Analysis of immunogenicity of therapeutic protein drugs

TextSunshine Demei

The immunogenicity of a therapeutic protein drug refers to the ability of the therapeutic protein and/or its metabolites to induce an immune response or immune-related events against itself or related proteins.

Immunogenicity research has always been a difficult point in drug evaluation, involving immunology, pharmacology, toxicology, clinical medicine, statistics and other disciplines.

The formation mechanism of immunogenicity and its influencing factors

1.

Figure 1 Schematic diagram of the FDA on the immunogenicity production mechanism of therapeutic proteins

The general process of cellular immunity includes the differentiation of T cells and the release of cytokines, which can cause changes in the ratio of different T cell subtypes and changes in peripheral cytokine levels.

Humoral immunity mainly works through anti-drug antibodies (ADAs) produced by B cells against therapeutic proteins.

T cell dependent pathway

The immune response generated by the T cell-dependent pathway is characterized by high intensity and long duration.

T cell-independent pathway

Without the participation of T cells, the body can also produce ADA for therapeutic protein aggregates, that is, a non-T cell dependent pathway.

Figure 2 The production mechanism of anti-drug antibodies

2.

Table 1 Factors affecting immunogenicity

2.

The activation state of the patient's immune system is also related to the occurrence of immunogenicity.

Whether there is a history of exposure to related drugs also affects the occurrence of immunogenicity.

2.

Therapeutic protein is regarded as a foreign substance by the body's immune system.
The efficacy of the drug is usually determined by the integrity of the epitope sequence on its structure, and the oxidation, deamidation, deamination, degradation, conformation and other structural changes and sequence differences of the drug All are considered to be risk factors for immunogenicity, because they can change the conformation of proteins, aggravate the aggregation of drugs, or change the kinetics of antigen uptake, processing and presentation of antigen peptides.
The chemical changes of the above-mentioned drugs are similar to biotransformation and may occur in damaged physiological environments, such as inflamed tissues.
Structural changes in recombinant endogenous protein drugs will also break immune tolerance, leading to the production of ADA.

Glycosylation in the structure of therapeutic proteins can also affect the immunogenicity of the drug.
The post-translational modification of the therapeutic protein is determined by the expression system or cell line of the selected product.
At present, most of the products under research or on the market use Chinese hamster ovary cell lines (CHO).
Unlike other cell lines such as yeast and insect cell lines, the glycosyl structure of products obtained from mammalian cell lines is considered to be closer to the natural human glycosyl structure.
Maintaining the same glycosylation modification as human is particularly important for recombinant human protein therapeutics.
The type of glycosyl not only affects the function of the therapeutic protein, but also affects the stability of the product, the folding of the protein, and the optimization of biophysical characteristics.
Therefore, the glycosylation structure needs to be detected during the production process.
However, although some glycosylated structures are antigenic, their aggregation is reduced and immunogenic epitopes are blocked from time to time.

Aggregate

Aggregate is used to describe the formation of therapeutic protein polymer, which can be dissolved in solution or formed into insoluble particles.
The ideal situation for therapeutic proteins is that they exist in their smallest natural and uniformly dispersed form.
However, they usually form particles that are as small as a dimer structure that is large enough to be visible under a microscope.
Moreover, various factors, from the modification of the chemical structure to environmental factors such as storage problems, concentration, pH, ionic strength, freezing and thawing, and mechanical exercise, will affect the existence of therapeutic proteins.
Although aggregation has a high risk of immunogenicity, its interaction mechanism with the immune system, as well as their size, number, structure, and number effects are unknown, and further research is needed.

Impurity

Impurities are another important factor and have adjuvant characteristics.
If present in the final product, even if the therapeutic protein drug itself does not produce classic immunogenicity, impurities may also cause immunogenicity.
These impurities may be derived from host cell components, resins on the chromatographic column, enzymes, endotoxins, bacterial DNA, and monoclonal antibodies used for purification may remain in the final product.
They can activate pattern recognition receptors (PRR), produce pro-inflammatory cytokines, and increase the processing and presentation of antigens by APC.
It should be emphasized that the non-specific immune response caused by pollutants can be used as a costimulatory factor to break immune tolerance, increase the incidence of ADA produced by the body, and increase the titer level.
However, some studies have shown that in some monoclonal antibody products, the host cell level is higher, but it does not increase the production of immunogenicity.
Immune reactions caused by impurities are generally skin reactions, allergic reactions, and serological diseases.
With the application of mammalian cell lines, as well as the improvement of purification technology and other downstream processes, microbial-related pollution can be controlled to the lowest level.
This is a key link in drug development.
For example, host cell protein is a routine monitoring indicator at different stages of production.

other

The composition of the stabilizer is also particularly critical.
A typical example is that the protein stabilizer of the Epogen preparation changes from human serum albumin to PS80 (polysorbate) to produce ADA, which causes serious side effects of pure red blood cell aplasia.

The interaction with the container, or the interaction with the various parts encountered in the potting system may produce changes in protein structure and produce immunogenicity.
Interaction with glassware or air may denature the protein, cause protein aggregation, and may produce immunogenicity.
Therefore, in order to avoid the production of immunogenicity, a material that does not cause protein drugs to aggregate should be selected as the container.

2.
3 Treatment-related factors

Route of administration

The route of administration can affect immunogenicity.
Because subcutaneous administration may be easily ingested and processed by DCs, it is easier to produce immunogenicity than intravenous administration; however, studies have found that 16 non-immunomodulators are administered subcutaneously, and the incidence of ADA is less than 15%.
The mucosal and pulmonary routes of administration also produce immunogenicity.
Usually high-dose drugs produce low immunogenicity, but it should be noted when interpreting data that it may be due to the presence of high-dose drugs in the circulation that interfere with detection or the level of anti-drug antibodies will decrease with the increase of immune complexes.

Dosing frequency

Repeated or intermittent administration is more likely to induce immunogenicity than single-use monoclonal antibodies.

Clinical risk of immunogenicity

The reason why we should pay attention to immunogenicity is because once the therapeutic protein has an immune response, it may bring risks, ranging from affecting the efficacy of the drug, and threatening the life of the patient.
When immunogenicity is unavoidable, it is particularly important to fully understand the possible risks of immunogenicity, to monitor the immunogenic response in time, and to intervene early.
The United States Pharmacopoeia General Chapter 1106 summarizes the possible clinical risks of immunogenicity.

Table 2 Factors affecting the clinical consequences of immunogenicity

From the perspective of clinical manifestations, the immunogenicity may have no obvious impact on the patient, may affect the pharmacokinetics or efficacy of the drug, and may also bring additional safety issues.
In 2016, the FDA summarized and published 121 biological products approved at the time, 89% of which had immunogenic reactions, 49% of which affected the efficacy and 60% had safety-related issues.

Figure 3 Statistical analysis of the incidence of immunogenicity and clinical impact of listed drugs

Immunogenicity evaluation strategy

Based on the effectiveness and safety issues that may be caused by the body's immune response caused by the therapeutic protein, it is necessary to evaluate and monitor the immunogenicity of the therapeutic protein.

The evaluation of immunogenicity can range from theoretical simulation calculations, to pre-clinical in vivo and in vitro experimental predictions, to direct monitoring of the subject's immune response after clinical administration.

Computer simulation calculation is mainly based on T cell epitope prediction to evaluate the immunogenicity of biopharmaceuticals.
There are many tools for T cell epitope prediction.
The FDA has developed a TCPro prediction tool for T cell epitope prediction to assist in immunogenicity assessment.

Figure 4 TCPro work flow chart

The immunogenicity in vitro experiment is mainly evaluated by detecting the cytokines released by the activation of immune cells.
MHC II antigen binding is also an optional in vitro immunogenicity evaluation method.

The best way to reflect the immunogenicity of a drug is to observe the response of the immune system after administration.
Before administration to humans, the immune responses of transgenic animals and non-human primates are most similar to those of humans, and they are now more models of immunogenicity analysis in vivo.
The advantages and disadvantages of various preclinical immunogenicity evaluation methods are summarized as follows.

Table 3 Preclinical immunogenicity evaluation tools

Although various preclinical methods for predicting immunogenicity can provide more information, the correlation with the true immune response of clinical subjects has not yet formed an industry consensus.
Direct detection of anti-drug antibodies in the serum of clinical subjects is still The currently recognized immunogenicity evaluation method is also the regulatory requirement of the drug administrations of various countries.
At present, FDA, USP, EMA, and NMPA generally recommend the detection of anti-drug antibodies to evaluate the immunogenicity of therapeutic proteins.
The anti-drug antibody test is also different from other quantitative tests, but is based on a hierarchical analysis strategy to comprehensively evaluate the production of anti-drug antibodies.
At present, the industry-recognized analysis level includes three levels of screening-confirmation-identification.
In the third analysis level-identification level, the current research is mostly titer analysis and neutralizing antibody analysis.

Figure 5 Immunogenicity evaluation strategy

The screening experiment is the beginning of the entire level of analysis.
All potential positive samples should be detected as much as possible, and a certain amount of false positive samples (generally 5%) should be allowed.
The result of the screening experiment is judged based on the statistical method.
The serum of the unexposed population is used as a reference to set the cut-off value of positive samples, and the samples higher than the cut-off value are judged as positive.

Since the screening experiment naturally has 5% false positive samples, the positive samples obtained in the screening experiment need to be further confirmed.
The confirmation experiment generally adopts the same detection method as the screening experiment, and a certain amount of drug is added to compete with the anti-drug antibody during the sample detection process.
If anti-drug antibodies are present, the detection signal will decrease significantly.
In order to improve the credibility of positive samples, the false positive rate of confirmation experiments is generally set at 1%.

The positive sample obtained after screening and confirmation is the anti-drug antibody positive sample, and the concentration level of the anti-drug antibody can be semi-quantitatively detected-titer experiment according to research needs.
The titer experiment and the screening experiment use the same detection method, and the maximum dilution factor of the sample detection value higher than the cut-off value is used as the sample titer.

In addition, in order to evaluate whether the produced anti-drug antibodies affect the biological activity of the drug, neutralizing antibody testing is required.
The detection method of neutralizing antibody must be designed based on the mechanism of action of the drug.

Anti-drug antibody detection method

Immunogenicity research mainly focuses on the detection and characterization of anti-drug antibodies, and data on the incidence, titer, duration and neutralization ability of anti-drug antibodies should be obtained.

The detection of anti-drug antibodies can be done using a variety of biological analysis platforms, including ELISA (Enzyme-linked immunosorbent assay), ECL (Electrochemiluminescence), RIA (Radioimmunoassay), and SPR (Surface plasmon resonance).
In addition, immunoassays for screening and confirming ADAs can be performed using a variety of formats and detection systems, each of which has its own advantages and disadvantages.

Figure 6 Principles of various anti-drug antibody detection platforms

In addition, human ADA is a complex analyte.
They are usually polyclonal, bind to different regions, and have different isotypes and binding affinities.
Therefore, the selection of ADA analysis methods should be based on their specific binding regions as much as possible.
To fully evaluate the true immunogenicity of the drug.

In the choice of analysis platform, you can consider the advantages and disadvantages of the analysis methods of each platform listed in the table below to avoid possible false negatives.

Table 4 Advantages and disadvantages of various anti-drug antibody detection platforms

Neutralizing antibody testing is generally designed based on the mechanism of action of drugs.
USP1106 classifies common drug action mechanisms and gives recommended method selections.
In order to more comprehensively reflect the mechanism of action of drugs, the selection of neutralizing antibody testing is generally based on Cell analysis method platform.

Figure 7 Classification and application methods of therapeutic proteins

In 2018, the FDA counted the application methods for the past five years.
Among them, the detection of ADA was mainly based on ECL and ELISA platforms, and the detection of neutralizing antibodies was mainly based on cell experiments and competitive ligand binding experimental platforms.

Figure 8 Statistics of therapeutic protein detection methods declared by the FDA in the past five years

case study

A clinical trial of an antibody project under development enrolled 60 patients, and each patient collected a total of 6 time points (including pre-administration) serum samples for the detection of anti-drug antibodies.

First, through the screening experiment, 9 samples of 6 patients were found to be positive.
Through further confirmation experiments, it was determined that 3 of them were false positives, and 6 of the sera of the other 3 patients were positive for ADA.

Table 5 Summary of screening experiment results

Note: -: screening test is negative; ±: screening test is positive, confirmation test is negative; +: screening test is positive, confirmation test is positive.

The titer experiment was used to characterize the relative content of anti-drug antibodies, and the results showed that the titers of 6 samples ranged from 160-1280.

Figure 9 Titer analysis results of ADA positive samples

To further study whether the generated anti-drug antibody affects the drug efficacy of the drug, an activity inhibition neutralizing antibody detection experiment was established.
All 6 ADA positive samples have neutralizing activity and can block the activity of the drug.

Conclusion

Immunogenicity analysis is getting more and more attention in the research and development of therapeutic protein drugs, and the detection of anti-drug antibodies in clinical trials has become a regulatory requirement of the Food and Drug Administration.
Immunogenicity research runs through the entire drug research life cycle, and has become the consensus of the Food and Drug Administration and the field of medical research.
In order to develop more competitive protein drugs, researchers have moved forward the study of immunogenicity, from the drug design stage through various techniques to reduce the immunogenicity of drugs and increase the success rate of drug research.
At the same time, the development and establishment of reliable non-clinical immunogenicity analysis and evaluation methods also provide early guidance for clinical immunogenicity.
It is believed that with the further promotion of the immunogenicity analysis strategy with clinical anti-drug antibody analysis as the core, and the further improvement of the cell immune monitoring and analysis methods, it is bound to promote the process of drug research and development more comprehensively and in-depth.

references

[1] FDA.
immunogenicity testing of therapeutic protein products-developing and validating assays for anti-drug antibody detection.

[2] NMPA.
Technical guidelines for drug immunogenicity research (draft for comments).

[3] Ulitzka M.
, Carrara SC, Grzeschik J.
, et al.
Engineering therapeutic antibodies for patient safety: tackling the immunogenicity problem[J].
Protein Engineering Design and Selection, 2020: 33.

[4] Krishna M.
Product-Related Factors and Immunogenicity of Biotherapeutics[J].
Journal of Pharmaceutical Innovation, 2019, 15(49):1-13.

[5] Erik D.
and Huub S.
.
Immunogenicity of Innovative and Biosimilar Monoclonal Antibodies[J].
Antibodies, 2019, 8: 21.

[6] Schellekens H.
and Jiskoot W.
.
Immunogenicity of Therapeutic Proteins[J].
Pharmaceutical biotechnology, 2013, 6: 133-141.

[7] Davda J.
, Declerck P, Hu-Lieskovan S.
, et al.
Immunogenicity of immunomodulatory, antibody-based, oncology therapeutics[J].
Journal for ImmunoTherapy of Cancer, 2019 7:105.

[8] Schreitmüller, Thomas, Barton B, Zharkov A, et al.
Comparative immunogenicity assessment of biosimilars[J].
Future Oncology, 2018, 15.

[9] Shankar G.
Current challenges in assessing immunogenicity[J].
Bioanalysis, 2019, 11(17):1543-1546.

[10] Osman N Yogurtcu, Zuben E Sauna, Joseph R McGill, et al.
TCPro: an In Silico Risk Assessment Tool for Biotherapeutic Protein Immunogenicity[J].
The AAPS Journal, 21(5):96.