of the zero model is:

[0199] ln( L 0 ) =∑[y i ×ln(p) + (1-y i )×ln(1-p)]= n1×ln(p) + n0×ln(1-p)= -248.928

[0200] Where n1 is the number of y=1, n0 is the number of y=0, and n0+n1=390

[0201] The log-likelihood of the current model is: ln( L 1 ) =∑[y i ×ln( p i ) + (1- y i ) ×ln(1- p i )]= -171.833

[0202] in, p i = e (β0 + β1×x i ) / (1 + e (β0 + β1×x i ) )

[0203] so, χ 2 = -2 ln L 0 -(-2 ln L 1 = 154.190

[0204] (2) Significance test of logistic regression equation coefficients (Wald test) The Wald statistic was used as the test statistic to calculate the standard error and Wald value of the intercept and slope. The results showed that the p-values ​​of the intercept and slope were both less than 0.001, and the 95% confidence intervals did not contain 0, indicating that they had good significance and the model parameters were reliable (Table 2).

[0205] Table 2. Significance test results of the logistic regression equation coefficients.

[0206] (3) Goodness-of-fit test of the binary logistic regression equation: Based on the results of the logistic regression significance test, the log-likelihood value of the zero model (ln) L 0 The log-likelihood value of the current model is -248.928, and the log-likelihood value (ln) is -248.928. L 1 The value is -171.833, therefore, the value of R in Negorko is... 2 The statistical results are as follows:

[0207] R 2 Nagelkerke =[1-( L 0 / L 1 )] 2 / n / [1- L 0 ) 2 / n ]=[1-exp(2 / n(ln L 0 -ln L 1 )] / [1-exp(2 / nln L 0 ]

[0208] have to, R 2 Nagelkerke =0.453

[0209] Given the inherent variability and complexity of biological systems, and the fact that foot-and-mouth disease immune responses are influenced by multiple factors (strain differences, adjuvants, individual animal differences, etc.); and considering that antibody titer is only one of the relevant protective factors, not the sole determinant, model fit is often difficult to achieve a high level. Based on existing research experience, when... R 2 A value between 0.4 and 0.6 is generally considered to indicate a good fit. This model's... R 2 The value is 0.453, which falls within the generally accepted range, indicating that it has good explanatory power and good fit to the observed data.

[0210] 2.2.4 Model prediction performance varies depending on the calculated logarithm. 10 The predicted protection rate and actual protection rate within the titer range were used to calculate and determine that the overall prediction accuracy of the model reached 81.0%, with an AUC of 0.864, indicating that the model has good discriminative ability (Table 3).

[0211] Table 3 Comparison between model predictions and actual observations

[0212] Therefore, the logistic regression model performs well overall, exhibiting high statistical significance and a good goodness of fit. The model can be used to predict the probability of protection against foot-and-mouth disease vaccine, and the independent variable log... 10 Titer is a significant predictor of protection status.

[0213] 2.2.5 Application of the Binary Logistic Regression Model to PD 50 calculate

[0214] 2.2.5.1 Calculating the PD of a vaccine using a binary logistic regression model 50 The method for calculating the value is as follows:

[0215] Step 1: For each dose group (e.g., 1 dose, 1 / 3 dose, 1 / 9 dose), detect the neutralizing antibody titer (log). 10 (titer);

[0216] Step 2: Calculate the probability of protection (P) for each animal using a unified model;

[0217] Step 3: Calculate the mean protection probability (P) for each dose group. 平均 );

[0218] Step 4: Perform linear regression with the logarithmic dose value as the independent variable and the logit value of the p-mean as the dependent variable: logit(P) 平均 ) = a + b × log 10 (dose);

[0219] Step 5: When P 平均 When = 0.5, logit(P) 平均 Given ln[P / (1-P)]=0, solve for PD. 50 Dosage: 0 = a + b × log 10 (PD 50 Therefore, PD 50 = 10 (-a / b) Each dose of PD 50 Quantity = 1 / PD 50 dose.

[0220] 2.2.6 Data Derivation and Case Application: PD 50 Calculation process

[0221] Using actual test data from a batch of foot-and-mouth disease mineralized virus-like particle vaccines as an example, this demonstrates PD. 50 The complete calculation process for this batch of vaccines, based on the challenge protection results, involves calculating the PD (progression point).50 The value is 11.84 per dose (Table 4).

[0222] Table 4. Case Study: Neutralizing Antibody Detection Results and Protective Status

[0223] Note: "+" indicates protection against viruses, and "-" indicates no protection against viruses.

[0224] Step 1: Calculate the protection probability (P) value for each individual animal.

[0225] Taking animal 1 (dose group 1) as an example:

[0226] Its antibody titer value log 10 =2.55 Substituted into the logistic regression model equation

[0227] logit (P)= ln[P / (1-P)]= -6.240 + 4.238×log 10 (Antibody titer)

[0228] We get logit(P) = -6.240 + 4.238 × 2.55 = 4.567

[0229] Because logit(P) = ln(P / (1-P))

[0230] Then P= e logit (P) / (1+ e logit (P) )

[0231] Therefore, P = e 4.567 / (1+ e 4.567 ) = 96.263 / (1 + 96.263) = 0.990

[0232] By analogy, the protection probability value (P) and the average protection probability value (P0) of other individual animals can be calculated. 平均 (Table 5)

[0233] Table 5. Individual protection probability (P) and average protection probability (P0) of animals 平均 )result

[0234] Step 2: Calculate logit(P) at each dose point 平均 Values ​​(Table 6)

[0235] Taking a 1-head dose as an example, calculate logit(P) 平均 )value:

[0236] P 平均 =0.966 Substitute into the equation logit(P) 平均 )=ln[P 平均 / (1-P 平均 )]

[0237] We get ln[0.966 / (1-0.966)] = ln(28.412) = 3.343

[0238] Table 6. Logit(P) at each dose point 平均 ) value result

[0239] Step 3: Fitting the dose-response curve

[0240] The model is in the form: y = a + bx

[0241] Given x = log 10 (Dosage) = [0, -0.477, -0.954]

[0242] y = logit(P 平均 = [3.343, 2.124, 0.273]

[0243] Use linear regression to calculate the values ​​of a and b.

[0244] First, calculate and The average value.

[0245] = (0 + (-0.477) + (-0.954)) / 3 = (-1.431) / 3 = -0.477

[0246] = (3.343 + 2.124 + 0.273) / 3 = (5.740) / 3 = 1.913

[0247] Then, calculate and

[0248] (1) Calculate the difference:

[0249] for :

[0250] When x i When = 0, 0 - (-0.477) = 0.477;

[0251] When x iWhen the value is -0.477, then -0.477 - (-0.477) = 0;

[0252] When x i When the value is -0.954, then -0.954 - (-0.477) = -0.477;

[0253] for :

[0254] When y i When = 3.343, 3.343 - 1.913 = 1.430;

[0255] When y i When = 2.124, 2.124 - 1.913 = 0.211;

[0256] When y i When = 0.273, 0.273 - 1.913 = -1.640;

[0257] so,

[0258] = (0.477)(1.430) + (0)(0.211) + (-0.477)(-1.640) = 1.464

[0259] = (0.477) 2 + (0) 2 + (-0.477) 2 = 0.2275 + 0 + 0.2275 = 0.455

[0260] so,

[0261]

[0262]

[0263] so,

[0264] Fitting equation: logit(P) = 3.448 + 3.219 × log 10 (dose)

[0265] Step 4: Calculate PD 50

[0266] When P = 0.5, logit(P) = ln[P / (1-P)] = 0.

[0267] We get 0 = 3.448 + 3.219 × log 10 (PD50 )

[0268] log 10 (PD 50 = -3.448 / 3.219 = -1.071

[0269] so,

[0270] PD 50 = 10 -1.071 =0.0849 heads

[0271] so,

[0272] Each PD 50 Quantity = 1 / 0.0849 = 11.78

[0273] so,

[0274] Regarding vaccine SZ202201, PD 50 Approximately 0.0849 heads per dose, with each dose containing 11.78 PD. 50 This result is consistent with the PD obtained from traditional challenge experiments for this batch of vaccines. 50 The values ​​(11.84 / head) were basically consistent, with a relative error of 0.51%, which is within the acceptable range of methodology.

[0275] The above calculates PD 50 The calculation process can be easily performed using Excel. The relevant functions used when calculating each parameter using Excel are summarized in Table 7.

[0276] Table 7 Summary of corresponding functions in Excel

[0277] 2.2.7 Verification of the equivalence between the neutralizing antibody alternative method and the challenge method

[0278] The PD calculated using this neutralizing antibody replacement method 50 Values ​​compared to PD obtained from classic virus challenge experiments 50 A comparative analysis was conducted. The results showed that the Pearson correlation coefficient (r) between the two methods was 0.980, indicating a significant positive correlation between the alternative method and the challenge experiment. The logistic regression model based on neutralizing antibody titers was also used to assess vaccine efficacy (PD). 50 The predicted trend is consistent with the challenge experiment, indicating that the alternative method has a strong predictive ability for the challenge experiment. This means that the alternative method can accurately determine whether the vaccine efficacy meets the target (not less than 6 PD per dose). 50 A p-value < 0.001 indicates that the correlation is statistically significant.

[0279] The average relative error was 4.06%, which is within the acceptable range for bioassays. Larger relative errors in individual batches may be due to the following reasons: individual differences in animal immune responses may lead to uncertainty in the relationship between antibody titers and protective efficacy; batch-to-batch variation or operational errors when using liquid-phase blocking ELISA for neutralizing antibody detection (Table 8).

[0280] Table 8 Comparison of Alternative Methods and Challenge Experiment Results

[0281] 3. Methods for determining vaccine efficacy

[0282] Based on the results of this study, the following method was determined to be used to determine vaccine efficacy using neutralizing antibody titers:

[0283] Step 1: Detect neutralizing antibody titers in immunized animals

[0284] Fifteen healthy pigs, each weighing approximately 40 kg, were divided into three groups: a 1-dose group (5 pigs), a 1 / 3-dose group (5 pigs), and a 1 / 9-dose group (5 pigs). The pigs were inoculated intramuscularly behind the ear. Two unimmunized control pigs were included. Serum was collected 28 days after immunization, and the neutralizing antibody titers were detected using an established liquid-phase blocking ELISA method for foot-and-mouth disease neutralizing antibodies.

[0285] Step 2: Substitute the neutralizing antibody titer into logit(P) = -6.240 + 4.238 × log 10 The (antibody titer) formula is used to calculate the protection probability (P) for each animal.

[0286] Step 3: Calculate the average protection probability (P) for each dose group (1 dose, 1 / 3 dose, 1 / 9 dose). 平均 ) value, logit(P 平均 ) value and log 10 (dose);

[0287] Step 4: Fitting the dose-response curve

[0288] The model is in the form: y = a + bx

[0289] Given that x = log 10 (Dosage) = [0, -0.477, -0.954], y = logit(P 平均 );

[0290] in,

[0291] Step 5: Calculate PD 50

[0292] When P = 0.5, logit(P) = ln[P / (1-P)] = 0

[0293] Okay, log 10 (PD 50 ) = -a / b

[0294] Each dose of vaccine contains PD 50 =1 / (10 -a / b )

[0295] 4. Summary

[0296] The alternative method for evaluating the efficacy of foot-and-mouth disease vaccine based on liquid-phase blocking ELISA detection of neutralizing antibodies, established in this application, has been systematically validated and demonstrated to have high reliability and feasibility. This method constructs a quantitative relationship between neutralizing antibody titers and protection probability using a logistic regression model, providing a scientifically effective alternative to traditional challenge experiments.

[0297] The logistic regression model established in this application demonstrates excellent predictive performance. The likelihood ratio statistic χ² is 154.190 (P < 0.0001), and the intercept ( β 0 The Wald value of ) was 72.628 (P<0.0001), and the slope ( β 1 The Wald value for ) was 87.500 (P<0.0001), R 2 The statistical statistic was 0.453, the overall prediction accuracy of the model reached 81.0%, and the AUC was 0.864. In validation experiments with multiple batches of vaccines, the Pearson correlation coefficient between the alternative method and the challenge experiment results reached 0.980, and the average relative error was controlled within the acceptable range of 4.06% for bioassays, fully demonstrating the accuracy and reliability of the method.

[0298] First, by avoiding live virus handling, this application eliminates biosafety risks to some extent; second, it shortens the testing cycle by one month, significantly improving testing efficiency; furthermore, it eliminates the need for challenge experiments, thus meeting animal welfare and ethical requirements while significantly reducing costs. These features make it particularly suitable for batch testing and quality control applications of vaccines.

[0299] In summary, this application can accurately determine whether the vaccine efficacy meets the standard (not less than 6 PD per dose). 50 Therefore, this application can be used as one of the methods for evaluating the efficacy of foot-and-mouth disease virus-like particle vaccines.

[0300] The above examples illustrate the present invention only to aid in understanding it and are not intended to limit the scope of the invention. Those skilled in the art can make various simple deductions, modifications, or substitutions based on the principles of this invention.

Claims

1. A method for constructing a model to predict the protection probability of type O foot-and-mouth disease vaccine, characterized in that, Includes the following steps: S10. Select at least three batches of type O foot-and-mouth disease vaccine and immunize experimental animals with at least three dose groups of gradient dilution. Immunize at least five experimental animals with each dose group. Collect serum 28 days after immunization and conduct a virulent challenge experiment simultaneously to record the challenge protection results of each experimental animal. S20, using a liquid-phase blocking ELISA method based on foot-and-mouth disease neutralizing nanobodies, the neutralizing antibody titer of each serum sample was detected, and the neutralizing antibody titer was logarithmic. 10 Conversion; S30, in log 10 The converted neutralizing antibody titer is used as the independent variable X, and the challenge protection result is used as the dependent variable Y. A binary logistic regression analysis is performed, yielding the binary logistic regression model logit(P) = ln[P / (1-P)] = β 0 + β 1 ×log 10 ( X In ) β 0 and β 1 The value of ; where P is the protection probability, X For antibody titer, β 0 The intercept is... β 1 The slope; S40, Perform significance tests, coefficient significance tests, and goodness-of-fit tests on the binary logistic regression model to verify the model parameters. β 0 and β 1 This yields a model for predicting the protection probability of type O foot-and-mouth disease vaccine, known as the unified model. The experimental animal is a pig, and the model parameters are... β 0 =-6.240, β 1 =4.

238.

2. The method as described in claim 1, characterized in that, In step S10, experimental animals are immunized with three dose groups: 1 dose, 1 / 3 dose, and 1 / 9 dose, respectively. Alternatively, in step S10, at least ten batches of type O foot-and-mouth disease vaccine are selected; Alternatively, in step S10, a total of no less than 200 experimental animals may be immunized.

3. The method as described in claim 1, characterized in that, In step S40, the significance test is a likelihood ratio test, the coefficient significance test is a Wald test, and the goodness-of-fit test is a Nagelkerke R-squared test. 2 test.

4. The method as described in claim 1, characterized in that, The construction method further includes step S50 after step S40: calculating the PD based on the unified model... 50 The value compared to the PD obtained from traditional challenge experiments 50 Pearson correlation analysis was performed on the values.

5. A method for testing the in vitro efficacy of type O foot-and-mouth disease vaccine, characterized in that, Includes the following steps: S10, the O-type foot-and-mouth disease vaccine to be tested was immunized in experimental animals according to at least three graded dilution dose groups, with at least five experimental animals immunized in each dose group, and serum was collected 28 days after immunization; S20, using a liquid-phase blocking ELISA method based on foot-and-mouth disease neutralizing nanobodies, the titer of type O foot-and-mouth disease neutralizing antibodies in the serum of each experimental animal was detected; S30, the neutralizing antibody titer is log 10 After conversion, the model is substituted into the unified model constructed by the method described in any one of claims 1 to 4 to calculate the protection probability P of each experimental animal; S40, Calculate the average protection probability P for each dose group. 平均 And with the logarithm of dose as the independent variable and P as the variable. 平均 The logit value is used as the dependent variable for linear regression to fit the dose-response curve; S50, based on the dose-response curve, calculate the PD according to the dose corresponding to the mean protection probability of 0.

5. 50 Value, and based on the PD content of each dose of vaccine. 50 Quantity determines vaccine efficacy.

6. The method as described in claim 5, characterized in that, In step S50, when each dose of vaccine contains PD 50 A result of ≥6 indicates that the vaccine is considered to be of acceptable efficacy.

7. The method as described in claim 1 or 5, characterized in that, The liquid-phase blocking ELISA method includes: S21, the nanobody Ab2 that neutralizes foot-and-mouth disease virus was diluted to the working concentration with carbonate buffer, added to the ELISA plate, coated overnight at 4°C, washed with PBST and patted dry; S22, the serum to be tested is serially diluted 2-fold with PBST, mixed with an equal volume of PBST buffer containing VLPs, vortexed and mixed, and reacted overnight at 4°C to obtain serum-antigen mixture; S23, Add the serum-antigen mixture obtained in step S22 to the enzyme-labeled plate obtained in step S21, and react at room temperature; After washing with S24 and PBST, add biotin-labeled detection antibody and react at room temperature; After washing with S25 and PBST, HRP-labeled streptavidin was added and the reaction was carried out at room temperature. S26, after washing with PBST, added the substrate TMB for color development at room temperature, and terminated the reaction with a stop solution; S27, read the OD of each well using a microplate reader at a wavelength of 450 nm. 450 nm The value is used to calculate the neutralizing antibody titer based on preset judgment criteria.

8. The method as described in claim 7, characterized in that, In step S21, the working concentration is 1 µg / mL; Alternatively, in step S22, the final concentration of VLPs is 5 µg / mL; Alternatively, in step S24, the reaction concentration of the biotin-labeled detection antibody is 1.5 µg / mL; Alternatively, in step S25, the reaction concentration of HRP-labeled streptavidin is 0.5 µg / mL.

9. An in vitro potency testing system for type O foot-and-mouth disease vaccine, characterized in that, include: The input module is used to receive neutralizing antibody titer data from experimental animals immunized with different dose groups of the same vaccine; The probability calculation module is used to substitute the neutralizing antibody titer into the unified model constructed by the method as described in any one of claims 1 to 4 to calculate the individual protection probability of each experimental animal and the average protection probability of the dose group. The curve fitting module is used to perform linear regression with the logarithm of the dose as the independent variable and the logit value of the average protection probability as the dependent variable to fit the dose-response curve. The efficacy determination module is used to calculate the PD based on the dose-response curve. 50 The value was calculated and converted into the PD (PD) per dose of vaccine. 50 The quantity is considered sufficient; when the quantity is ≥6, the vaccine is deemed to be of acceptable efficacy.

10. A product for testing the in vitro efficacy of type O foot-and-mouth disease vaccine, characterized in that, include: Memory, used to store computer programs; A processor for implementing steps S30-S50 of the method as described in claim 5 by executing the computer program stored in the memory.