Early-Development Assessment Tools: De-Risking Strategies for Biologics
Tech Snapshot captures today’s cutting-edge tools and technologies that will help drive drug discovery tomorrow. This installment was written by Yvette Stallwood, Ph.D., Head of Applied Protein Services, Lonza Biologics, and was first published by MassBio on October 20, 2020.
The complexity of today’s biologics means the risks of failure in the development process are higher than ever. With new molecule classes and more variations in functional structures, there are many unknowns to determine in the development process.
One way of de-risking this process is to employ early-development assessment tools. Utilizing early-stage assessment and characterization analysis at the right times and for the right parameters for your molecule and program can be a difficult decision to make. The development activities you perform is always a balance of budget, timing, and risk-tolerance. Many of these assessments can be relatively inexpensive, particularly when you consider the costs of re-work, delay to IND timeline or even program failure if you miss critical molecule information at an early stage. In any case, the net savings in time and funding can make the difference between keeping a development program moving forward and cancelling it.
There are many types of assessments that can be carried out to help de-risk your molecule. These include in silico and in vitro immunogenicity, manufacturability (aggregation, post-translational modifications, and candidate engineering), and protein production. In silico and in vitro tools are faster methods that can be completed at a lower cost, so they are often more suited to a large number of sequences or products/molecules and are more beneficial if used in earlier stages before committing to a final lead. However, any of these methods can be used even if you have already decided on your lead candidate.
There are five key areas to consider for early-development molecule assessments:
The drug discovery and development process is expensive and time-consuming, making efficiency and speed crucial at every stage. If the expression system you select cannot generate the desired attributes, you risk having a product later that is of different quality and/or potentially not easily manufacturable, creating setbacks that lead to added costs and delays. The type of platform you use should be influenced by the type of molecule to be expressed.
Utilizing the platform best suited for the type and complexity of the target molecule allows for further assessment of the product using the same vectors, regulatory approved cell lines, and strains that will be used for manufacturing. Among other factors, material expressed at an early stage can then be used to assess aggregation, functional activity, and immunogenicity.
An in silico manufacturability assessment should include the use of structural models that represent your candidate molecule or techniques to investigate potential post translational modifications (PTMs) and undesirable properties, such as glycosylation and deamidation.
This type of assessment can be run on a single molecule or, preferably, several candidates in order to pinpoint and understand the risks. Identifying a high-risk issue or feature early allows you to put a mitigation plan in place before a problem occurs, such as selecting an alternate lead, re-engineering the molecule, or modifying your process development approach.
In Silico Immunogenicity
All therapeutic proteins display some level of immunogenicity. There are several causes of immunogenicity that stem from both clinical and product-related factors. In silico immunogenicity assessments serve as a predictive tool for identifying whether the molecule presents a high level of risk to the patient. In silico immunogenicity predictions typically focus on identifying high-risk/critical T-cell epitopes. The use of filters can be employed to exclude self-epitopes, thereby reducing over-predictiveness of such tools. In silico and in vitro immunogenicity screening is commonly used during the pre-clinical immunogenicity risk assessment with the data included in regulatory filings.
Humanization of antibodies involves substitution of as many non-human regions for human sequences as possible, resulting in a much less immunogenic molecule. There are multiple approaches for humanizing an antibody, such as germlining or CDR grafting, which reduce the immunogenicity risk. However, greater gains can be achieved by combining humanization with deimmunization, where selected critical T-cell epitopes are also re-engineered, reducing the immunogenicity risk further than humanization alone. By combining antibody humanization and deimmunization with manufacturability assessments in a single round of design, you can further improve your molecule by eliminating/reducing high-risk, T-cell epitopes and liabilities.
By using in vitro assessments ― the analysis of primary human immune cells for their response to the test product ― in addition to earlier in silico immunogenicity assessments, you can evaluate the immune response in a more relevant human system. This provides better correlation with clinical response than animal studies, as well as a more cost-effective system.
By utilizing a variety of early development assessment tools, you can significantly de-risk biologic products before spending significant time and resources on advancing the development of your candidate molecule. The more data and information you can generate early, the better chance you have of selecting only the most promising molecules, optimizing the ones that require it, and eliminating those with the highest risk profiles. Whether you have the in-house resources to do it on your own or you work with an experienced partner, performing even just one of these assessments adds value to your development process and reduces the chance of costly setbacks. And in the race to the clinic, these benefits serve as key advantages that can help increase the value of your asset while simultaneously developing products with the highest level of safety and efficacy.
This installment was was first published by MassBio on October 20, 2020.