Monitoring Antibody Drug Safety and Efficacy in Preclinical and Clinical Development
John Cardone, Marketing Manager, Custom Antibodies, Bio-Rad Laboratories.
Antibody-based therapeutics are a powerful modality developed for a range of diseases, including cancer, autoimmune disorders, and infections with over 170 antibodies in regulatory review or approved currently. An inherent high target specificity and affinity, often associated with fewer adverse events, and potential for modification and refinement through genetic engineering has aided in the recent clinical success of antibody drugs.
However, antibody therapeutic development remains a long, complex, and costly process. Close observation of drug safety and efficacy through pharmacokinetic (PK) assessment is critical to progressing candidates from discovery to clinical trials. At the same time, regulatory agencies now require patient immune response to the drug to be monitored throughout this process. An immune response resulting in the generation of anti-drug antibodies (ADAs), which bind the drug and neutralise its activity or accelerate its biological removal, can cause adverse events, impacting drug safety, and could cause a loss in efficacy.
To determine the best antibody candidates for preclinical and clinical progression, highly specific and highly sensitive assays that inform on a drug’s safety and efficacy, as well as the potential to trigger an immune response, are essential.
What are the Challenges in PK and ADA Assay Development?
The development of robust PK and ADA assays comes with several challenges, especially for human or humanised therapeutic monoclonal antibodies. When administered to the patient, the concentration levels of human or humanised antibody drugs can be a million times lower than the human serum antibody concentration, making the drug difficult to detect and quantify. PK assays, used to measure free and/or target-bound drug levels in patient serum samples, therefore rely on highly specific and sensitive detection reagents that bind the drug and not other similar antibodies.
Likewise, an appropriate surrogate positive control or reference standard is used to accurately assess anti-drug immune response in an ADA. In the clinical stage of development, these assays should use multiple antibody controls with varying affinities and/or immunoglobulin subclasses to offer a more accurate representation of a patient’s natural immune response. Reflecting on ADA assay sensitivity, which is defined as the lowest concentration at which the antibody concentration consistently produces either a positive result or is equal to a pre-determined readout, the US Food and Drug Administration (FDA) recommends that such assays achieve a sensitivity of at least 100 ng/ml. This level of sensitivity should be sufficient to enable detection of ADA’s before they reach hazardous levels associated with altered pharmacodynamic, PK, safety, or efficacy profiles. Traditional methods, such as electrochemiluminescence, surface plasmon resonance, and enzyme-linked immunosorbent assay (ELISA), have struggled to meet this guideline. Such assays are often faced with low sensitivity and low drug tolerance and may detect only unbound ADAs, resulting in false negative results.
Anti-idiotypic antibodies (i.e., an antibody specific to the idiotope of another antibody) are recognised as optimal detection reagents in PK assays and as surrogate positive controls or reference standards in ADA assays. However, traditional sources for anti-idiotypic antibodies, including rodent monoclonal anti-idiotypic antibodies and primate polyclonal serum, bring several challenges. For example, rare specificities for drug-target complex specific antibodies would be extremely challenging to identify through immunisation. Moreover, affinities of IgG antibodies generated through immunisations are difficult to determine given their bivalent format (i.e., each Ig monomer contains two antigen-binding sites); monovalent formats (e.g., monovalent Fabs) are the ideal antibody format to define intrinsic affinity given each Fab monomer contains one antigen binding site.
Alongside ethical considerations for the use of laboratory animals, successful immunisation is not guaranteed with primate polyclonal serum. Although human sera obtained from early clinical trials does present an alternative, obtaining material consistently and in the quality and quantity needed is difficult, and continual assay validation is required throughout the clinical development stage.
The Solution: Recombinant Monoclonal Anti-Idiotypic Antibodies
Sophisticated antibody libraries, specifically created by de novo synthesis to represent the structural diversity of the human antibody repertoire and optimised for expression in E. coli, combined with phage display technology have paved the way for recombinant monoclonal anti-idiotypic antibodies. These antibodies are already well-established in the development of human therapeutics and various other research settings and alleviate some of the common challenges in antibody production. Generated using fully in vitro processes, recombinant anti-idiotypic antibodies offer greater predictability, flexibility, and opportunity for optimisation, and guarantee a long-term supply of sequence-defined, fully human antibodies, without the use of animal-derived components or laboratory animals.
By implementing different screening strategies, phage display technology enables the generation of anti-idiotypic antibodies with varying binding modes specific to different forms of the antibody drug (i.e., free, target-bound), and with varying affinities (Figure 1). Selection carried out in the presence of isotype subclass-matched antibodies leads to the creation of paratope specific (i.e. antigen binding site-specific) anti-idiotypic antibodies (i.e., Type 1). These antibodies are inhibitory as they do not allow the drug to bind its target and can be used to measure free drug as well as positive controls in ADA assays. Alternative selection strategies can produce a non-inhibitory anti-idiotypic antibody specific to an idiotope outside the drug’s antigen binding site (i.e., Type 2), which can be used to detect total drug – free, partially bound, and fully bound. Finally, guided selection can be used to create antibodies capable of detecting rare specificities, such as drug-target complex binders (i.e., Type 3). By using a drug-target bound antibody as bait throughout phage display, whilst the library is alternatingly blocked using individual components of the complex, antibodies that recognise a feature unique to the drug-target complex can be isolated.
Type 1
Inhibitory
Paratope-specific
Detects free drug
Type 2
Non-inhibitory
Not paratope-specific
Detects total drug (free, partially bound, fully bound)
Type 3
Non-inhibitory
Drug-target complex specific
Detects bound drug exclusively
Figure 1. Binding modes and properties of anti-idiotypic antibodies.
Armed with a complete toolbox of Type 1, Type 2, and Type 3 anti-idiotypic antibodies, PK and ADA assay developers have greater flexibility and, for PK assessment, a more complete analysis of the availability and state of the antibody candidate. For example, combining Type 2 or Type 3 anti-idiotypic antibodies with an antigen capture ELISA provides a solution to detecting monovalent antibodies or Fab fragments in serum, such as ranibizumab, which cannot be detected using Type 1 anti-idiotypic antibodies in an ELISA bridging format due to the drug only have one binding domain.
Similarly, quantitating levels of intact natalizumab has been historically challenging. Natalizumab, a humanised IgG4 monoclonal antibody drug used to treat multiple sclerosis and Crohn’s disease, undergoes half-antibody exchange in vivo with other antibodies of the same isotype, resulting in molecules comprising one drug heavy-light pair coupled to one heavy-light chain pair of unknown specificity. Using recombinant anti-idiotypic antibodies instead of the antigen to capture antibody drug, two ELISAs have been designed to successfully assess natalizumab levels in vivo, one ELISA measuring total antibody drug (i.e., intact and exchanged molecules) and the other measuring intact antibody drug only. Validation studies confirmed that the PK assays were specific, accurate, and precise.
Phage display technology also allows a panel of anti-idiotypic antibodies with high, medium, and low affinity to the target drug to be selected, representing the variation in human immune responses to a drug, critical to ADA assay development.
Conclusion
Robust drug safety, efficacy, and immunogenicity assessment are essential to therapeutic antibody development. Recombinant anti-idiotypic antibodies, generated in vitro through sophisticated antibody libraries and phage display technology, provide a reproducible, long-term supply of the highly sensitive and highly specific detection reagents and controls needed for PK and ADA assays respectively. These anti-idiotypic antibodies are an innovative solution for faster assay optimisation and more effective resource and cost management during preclinical and clinical stages, avoiding continual assay validation.