Bioconjugate process development is inherently complex, requiring a careful balance between biomolecule stability, optimal reaction conditions, payload solubility, and product-related impurity clearance. As these variables are interconnected, decisions made early in development have important downstream considerations.
In many programmes, process development is separated from full optimisation due to material constraints and stringent timelines leading up to first-in-human clinical trials. While often inevitable, this approach can introduce risk if not properly supported by strong platform knowledge, targeted experimentation, and robust data.
Early decisions can have far-reaching implications for scalability, process robustness, and overall timelines. Without a clear understanding of how different variables interact, challenges that are difficult or costly to correct later may arise. Thoughtful process design and a Quality by Design (QbD)-focused strategy play an important role in mitigating these risks.
In this blog, we’ll outline five important considerations for a risk-based, QbD-aligned approach to ADC process development that balances speed with long-term success.
1. Design of Experiments (DoE)
Why it matters
ADC processes involve multiple interacting parameters, such as pH, temperature, and mixing rate, and testing a single variable at a time is not sufficient for building deep process understanding. DoE is essential to clearly and efficiently understand how critical process parameters (CPPs) influence critical quality attributes (CQAs).
Key considerations for effective DoE:
- Clearly linking CPPs to CQAs
- Identifying and defining acceptable operating ranges that balance reaction performance with molecular stability across development stages
- Avoiding unnecessary stress on the molecule while achieving key process objectives
DoE enables simultaneous evaluation of multiple parameters and their interactions, supporting evidence-backed decisions and reducing reliance on trial-and-error.
In a recent enzymatic conjugation project at Sterling, DoE was used to optimise a single-pot reaction containing three enzymes, one co-factor, and one substrate. This ultimately enabled complete conversion of the antibody to the activated form along with a 40% reduction in cost of goods, demonstrating how targeted experimental design can deliver both technical and economic advantages in bioconjugation programmes.
2. Purification
Why it matters
Early-stage purification strategies are generally designed to be fit-for-purpose to ensure effective impurity control while maintaining process scalability and robustness. Full optimisation can later follow.
Key considerations for optimising purity:
- Understanding the criticality of process and product related impurities
- Prioritising clearance of the most critical impurities
- Leveraging platform knowledge to establish conservative purification strategies early on
- Applying core techniques such as tangential flow filtration (TFF), chromatography, and activated carbon filtration
- Demonstrating purification performance under worst-case conditions to de-risk scale-up
The early use of conservative, well-understood purification approaches helps to ensure reliable impurity control while enabling later-stage optimisation, such as increasing TFF membrane load, to further optimise efficiency and cost-effectiveness. Additionally, established platform knowledge can significantly reduce purification development effort.
In one example involving auristatin payloads, Sterling applied predefined TFF conditions, requiring only two or three gram-scale runs to modify the process prior to 250-fold scale-up. In another involving exatecan payloads, where TFF clearance is known to be poor, GMP-compatible, cost-effective carbon filters were used for rapid screening to allow for reliable initial scale-up runs.
3. Mixing
Why it matters
For bioconjugation processes where reactants can vary significantly, the criticality of mixing rates depends on the particular project. Highly soluble reagents with slow reaction kinetics, such as many enzymatic reactions, may only require effective initial mixing, while lower solubility payload linkers must often be “pulled” into solubility during conjugation. This makes constant mixing critical during this phase to prevent solids settling and incomplete reactions.
Key considerations for ensuring appropriate mixing:
- Using scale-down mixing models that accurately represent large-scale manufacturing equipment
- Understanding the relationship between mixing time, homogenisation, and reaction rate; fast stochastic reactions, for example, require more control over mixing
- Monitoring reaction progress and product quality as scale increases and mixing equipment design and characteristics change
There is no one-size-fits-all approach to mixing for bioconjugation programmes, and mixing must be carefully controlled to prevent shear-induced damage to the conjugate.
In a project with a low-solubility payload; a scaled-down mixing model was developed to optimise addition rate and mixing conditions. Initially, inappropriate mixing resulted in precipitation of the payload and low drug load. After identifying critical mixing parameters, the characteristics were replicated in manufacturing-scale equipment through surrogate experimentation, and the process successfully scaled to ~1kg input.
4. Filtration
Why it matters
A well-designed filtration strategy is required for all bioconjugation processes to support clarification and microbial control. Mapping process risks at each stage is necessary to identify where filtration is required to reduce process or quality-related challenges in subsequent steps.
Key considerations for successful filtration:
- Selecting filters based on specific process risk (e.g. 0.2 μm filters for microbial risk or depth filters for particulate risk)
- Sizing filters using real process intermediates, in-real time
- Establishing overload or challenge data to confirm sizing
- Standardising on ready-to-process, single-use filters with acceptable chemical compatibility for typical co-solvents
- Oversizing filters or installing redundant parallel capsules
Filter sizing should be continuously reassessed across protein lots, payloads, equipment types (e.g. peristaltic and quaternary diaphragm pumps), and scales.
For example, in one ADC process at Sterling, development filtration data did not scale during pre-clinical toxicology lots. Extensive root cause analysis identified shear sensitivity to peristaltic pumps, which increased particulate formation. Quaternary diaphragm pumps were subsequently standardised across development and manufacturing scales to ensure filtration robustness.
5. Process intermediate stability
Why it matters
Conditions that are optimal for conjugation chemistry are not always optimal for product intermediate stability. Understanding where process intermediates can and cannot be held is imperative for process design and manufacturing scheduling.
Considerations for maximising stability:
- Determining stability profiles of the ADC, starting materials, and intermediates
- Using stability data to design manufacturing controls and schedules
- Leveraging literature and experiential learning to identify known stress factors
- Studying extended process hold times to support quality investigations in case of unexpected delays
In one programme, a conjugate required column chromatography with a low-pH elution step to achieve a specific quality attribute. This conjugate was shown, through hold studies, to be colloidally unstable at low pH. This data was used to develop a neutralisation step prior to forward processing and establish stringent time limits, ensuring product quality and manufacturing control.
Building robust and scalable ADC processes
Bioconjugate process development requires the right balance between speed, robustness, and scalability. Mixing, stability, purification, DoE, and filtration, must be addressed holistically to reduce downstream risk.
A QbD-aligned strategy that is supported by representative scale-down models and phase-aligned decision making helps scientists build understanding early and progress projects confidently. Close collaboration across sites and teams is equally important, whether supporting mAb production, payload linker manufacture, bioconjugate process development or manufacturing.
This risk-aware, collaborative approach is core to ADC process development and clinical manufacturing at Sterling. By applying expertise in process design, purification strategy, and ADC discovery through clinical manufacturing, we help customers proactively de-risk scale-up early in their ADC programmes. If you’d like to learn more about how we can support your project, speak to an expert.
