Part 3 of 3: The Importance of Developability Assessments – Utilizing Late-Stage Assessments to Ensure Clinical through Commercial Success

In the first two parts of this 3-part series, our experts, Dr Rob Holgate & Dr Gary Watts, explored how early and mid-stage developability assessments underpin the rational design and selection of biologic drug candidates. However, as you get closer to lead candidate nomination, the purposes of assessments change. Rather than seeking to eliminate large numbers of unsuitable candidates as quickly as possible, the aim is now to provide a much more detailed deep dive into a smaller number of potential candidates to ensure they are fit for purpose. 

This becomes increasingly important as molecules scale up as part of their CMC strategy and move toward the clinic since new challenges emerge – ones that can only be addressed by rigorous later-stage developability evaluations.  

In this final article, we detail how these assessments enable lead candidate selection, de-risk clinical progression, support regulatory submissions, and ultimately help drive late phase commercial success. 

Developability

As described in Part 2, the best molecule is not always just the one with the best in vitro potency; rather, it is the candidate with the optimum combination of biological activity, safety, and manufacturability.

To achieve this, a holistic view is required, focusing on three primary aspects:

1. Binding & Function: Whatever the format, whether it’s an antibody, bispecific, or bioconjugate, we need to ensure it binds appropriately to the target. In some cases, additional functionality may also be required – e.g., the presence (or absence) of effector function may be key to the molecule’s function or safety profile. Addition of a payload will further change the functionality.
2. Immunogenicity: We need to ensure that the molecules that we’re developing show no undesirable immune responses and have an acceptable safety profile.
3. Manufacturability: We are looking for issues that could affect the ability to be manufactured on a large scale as well as any impact on long-term stability and integrity.

Transient vs Stable Production

As discussed in Part 2 of this blog series, early-stage studies are often performed using material from transient expression systems because they are often faster and more cost-effective. However, these systems are typically used as surrogates for the final manufacturing cell line. This distinction becomes especially important for complex molecules, where transitioning between different hosts can reveal unexpected issues. Moving into the final manufacturing system as early as possible therefore improves translatability into CMC and helps derisk the overall program.

Adopting stable pools can offer several important advantages:

• Higher product titres: Stable systems typically improve expression levels.
• Improved product quality: This is particularly valuable for complex biologics such as bispecifics.
• More representative material: Stable pools based on the manufacturing cell line provide material that better reflects later-stage development needs.

For this reason, it is becoming increasingly common to introduce stable pools based on the manufacturing cell line earlier in development – for example, by taking multiple clones into stable pools, such as with our AbZelectPRO™ CHO-K1 cell line.

Functional & Immunogenicity Assessments

Whereas early-stage assessments focus on “simple” biochemical or cell-based assays, assays and assessments at later stages tend to be more in-depth. However, as you progress to more relevant biological models such as 3D models or organoids, the complexity, and variability (and cost) increases. While the functional assays are likely to be very target and MoA-specific, there are a number of common assessments that should be considered at this stage, especially from a safety perspective.

Examples of these types of assays include:

• Effector function assays: For antibodies, testing ADCC (antibody-dependent cellular cytotoxicity), CDC (complement-dependent cytotoxicity), and other Fc-associated mechanisms.
• Cytokine release assays: Detecting the risk of cytokine storm, especially important for immunomodulatory therapies.
• Off-target activity: Screening for unintended binding or effects in human tissues.

When it comes to Immunogenicity, even well-engineered and stable molecules can trigger immune responses in humans. Later-stage immunogenicity assessments go beyond the in silico predictions used at earlier stages of R&D and focus on using in vitro or ex vivo cell-based assays, using e.g. PBMCs from a diverse panel of donors, to assess risk:

• In vitro assays: Human PBMC (peripheral blood mononuclear cell) assays such as EpiScreen™ 2.0 measure T cell activation and cytokine release in response to the candidate.
• Ex vivo analyses: Use samples from human donors to detect rare but clinically relevant immune reactions.
• Anti-drug antibody (ADA) assays: Developed and validated to detect and characterize patient antibodies against the biologic during clinical trials.

Derisking Manufacturability

Manufacturability and Lead Candidate Selection

Early-stage manufacturability focuses on screening many potential candidates but typically where quantities of material available are very often limiting (typically μg to low mg). The progression to clinical and commercial manufacturing moves things on to a completely different scale, often dealing with Kgs of material and 1000s of liters of culture. Early-stage manufacturability studies, while valuable, may not fully predict behavior in larger-capacity bioreactors or during large-scale purification, where the stresses placed on the molecule are quite different from those at bench scale. Later-stage manufacturability assessments focus on stress-testing panels of candidates using studies designed to simulate a range of CMC conditions that they will face:

• Thermal stress: Incubation at elevated temperatures is used to assess susceptibility to unfolding, aggregation, and degradation and is often used as a predictor of long-term stability
• Freeze-thaw cycles: Assess susceptibility to aggregation due to temperature cycling by performing multiple freeze-thaw cycles and are used to mimic material handling and storage during manufacture & early clinical studies.
• pH excursions: These are used to simulate exposure to acidic or basic conditions during purification. Low pH studies (~pH 3.5) are used to assess a molecule’s ability to withstand a low pH hold (e.g., during viral inactivation), whilst high pH studies may be used to assess the molecule’s susceptibility to deamidation and to identify degradation pathways.
• Mechanical agitation: Assesses molecule sensitivity to downstream processes (such as chromatography and filtration) used during manufacturing as well as shipping stresses.
• Light exposure: Assesses samples for susceptibility to daylight illumination, and is especially relevant for products in transparent containers.

Along with the Immunogenicity and Functional activity studies described, these Manufacturability studies provide a deep insight into how a molecule behaves and allows effective triaging and lead candidate selection from a larger panel of leads.

Formulation Development

The manufacturability studies do not, however, simply stop because you have a lead candidate, and this is where formulation development steps in. To be viable as a potential candidate, a protein must maintain its colloidal stability, chemical integrity, and therapeutic potency not just during the manufacturing process, but also through long-term storage, shipping, in-use handling etc.

With a recommendation that molecules have a minimum solution stability of at least 2 years, obtaining real-time stability data is inherently time-consuming, and so the early stages of formulation development focus on using a series of accelerated stability stress studies that aim to mimic and predict how the drug will behave in over an extended time period, ensuring that any potential issues can be addressed before significant investments are made.

Central to formulation development is determining the optimal balance of different components. Initially, an assessment is performed to identify the best buffer, pH and excipient combinations and explores a wide range of conditions typically spanning pH, charge and tonicity, before performing an extended stability assessment on the leading formulation.  If required, additional components such as viscosity reducing agents and antioxidants may also need to be considered. Once established, the stability studies applied are broadly similar to those performed as part of candidate selection (e.g. thermal stress and freeze thaw) however are likely to be more extensive in scope(e.g. broader temperature range and extended study duration).

Viscosity and High-Concentration Formulation

With the rise of subcutaneous biologics, high-concentration formulations are becoming increasingly desirable. However, many proteins exhibit problematic viscosity at high concentrations, hindering syringeability and patient compliance. Late-stage assessments designed to assess suitability include:

• Maximum Solubility Determination: The highest concentration at which a biologic stays fully dissolved, monophasic, and stable in a formulation.
• Colloidal stability and B22 and kD Analysis: Provide an assessment of the propensity to self-associate.
• Viscosity profiling: Measuring at clinical and commercial concentrations.
• Formulation optimization: Screening excipients and buffer systems to minimize viscosity and aggregation.
• Injectability & Syringeability Studies: Provide an objective and subjective assessment, respectively, of the injection forces involved as well as the physical practicality of injection.

If material is available, performing these studies as part of lead candidate selection is highly recommended; however, due to the high material requirements, this is not always feasible.

Considerations for Complex Modalities – Moving Beyond Monoclonal Antibodies (mAbs)

The biologics landscape has progressed from traditional monospecific mAbs and now encompasses bi- and multi-specific antibodies, Antibody-drug conjugates (ADCs), Antibody-oligonucleotide conjugates (AOCs), antibody fusion proteins, and other novel modalities that support the next generation of medicines.

However, each advance in design brings unique manufacturability challenges:

• ADCs & AOCs: Bioconjugates typically require precise conjugation and rigorous monitoring for payload stability, free drug, and drug-to-antibody ratio (DAR). Additionally, conjugation of different payloads to mAbs will change their biophysical properties – whether through increased hydrophobicity (as often the case with cytotoxic payloads) or increased charge (as with oligos). Understanding how this will impact the stability of the overall molecule is essential.
• Bispecifics & Multispecifics: Many BsAbs utilize antibody building block fragments such as scFv as part of their design. However, when compared with mAbs or Fabs, it is well accepted that scFvs provide challenges due to their lower stability and increased propensity for aggregation. Assessing how these molecules behave during lead candidate selection is even more important to avoid unwanted surprises.

Similarly, the asymmetric designs often found in bispecifics can lead to complex manufacturing processes. Often, the potential for mispairing represents a particular challenge due to multiple heavy or light chains, and so, characterization techniques such as SEC, CE-SDS, cEIF and LC-MS are essential to assess product heterogeneity and help guide purification strategies.

• Fusion proteins and novel scaffolds: Fusion proteins, including Fc-fusion proteins, typically contain multiple domains from different proteins joined together. These domains are often taken out of their natural context and as a result, when combined together, can lead to instability, creating manufacturing challenges. For example, different domains may demonstrate differing pH sensitivity which could be problematic, e.g. for some Fc-fusion proteins where a low-pH elution step is typically used as part of the Protein A capture during purification.

Similarly, some fusion proteins may also be more prone to proteolysis during production, potentially due to the junction sequence or linker sequence between domains. Investigating different linker sequences and lengths may improve this.  Protein domains, emphasizing the importance of investing time upfront at the design stage.

Analytical Challenges

Complex modalities necessitate a broad toolbox of analytical techniques, with orthogonal methods – each measuring a different property – being critical to building a comprehensive picture of stability and manufacturability. These analytical methods are continually evolving, with increasingly complex modalities requiring new methods to be developed. However, these analytics are not just a nice-to-have! As well as being invaluable for supporting candidate selection, regulatory agencies increasingly expect such depth of analysis in IND/IMPD submissions prior to clinical trials.

Embedding Developability into the Biologics Pipeline

The Cost of Finding Liabilities Late

Failures discovered during, or after, scale-up can delay programs by months or years and potentially cost millions. In contrast, late-stage developability assessments enable informed, timely decision-making – whether to proceed, re-engineer, or halt a candidate, allowing effective risk management.

A Culture of Continuous Assessment

For a program to be successful, it needs to be recognized that developability is not a one-time hurdle to overcome but a continuous discipline. Embedding developability at every phase – from discovery to IND and beyond – requires:

• Regular cross-functional review
• Transparent data sharing and risk registers
• Clear “go/no-go” criteria based on predefined TPP and CDTP specifications

Best Practices

• Start early, reassess often: Make developability an integral part of every milestone. Don’t put it off so that it becomes a late-stage CMC problem.
• Combine multidisciplinary expertise: Leverage the combined knowledge of protein engineers, analytical scientists, formulation scientists, and regulatory strategists.
• Document and communicate: Maintain detailed records of liabilities, mitigation steps, and decision rational for internal alignment and regulatory defense.

Summary

Going back to the beginning and to Part 1 of our developability series, it is important to establish a clear vision from the outset of where you want your molecule to go. Having a promising idea and target product profile (TPP) with a clear vision allows us to anticipate early on what the key risks may be and allows for the mitigation of those as soon as possible.

Developability is not a static process; it must be applied continually, taking many forms spanning manufacturability, immunogenicity, binding, and function. Later-stage developability assessments are the linchpin that connect promising laboratory discoveries to successful clinical and commercial products. They transform risk into actionable insight, de-risking scale-up, ensuring patient safety, and maximizing the likelihood of regulatory approval.

Companies that prioritize continuous, rigorous developability assessments across all stages of development are best positioned to deliver safe, effective, scalable, and accessible biologic medicines to patients in need.

Developability Series Q&A

• How does manufacturability assessment impact formulation strategy?Manufacturability assessments can be used to identify risks such as deamidation or oxidation. If these can’t be engineered out, formulation can potentially be used as a way of mitigating them (e.g., adding antioxidants for oxidation, adjusting pH for deamidation). Early detection ensures you select the right excipients and formulation conditions to manage these liabilities for robust drug development.
• If all candidates look similar after manufacturability assessment, how do you select a lead?There are many factors to consider beyond simply manufacturability, for example, immunogenicity as well as binding and efficacy, and often you are looking to select the candidate with the optimal balance of developability properties. There may also be other factors to take into account, such as IP considerations.
• Is titer part of the manufacturability assessment?Not formally, but titer can act as an indicator of stability. Low titers can flag manufacturing challenges and can help deprioritize weaker candidates. It is, however, important to ensure you have an appropriate benchmark or historical data to determine comparative expression in the relevant system.

——————————————-

This concludes our three-part series on Developability Assessments. If you would like a compiled version, visual summaries, or further expansion on any topic, please get in touch with the team at Abzena today by clicking here.