Navigating humanization, isotype selection, advanced therapeutic antibody development, and more…
Protein engineering is a critical step in the development of therapeutic antibodies. Almost every facet of an antibody can be modified in one way or other and this FAQ aims to simplify some of the key concepts such as antibody humanization, the importance of affinity and function, isotype selection, and the development of bispecific and antibody-drug conjugates (ADCs). By understanding these fundamental aspects, you can better navigate the advancements and strategies involved in creating effective and safe antibody-based therapies.
Do all antibodies need humanization?
Humanization is the process of changing the sequence of the antibody variable domains to make it appear more human-like and therefore reduce the risk of immunogenicity. This process is crucial for therapeutic antibodies derived from non-human sources to prevent the body from developing anti-drug-antibodies (ADAs) as a result of an immune response against the antibody and which, in turn, could lower the effectiveness of the treatment.
Humanization is typically necessary when therapeutic antibodies are developed using animals like mice, rabbits, or llamas. Advancements such as phage libraries and humanized mice have lessened the need for full humanization. However, even with these methods, some antibodies may still need optimizing to ensure they don’t trigger an immune response.
The challenge for humanization lies in maintaining the antibody’s structure and binding properties while altering its sequence to make it more human-like. At Abzena we have developed a unique approach called Composite Human Antibody™ technology which combines multiple human germline segments to create humanized antibodies with high homology to human germlines and high functionality. In addition, to further assist with the humanization process, we have also developed a proprietary in-silico immunogenicity assessment tool, iTope-AI. iTope-AI scans protein sequences to identify MHC Class II binding peptides (known as T cell epitopes and considered one of the key drivers of immunogenicity!). Abzena’s dual approach to humanization combines removing high-risk MHC Class II binders together with increasing human similarity. Taken together, this approach directly targets the core issue of ADA generation.
What is more important – affinity or function?
The simple answer is that both are as important as each other. Affinity refers to how strongly an antibody binds to its target, while function is about the biological effect of that binding. In some cases, the two go hand in hand, however there are also examples where a higher affinity doesn’t always mean better therapeutic results (e.g. as observed in the binding-site barrier in certain tumours).
Techniques like surface plasmon response (SPR) or biolayer interferometry (BLI) are used routinely in antibody development to measure affinity and are rapid ways of screening antibodies, and so are typically used as an initial readout during development. Functionality is often harder to measure and can require complex bioassays or in vivo models. However, biological assays typically provide the more relevant information.
Both attributes are important and developing an appropriate assay screening cascade is essential to select only the best candidates.
How important is the choice of isotype?
The Fc portion of an antibody is much more than just a scaffold. It can influence the overall function of an antibody and so, choosing the right isotype is a critical decision in the overall development of your antibody.
Ig isotypes can show a broad range of effector functions. For example, IgG1 is capable of exerting potent effector functions (such as Complement Dependent Cytotoxicity (CDC) or Antibody Dependent Cellular Cytotoxicity (ADCC)) which may be desirable for cell killing, whereas IgG2 and IgG4 are preferred when Fc-mediated cell depletion is to be avoided. In addition to the inherent properties of the Ig isotype itself Fc effector function can, in some cases, be dialled up or dampened down even further or differentially altered through engineering of the Fc region to change how the Fc interacts with C1q or Fcγ receptors.
Similarly, various Fc substitutions can also change an antibody’s PK profile through altering the pH-dependent binding to FcRn, resulting in either increased or decreased half-life.
Understanding how you want your molecule to work is essential to allow the right decisions to be made regarding choice of isotype from both an effector function as well as a half-life perspective.
Can amino acids in complementarity determining regions be changed?
The six unique complementarity determining regions (CDRs) of an antibody are critical for the antibody’s ability to bind to its target.
The CDRs are typically found as exposed loops sat on top of a framework scaffold and modification of residues within these loops can have a significant impact on binding /function. In silico analysis of antibody Fv regions is a useful means of identifying a number of potential sequence liabilities (such as deamidation, isomerization and glycosylation) within the CDRs. If potential liabilities are found, it may be possible to surgically remove them by selective substitution, aided by homology models and in silico tools to ensure binding and function is maintained.
By contrast, in some cases a less precise approach may be more appropriate. As mentioned, CDRs are typically responsible for binding however, sometimes the affinity is not sufficient to drive an adequate biological response. In this case, a technique called affinity maturation is often used. Affinity maturation typically uses a library-based approach that allows the generation of large numbers (>108) of variants and the investigation of large regions of sequence space. Library designs can also be customized if needed to maximize the space searched across multiple CDRs.
Other engineering strategies may use a combination of the two and the approach you take is dependent on the question you are trying to answer. However, no matter the approach or the reason for making a particular change, even one amino acid substitution could impact other properties, and so changes should always be viewed in the wider context of developability.
Is it easy to adapt my antibody into a bispecific or ADC format?
Reformatting antibodies into bispecifics or ADCs can extend the therapeutic potential of any given antibody, however the design often requires careful consideration.
When developing bispecifics, there is no “right” format. The overall shape, the avidity of each targeting component and getting the correct balance of affinities for each targeting moiety are just some of the challenges that can influence how well a bispecific works. Manufacturing challenges are also a key consideration since chain mispairing can provide purification challenges and drastically reduce the overall yield of a given bispecific. Typical questions that often arise include what heavy chain heterodimerisation is required (e.g. knob-into-hole technologies?) and how do you maintain the correct VH-VL pairings? Unfortunately, predicting the best format is not simple and so, multiple designs are often required to be made and tested.
When developing ADCs, there are also multiple factors to consider, including choosing the most appropriate conjugation strategy. These can broadly be considered as either using antibodies without (i.e. native) or with engineering. Native antibody approaches include stochastic methods such as lysine conjugation however, the potential for lysines in certain positions (e.g. CDRs) may affect target binding of some mAbs post-conjugation. Other technologies include using interchain cysteine residues, and platforms such as ThioBridge™ have been developed to target and re-bridge the native interchain cysteines, resulting in a stable and homogeneous product. Finally, engineered approaches use site specific handles or tags into a sequence for efficient conjugation but require up front engineering to introduce the conjugation site.
What lessons can be applied at the early stages of antibody development?
The journey from discovery to a licensed product is long and complex. Identifying and solving potential issues early on in the development process can save time and resources and help towards de-risking the overall process.
Applying a good understanding of the desired mechanism of action and good design principles ensures that the desired biological function of an antibody can be maintained and sometimes even enhanced, without compromising safety. However, while there is an obvious focus on biological function of any engineered mAb, the ability of a candidate to be manufacturable should not be underestimated and so the up-front use of in silico tools at an early stage combined with appropriate analytical characterization is playing an increasingly key role in ensuring only the best candidates are progressed to the next stage.
About us
At Abzena, we appreciate that every antibody and every scenario is different. With our extensive experience, we guide customers through the complexities of antibody development, from early identification of potential challenges to successful product creation. Our tailored strategies and innovative approaches aim to meet your objectives and ensure the development of effective, safe, and high-quality therapeutic antibodies.
