An ada detection method for multispecific monoclonal antibody drugs

Abstract

The present application belongs to the technical field of biological detection, and relates to an ADA detection method for a multispecific monoclonal antibody drug, comprising the following steps: S1. coupling the drug and each functional domain thereof to different fluorescence-coded microspheres respectively to form a coupled microsphere system; S2. mixing a sample to be detected with the coupled microsphere system and biotin-labeled drug to incubate, to form a microsphere-ADA-biotin-labeled drug complex; S3. detecting the complex by using a streptavidin-labeled fluorescent probe to obtain the fluorescence signal intensity corresponding to each microsphere; S4. calculating the ADA concentration of the drug and each functional domain based on the fluorescence signal intensity to obtain common parameters that can be compared with each other; and S5. determining the distribution proportion of the ADA of each functional domain in the total ADA according to the corrected common parameters. The present application solves the problem of traditional methods that need multiple experiments and complicated operation, significantly improves the detection efficiency, and reduces the experimental cost.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of biological detection, especially the detection of ADA of drugs, and particularly relates to an ADA detection method for multispecific antibody drugs. BACKGROUND

[0002] The development of multispecific monoclonal antibody drugs is a technical route of monoclonal antibody drug development, which is characterized in that the multispecific monoclonal antibody can specifically bind to multiple antigens or multiple epitopes of one antigen, and has multifunctionality and multispecificity compared with traditional unit point monoclonal antibody drugs, and has potential therapeutic advantages in the following aspects, such as mediating the killing of immune cells on tumors, activating the immune system of the body, preventing the generation of drug resistance, enhancing the specificity and targeting of antibodies on tumor cells, reducing the adverse reactions caused by off-target effects, and mediating stronger endocytosis. In recent years, more and more multispecific monoclonal antibody drugs have been approved for marketing, and play an important role in clinical treatment.

[0003] With the continuous development of the research and development of multispecific monoclonal antibody drugs, the number of clinical trials related thereto has increased by several orders of magnitude. In the process of clinical trials, it is extremely necessary to detect and characterize the ADA induced by the immunogenicity of the drug. Due to the special structure of the multispecific monoclonal antibody drug, it has multiple functional domains (domain) and other structures, such as Fc segment, and each structure of the drug theoretically has immunogenicity and can induce immune response of the body to generate corresponding anti-drug antibodies (abbreviated as ADA). Among them, the domain is an important functional structure for the drug to recognize the target and play a role, and the ADA induced by the domain will significantly affect the efficacy and safety of the drug, so researchers pay special attention to it. The influence of other structures on the function of the drug is also important, but is weaker than the domain and is not convenient to subdivide, so it is generally classified into a category, and the ADA induced by the other structures is not directly detected in the detection process, but the difference between the total ADA induced by the drug and the sum of the ADA induced by each domain is represented. In the whole research process, researchers pay great attention to the proportion of each ADA, so as to better evaluate the efficacy and safety of the drug.

[0004] The commonly used detection method is enzyme-linked immunoassay. For example, for the drug with bispecific monoclonal antibody, the sample positive in the drug ADA screening experiment needs to be subjected to ADA confirmation experiment again, and at least three experiments are required, including one drug ADA confirmation experiment and one ADA confirmation experiment for each of the two different domains of the drug. For multi-specific monoclonal antibody drugs, each additional domain will increase one confirmation experiment. Currently, multi-specific monoclonal antibody drugs with four domains have been developed on the market. With the breakthrough of technology and the popularization of market application, more domains will be developed in the future. The limitations of using enzyme-linked immunoassay method for ADA detection will be more obvious.

[0005] In addition to the above problems, in the current ADA detection of bispecific monoclonal antibody drugs and trispecific monoclonal antibody drugs, the inhibition rates of the ADA confirmation experiments of multiple domains are often significantly greater than the inhibition rate of the drug ADA confirmation experiment, and even the inhibition rate of the ADA confirmation experiment of a single domain is significantly higher than the inhibition rate of the drug ADA confirmation experiment. This indicates that the inhibition rate cannot be used as a parameter to evaluate the data results of the ADA confirmation experiments of each domain. In the entire ADA detection experiment, there is a lack of a common parameter in the ADA detection experiments of each domain and the drug to evaluate and compare the data results and analyze the composition ratio of the ADA. The domain of the drug is the key structure of the drug efficacy and safety, and only the ADA detection results that can be compared with each other can effectively evaluate the differences in immunogenicity of each domain, and further evaluate the efficacy and safety of the drug.

[0006] It is speculated that the above phenomenon may be caused by the following three reasons: first, there is a certain difference between the ADA confirmation experiment of each domain and the ADA confirmation experiment of the drug, and the results in different reaction systems may not be directly compared with each other; second, multiple detections may produce detection errors, and after the mutual superposition of the errors, the error will exceed the acceptable range, thereby causing the abnormality of the results, and the more the detection times, the more significant the phenomenon; third, although the domain of the drug structure is the same as the amino acid sequence of the single domain and the structure and function are similar, due to the existence of steric hindrance, the natural conformation is necessarily different, resulting in a certain difference in the detection of ADA. For the consideration of scientificity, in the screening experiment, only the ADA screening experiment of the drug should be carried out, and the ADA screening experiment of different domains of the drug should also be carried out, so as to avoid false negative results and better compare the data with each other. However, the positive rate of ADA is generally between 5% and 20%, and if the ADA screening experiment of different domains is increased, the detection workload will be significantly increased. At present, most detection companies will not adopt the above screening method due to the comprehensive consideration of cost and other factors, and therefore there is a possibility of screening leakage in the ADA detection, resulting in false negative results.

[0007] In view of the large detection workload, the existing multi-association detection technology can be used, but due to the competition between the drug and each domain, the detection results of separate detection and joint detection are significantly different, so it is difficult to realize the joint detection of the ADA of the drug and each domain, and it is necessary to separate the ADA detection of the drug and the ADA detection of each domain. The above method can only achieve the purpose of partially reducing the detection workload. At the same time, due to the still unbridgeable differences in the reaction conditions of separate detection, it is still impossible to obtain common parameters and realize the mutual comparison of data. SUMMARY

[0008] In view of the special situation of the ADA detection of the multispecific antibody drug and the shortcomings of the current detection method, a kind of ADA detection method for multispecific antibody drug is developed to solve the problems of large experimental detection amount, large sample consumption, data results cannot be jointly analyzed and possible screening leakage in the prior art, and realize the accurate and efficient detection of the ADA of the multispecific monoclonal antibody drug.

[0009] The technical scheme adopted by the present application is as follows:

[0010] A kind of ADA detection method for multispecific monoclonal antibody drug, comprising the following steps:

[0011] S1. coupling the drug and each functional domain (domain) thereof to different fluorescently coded microspheres respectively to form a coupled microsphere system;

[0012] S2. mixing the sample to be tested with the coupled microsphere system and biotin-labeled drug to form a microsphere-ADA-biotin-labeled drug complex;

[0013] S3. detecting the complex by streptavidin-labeled fluorescent probe to obtain the fluorescence signal intensity corresponding to each microsphere;

[0014] S4. calculating the ADA concentration of the drug and each domain based on the fluorescence signal intensity and correcting the influence of the competition reaction by weighted calculation to obtain a common parameter that can be compared with each other;

[0015] S5. determining the distribution ratio of the ADA of each domain in the total ADA according to the corrected common parameter.

[0016] The detection principle of the present application is as follows: after pretreatment, the sample is specifically combined with polystyrene magnetic microspheres coupled with the drug or different domains of the drug and biotin-labeled drug to form a complex, i.e. microspheres coupled with the drug or different domains-ADA-biotin-labeled drug, and then washed multiple times by a magnetic separation device, combined with streptavidin-labeled phycoerythrin (SA-PE) to form the final immune detection complex, i.e. polystyrene magnetic microspheres coupled with the drug or different domains-ADA-biotin-labeled drug-streptavidin-labeled phycoerythrin. Through the detection of a multi-color flow cytometer or a liquid suspension chip detector, the corresponding detection results can be obtained, wherein different microspheres represent the detection of the ADA of the drug or different domains (wherein the drug represents the total ADA), the fluorescence intensity of phycoerythrin is proportional to the concentration of the ADA of the drug or different domains, the concentration of the ADA of the drug and each domain is calculated based on the fluorescence signal intensity, and the distribution ratio of the ADA of each domain in the total ADA is determined by correcting the influence of the competition reaction by weighted calculation.

[0017] As a preferred, the weighted calculation includes:

[0018] (i) correcting the detection result of the drug ADA by a pre-established influence parameter table;

[0019] (ii) using the formula:

[0020] C n % C n / (A*C 药 )*100%,

[0021] wherein C 药 represents the ADA detection result of the drug; C n represents the ADA detection result of the nth domain; A represents the influence parameter at the current ADA ratio; C n represents the ratio of the ADA induced by the nth domain in all ADAs. When the actual ratio has no corresponding data in the influence parameter table, the actual ratio is taken as the X axis and the corresponding influence parameter A is taken as the Y axis, two-point linear fitting is performed, and the actual ratio is substituted into the fitting formula to obtain the corresponding influence parameter A.

[0022] In the scheme of the present application, the ADA results of each domain are calculated first, and the ratio is calculated, the influence parameter A is determined by looking up the influence parameter table, and the reaction results of the drug-coupled microspheres are corrected to the optimal state without competition. After correction, the concentration of ADA can be used as a common parameter for the entire ADA detection experiment for comparing the reaction results of each microsphere and the ratio calculation.

[0023] The weighted calculation is to correct the competition in the entire detection process, thereby minimizing the influence on the detection results, and obtaining a common parameter that can be compared with each other, i.e., the concentration result of ADA.

[0024] The competition reaction mainly comes from two sources. The first is that the drug-coupled microspheres will also compete with the domain-coupled microspheres to bind the domain ADA, resulting in obvious competition. The second is that the affinity between the drug and the ADA of different domains is different, and the number of sites on the drug-coupled microspheres that can bind with the ADA is limited, so there is obvious competition when the ADAs of different domains specifically bind with the drug-coupled microspheres. The concentration and distribution ratio of ADAs will affect the detection results and are not uniform.

[0025] For the first kind of competition reaction, the present application takes into account the characteristics of liquid phase reaction: each microsphere is a reaction individual, and in the detection process, the same kind of microspheres also compete with each other to bind ADA, resulting in competition reaction and finally reaching equilibrium. When the binding capacity of the microspheres coupled with drugs and the microspheres coupled with domains to the ADA of the domain is consistent, the competition reaction between different kinds of microspheres can be converted into the competition reaction between the same kind of microspheres, thereby eliminating the competition reaction between different kinds of microspheres. Based on this principle, the present application takes the reaction performance of the microspheres coupled with drugs as a reference to adjust the concentration of each domain when coupled with microspheres. After a large number of tests and data evaluation, the microspheres coupled with domains with consistent reaction performance with the microspheres coupled with drugs are obtained, thereby avoiding the influence of this kind of competition reaction. It should be noted that the microspheres mentioned here have consistent performance, which is when the ADA is a single domain ADA; when the ADA is a mixed ADA, the performance of the microspheres is significantly different. This method only eliminates the competition reaction between two kinds of microspheres when binding to the ADA of one domain.

[0026] The data evaluation method is paired sample T test, which avoids the misjudgment caused by human intuition.

[0027] For the second kind of competition reaction, the processing method of the present application is to establish different ADA models to determine the influence of different concentrations and different proportions of ADA on the microspheres coupled with drugs, and finally to establish an influence parameter table to correct the detection results of drug ADA and minimize the deviation caused by such competition reaction.

[0028] As a preferred, the influence parameter table is established by the following steps: (1) configuring high concentration standard of different proportion combination of domain ADA: for a drug containing N domains, configure standard covering all proportion combinations, wherein the domain proportion coefficient of the kth combination is P k1 , P k2 , P k3 ... P kN , and satisfies P k1 + P k2 +... + P kN = 10, P k1 , P k2 , P k3 ... P kNTake the natural numbers in 0-10 in turn; (2) Dilute the standard samples of each proportion by gradient, obtain at least 7 concentration levels of detection samples; (3) Obtain the sum of the theoretical fluorescence signal values (X1+X2+...+XN) by detecting the ADA standard samples of each domain separately, and obtain the measured fluorescence signal value Y by joint detection of the total standard sample; (4) Calculate the ratio A=(X1+X2+...+XN) / Y under each proportion and concentration; (5) Repeat steps (1)-(4) at least 20 times, and after processing according to the outlier rejection rule, calculate the average value of the ratio A under the same proportion, and summarize it into the influence parameter table.

[0029] Note: According to the Technical Guidelines for Drug Immunogenicity Research, the detection sensitivity of IgG and IgM type anti-drug antibodies should generally reach 100 ng / mL. Therefore, the standard concentration of the detection sample at the lowest concentration level should be around 100 ng / mL. According to the 2-fold dilution, the standard concentration of the detection sample at the highest concentration level should be above 3200 ng / mL.

[0030] The influence parameter table is mainly set according to the ADA proportion of different domains. Although ADA of the same proportion and different concentrations also affects the detection results, through a large amount of data analysis, the average value of the influence parameters between different concentrations of ADA can be calculated to represent it uniformly. Taking a bispecific monoclonal antibody as an example, 11 high-concentration ADA standard samples are configured according to Table 1 below.

[0031] Table 1

[0032]

[0033] After gradient dilution of the high-concentration standard samples in the table, ADA standard samples of different concentrations under various proportions are obtained. According to the detection results, the ratio of the theoretical value to the test value is calculated, and the average value of the ratio under the same proportion is calculated, which is the influence parameter under this proportion level. Repeat the test 20 times, eliminate outliers, and calculate the average value of each test to summarize the final influence parameter table.

[0034] Similarly, 66 high-concentration ADA standard samples are required for a trispecific monoclonal antibody, and 66 influence parameters are obtained; 286 high-concentration ADA standard samples are required for a tetraspecific monoclonal antibody, and 286 influence parameters are obtained.

[0035] 66 high-concentration ADA standard samples are configured for a trispecific monoclonal antibody according to Table 2 below.

[0036] Table 2

[0037]

[0038] As preferred, the outlier rejection rule is to set a ratio D / R, where D = max-min, or D = min-min, and R is the range of all observations. If D >= 1 / 3R, the extreme value should be rejected. It should be noted that after each rejection of outliers, the above algorithm needs to be re-run to recalculate the corresponding results. The final rejected outliers should not exceed 5% of all data.

[0039] The common parameters require the use of standards, generally conventional standards or threshold standards, to reasonably quantify or semi-quantify the detection results. All standards are mixed solutions obtained by mixing purified specific antibodies of each domain at a certain concentration. The specific mixing concentration needs to be determined by considering the ADA detection sensitivity of different domains. Generally, the lowest concentration of the standard of the domain with the worst ADA detection sensitivity is used as the basis for setting the mixed standard concentration level. The threshold standard refers to the standard whose detection signal value is the cut-off signal value. The ratio of the detection signal value of the sample to the cut-off signal value is defined as AI, and the final test result of the sample is output in the form of AI.

[0040] As preferred, the microspheres include but are not limited to magnetic or non-magnetic coded microspheres of polystyrene, agarose, silica or high molecular copolymer materials, preferably polystyrene magnetic microspheres.

[0041] By implementing the above technical solutions, compared with the prior art, the present application has the following beneficial effects:

[0042] 1. Combined detection replaces multiple experiments: integrating drug and ADA detection of all domains into a single reaction (achieved by multi-fluorescent coded microspheres), completely solving the problem of multiple experiments and complicated operation in traditional methods; significantly improving detection efficiency and reducing experimental cost.

[0043] 2. By uniformizing microsphere performance, correcting affinity deviation with parameter table, and uniformly evaluating ADA distribution ratio, the ADA competition effect of the drug and each domain is quantified, and comparable concentration parameters are output, completely solving the problem of incomparable data and improving the scientificity of the results.

[0044] 3. The combined detection system simultaneously captures total ADA and domain-specific ADA, avoiding the missing screening caused by only detecting total ADA in traditional methods.

[0045] In summary, the problems of large experimental detection amount, large sample amount, inability to jointly analyze data results, and possible missing screening in the prior art are solved, and accurate and efficient detection of ADA of multi-specific monoclonal antibody drugs is achieved. DETAILED DESCRIPTION

[0046] The technical solutions of the present application will be described clearly and completely in combination with the embodiments.

[0047] Example 1: Establishment of a method for detecting ADA of a bispecific monoclonal antibody drug

[0048] 1.1, Sample MRD and acidification: 10 μL of sample per well was added to a dilution plate, followed by the addition of 190 μL of 0.472% H3PO4 per well, and the plate was sealed with a sealing film and incubated at room temperature with shaking at 800 rpm for 15-20 minutes.

[0049] 1.2, Microsphere washing: 100 μL of mixed conjugated drugs or 3 kinds of polystyrene magnetic microspheres of different domains of the drug were added to each well of a 96-well plate, followed by magnetic separation for 2 min, and the supernatant was removed.

[0050] 1.3, Sample incubation: 100 μL of biotin-labeled drug (concentration of 0.5 μg / mL) and 8 μL of 1M Tris solution were added to each well, and 50 μL of acid-treated sample was transferred to the well, and incubated at room temperature with shaking at 800 rpm for 1.5-2 hours.

[0051] 1.4, Washing after incubation: the 96-well plate was placed on a magnetic separator for 2 min, the supernatant was removed, 150 μL of PBS containing 0.05% tween-20 was added to each well, and the magnetic separation and washing were repeated 3 times.

[0052] 1.5, SA-PE incubation: 100 μL of streptavidin-labeled phycoerythrin (SA-PE) was added to each well, and incubated at 37°C with shaking at 1000 rpm for 30 min;

[0053] 1.6, Washing after incubation: repeat step 1.4;

[0054] 1.7, Machine reading: 100 μL of PBS containing 0.05% tween-20 was added to each well, and shaken at room temperature at 1000 rpm for 30 s, and then detected on a multi-color flow cytometer or a liquid suspension chip detection platform.

[0055] Example 2: Evaluation of the difference between different conjugated microspheres of a bispecific monoclonal antibody drug

[0056] 2.1, The Domain I standard was gradiently diluted using mixed serum to obtain a standard curve with a concentration range of 5000.00 ng / mL, 2500.00 ng / mL, 1250.00 ng / mL, 625.00 ng / mL, 312.50 ng / mL, 156.25 ng / mL, 78.13 ng / mL, and was denoted as a standard curve, with mixed serum as a blank control.

[0057] 2.2, The polystyrene magnetic microspheres coupled with drug and the polystyrene magnetic microspheres coupled with domain I were used respectively to detect the standard curve, and the two kinds of microspheres were also used to detect the standard curve in combination. The specific detection method was the same as that in Example 1 except that the microspheres used were different. The data were recorded as detection results A, B, C1 and C2 respectively.

[0058] 2.3, The data results are shown in the following table. The paired sample T test (two-tailed test) was performed on A and B, C1 and C2, A and C1, and B and C2 of the above four groups of data to compare the differences between the four groups of data. The results showed that there was no significant difference between the two kinds of microspheres. The detailed data results are shown in Table 3.

[0059] Table 3

[0060]

[0061] 2.4, The difference between the polystyrene magnetic microspheres coupled with drug and the polystyrene magnetic microspheres coupled with domain II was compared according to the same method. The results showed that there was no significant difference between the two kinds of microspheres. The detailed data results are shown in Table 4.

[0062] Table 4

[0063]

[0064] Example 3: Calculation of influence parameter table for ADA detection of bispecific monoclonal antibody drug

[0065] 3.1, The ADAs of Domain I and Domain II were mixed at a ratio of 5:5, and then the mixed serum was used to dilute the standard at a gradient to obtain a standard curve with a concentration range of 10000.00 ng / mL, 5000.00 ng / mL, 2500.00 ng / mL, 1250.00 ng / mL, 625.00 ng / mL, 312.50 ng / mL, 156.25 ng / mL, recorded as total standard curve. The ADA concentration corresponding to the highest standard curve point of 2 domains here is 5000.00 ng / mL, and the total ADA concentration is 10000.00 ng / mL after addition. The mixed serum was used as a blank control.

[0066] 3.2, using mixed serum, respectively, for each domain corresponding to the ADA gradient dilution, the standard curve concentration range of each domain corresponding to the ADA is 5000.00 ng / mL, 2500.00 ng / mL, 1250.00 ng / mL, 625.00 ng / mL, 312.50 ng / mL, 156.25 ng / mL, 78.13 ng / mL, respectively, mark curve 1 and mark curve 2, with mixed serum as blank control.

[0067] 3.3, according to the method of example 1, the total standard curve, standard curve 1 and standard curve 2 were detected respectively.

[0068] 3.4, the data analysis results are shown in table 5 below: test value and theoretical value are fluorescence value results (MFI)

[0069] Table 5

[0070]

[0071] 3.5, repeat the above experiment test 20 times, the data statistics and analysis results are shown in table 6:

[0072] Table 6

[0073]

[0074] 3.6, according to the same table 7, 11 kinds of high concentration ADA standard preparation

[0075] Table 7

[0076]

[0077] 3.7, according to the steps of 2.2, 2.3, 2.4, 2.5, test the influence parameters A of 11 kinds of different ADA ratio, establish the influence parameter table of the double specificity monoclonal antibody, see table 8:

[0078] Table 8

[0079]

[0080] The above describes the present application and its embodiments, which are not limited. The actual structure is not limited to this. In general, if a person skilled in the art is inspired by it, without departing from the purpose of the invention, without creative design, similar structure and examples of the technical scheme should belong to the protection scope of the present application.

Claims

1. A method for detecting ADA (adverse antibody-dependent agonist) against multispecific monoclonal antibody drugs, characterized in that, Includes the following steps: S1. The drug and its various domains are coupled to microspheres with different fluorescent codes to form a coupled microsphere system; S2. The sample to be tested is mixed with the coupled microsphere system and the biotin-labeled drug and incubated to form a microsphere-ADA-biotin-labeled drug complex; S3. The complex was detected using a streptavidin-labeled fluorescent probe to obtain the fluorescence signal intensity corresponding to each microsphere; S4. Calculate the drug and ADA concentrations in each domain based on the fluorescence signal intensity, and correct for the influence of competing reactions through weighted calculations to obtain common parameters that can be compared with each other; S5. Determine the distribution ratio of ADA in the total ADA of each domain based on the corrected common parameters; In step S4, the weighted calculation includes: (i) Correct the detection results of drug ADA by using a pre-established table of influence parameters; (ii) Using the formula: C n %=C n / (A*C 药 )*100% Where C 药 This indicates the ADA test results for the drug; C n Represents the ADA detection result for the nth domain; A represents the influence parameter under the current ADA ratio; C n % represents the proportion of ADA generated by the nth domain in all ADA; The influence parameter table is established through the following steps: (1) Configuring high-concentration standards with different ADA ratios of different domains: For drugs containing N domains, configure standards with different ratios of different domains; (2) Perform serial dilutions of the standard at each ratio to obtain at least 7 concentration levels of test samples; (3) Obtain the sum of theoretical fluorescence signal values ​​(X1 + X2 + ... + XN) by individually detecting ADA standards in each domain; obtain the measured fluorescence signal value Y by jointly detecting the total standard. (4) Calculate the ratio A = (X1 + X2 + ... + XN) / Y for each proportion and concentration; (5) After processing according to the outlier removal rules, the average value of the ratio A under the same proportion is calculated and summarized into an impact parameter table.

2. The ADA detection method according to claim 1, characterized in that, The standard products are configured in the following proportions, and the domain proportion coefficients for the k-th combination are P. k1 P k2 P k3 ... P kN Satisfying P k1 + P k2 + ... + P kN = 10, and P k1 P k2 P k3 ... P kN Take natural numbers from 0 to 10 in sequence, where N is the number of domains.

3. The ADA detection method according to claim 1, characterized in that, The outlier removal rule is as follows: set a ratio D / R, where D = maximum value - second largest value, or D = minimum value - second smallest value, and R is the range of all observations. If D ≥ 1 / 3R, the outlier is deleted.

4. The ADA detection method according to claim 1, characterized in that, When there is no corresponding data for the actual proportion in the influence parameter table, take the proportions at both ends of the actual proportion as the X-axis and the corresponding influence parameter A as the Y-axis, perform a two-point linear fit, and substitute the actual proportion into the fitting formula to obtain the corresponding influence parameter A.

5. The ADA detection method according to claim 1, characterized in that, Repeat steps (1)-(4) at least 20 times before proceeding to step (5).

6. The ADA detection method according to claim 1, characterized in that, Multispecific monoclonal antibody drugs are bispecific monoclonal antibody drugs, trispecific monoclonal antibody drugs, or tetraspecific monoclonal antibody drugs.

7. The ADA detection method according to claim 1, characterized in that, The microspheres are polystyrene magnetic microspheres.

Images