Optimizing CMC strategy for microbial processes
Managing process complexity requires tailored Chemistry Manufacturing and Control (CMC) strategies at all development stages. Successfully filing a license application with microbial-expressed molecules can therefore be complex and challenging, especially for small biopharma companies, which may have limited experience in filing. Fortunately, several strategies can address this challenge.
Managing Diversity In Microbial Processes
The molecules produced by microbial expression are diverse and include natural proteins of various sizes, such as enzymes, hormones or cytokines, as well as new molecular formats such as nanobodies or other antibody fragments, up to non-protein molecules and modifications such as plasmid DNA, PEGylated products or conjugated polysaccharide vaccines. All these molecules can be expressed in a variety of host-/vector combinations and purified through various unit operations, resulting in many different bioprocess shapes and sizes.
The range of complex biological molecules is on the rise and this propels growth of microbial expression platforms. Microbial manufacturing often offers advantages such as fast development timelines and accelerated manufacturing processes, high expression levels, and favorable media costs compared to mammalian cell culture.
Forming a strategic partnership with a seasoned contract development and manufacturing organization (CDMO) such as Lonza, with a history of filing success will instill confidence in the quality of the filing and can expedite the path towards dossier submission. This paper covers novel approaches to accelerate commercial filings and handle the complexity of CMC and the filing process.
Planning CMC Strategy
One of the most critical factors for any biopharmaceutical company is the time to market of a new molecule, which is driving efforts to reduce the time to drug approval. Conventionally, process transfer and production in launch scale would be in an advanced state before starting validation activities. Such a sequential approach minimizes risk, however it is also time consuming, with the time from transfer into a launch scale production asset to a submission-ready filing dossier easily taking three to four years.
Efforts for timeline reduction can additionally be triggered by, for example, the EMA's Priority Medicines Initiative (PRIME) and the FDA's Breakthrough Therapy (BT) designation program.
Reducing the timeline can be achieved by staggered approaches, where late-stage CMC activities (such as transfer to the final launch scale and process characterization) are initiated while late-stage CMC-activities are still ongoing. This strategy saves time but may increase risk, and so special care must be taken when following this parallel path.
Figure 1
Sequential and parallel approaches to commercialization.
A Structured Planning Approach
Meticulous planning is key to support commercial filing and to timeline acceleration. At Lonza we have structured this into five steps. Whilst in reality the projects are more fluid, by thinking in these stages we, in essence, tailor activities to each product and process and to our customers’ commercial strategy, timeline expectations and risk tolerance.
In the five phases, we specifically:
- Align the strategy with customers and carefully evaluate the possibilities for a serial versus a parallel approach, taking into account the customer’s appetite and the process readiness.
- Define the general approach in an initial plan and draw up timelines to determine which activities in the workflow can be staggered.
- Assess the process data and know-how to deep dive into detailed scope and determine if time-saving options available are viable, using a risk-benefit analysis.
- Refine the plan and implement options for optimizing timelines for all CMC activities again. This is because some of the options considered in step one may no longer be viable. We use this information to produce a detailed, customized project plan.
- Execute the practical work, balancing timeline objectives while managing risks.
Figure 2
Lonza’s five-step methodology for planning and executing the activities involved
in commercialization in a paralleled approach.
Three Case Studies: Timesaving Approaches
To demonstrate what the 5-step methodology with its tailored analysis and planning can achieve for a specific program, we have outlined three real-life examples, where time savings were identified and implemented to reduce the time to dossier submission.
Figure 3
Activities in commercialization program with highest potential for time savings.
In the case studies, we explain that the greatest potential for reducing timelines was identified in the activities we refer to as ‘Risk assessment 1’, ‘Scale Down Model Qualification (SDMQ)’ and ‘Process Characterization (PC)’. Process validation studies are already staggered and therefore do not offer great potential for time savings.
Case Study 1: Bring forward the risk assessment
In addition to practical validation, the theoretical assessment of potential failures and the associated risk of insufficient product quality and/or process performance is a key element. We undertake this in two distinct phases: ‘Risk Assessment 1’ is a starting point to all activities and defines the scope of PC studies, with ‘Risk Assessment 2’ (not shown in Figure 3) performed prior to at-scale validation batches (Process Performance Qualification (PPQ)-batches).
The risk assessments consider topics such as impurities and process- and equipment capabilities, and help to determine appropriate controls and mitigation strategies (to minimize risks). Cause-and-effect analysis software is used to map the process parameters versus the critical quality attributes (CQAs) and process performance, and we assign the effect of each parameter to each quality attribute. In addition, failure mode and effect analysis (FMEA) is applied for criticality ranking. With this information, each attribute is ranked numerically to establish how critical it is. If its ranking is high, it is considered a potential critical process parameter (pCPP), which should be included in the PC-studies.
There are opportunities here for time saving: In the traditional sequential approach, Risk Assessment 1 begins after all data of at least three GMP batches at launch-scale are available to serve as reference for the product’s critical quality attributes. However, process characterization can begin without completion of launch-scale batches. Therefore, if Risk Assessment 1 can start earlier, which can potentially saving one to two months. To do this requires leveraging process knowledge gained in the lab and the facility, along with historical data such as early development or manufacturing batch information. This knowledge can come from those scientists experienced on the process and often from external CDMO partners, if the process was developed outside of Lonza.
The time saving from bringing forward Risk Assessment 1 comes with a risk, since the assessment may not include all data from launch-scale batches. However, should minor parameter changes be needed to optimize quality or performance at scale, Risk Assessment 1 can be updated, ensuring accuracy. While there may be a need to repeat certain experiments during PC studies, our extensive experience with various molecules and process behaviors during transfer allows us to anticipate and effectively manage and mitigate this risk.
Figure 4
Timeline saving by bringing forward Risk Assessment 1.
Case Study 2: Bring forward scale down model qualification (SDMQ)
Besides the practical validation, the theoretical assessment Scale down models (SDMs) allow for the characterization of acceptable process parameter ranges at a small scale, which saves cost and time through parallelized experiments. To assure a representative parameter characterization valid to determine the full scale production process, a qualification of the SDM is required. It’s a prerequisite for every process characterization study and is called ‘Scale down model qualification’ (SDMQ).
SDMQ involves running processes at large and small scale and comparing the critical quality attributes of the product to determine if there is a scale-dependent difference. These potential differences can, however, be included in the data analysis and conclusions of the small-scale PC studies, provided it is accurately quantified, e.g. at least three data points per scale to obtain statistically meaningful results. SDMQ is really important because the process parameter setpoints and ranges derived in small scale are ultimately applied at-scale in the validation batches; Failure in SDMQ jeopardizes the success of the PPQ campaign.
It is becoming increasingly common for process characterization to start before three batches from the launch scale are available (Figure 5). In our experience, this approach is especially favored by small biotech customers seeking cost savings, since conducting three pre-GMP batches of material that may not be needed for clinical supply represents a financial burden. However, from a scientific point of view, fewer at-scale data points can be highly problematic if the parameter setpoints and ranges derived with this approach cannot be validated in the three batches of the PPQ-campaign.
Despite these risks, bringing SDMQ forward through use of as little as two GMP-batches at launch scale can save 1 month (or even more, depending on manufacturing schedules) in a commercial filing, and this is very appealing to our customers.
To accurately perform SDMQ predictions with data from just two GMP batches, several precautions must be taken. These include having a deep understanding of the correlation between the SDM and manufacturing scale, which is achieved through many years of technology transfer experience and an deep knowledge of the correlations between the two process scales. This ensures that conclusions drawn using two data points from at-scale GMP batches, are real and not scientific artifacts.
The limited statistical validity of the accelerated SDMQ must be compensated with additional measures. These are applied after the PPQ batches, where we re-evaluate the validity and conclusions of the early SDMQ assessment, at a stage with less timeline pressure.
Figure 5
Timeline saving by bringing forward Risk Assessment 1 and SDMQ.
Case Study 3: Minimizing process characterization scope with separate process understanding studies
PC studies define the set points and ranges of process parameters for the PPQ batches and later commercial production. These studies can be time consuming due to the number of unit operations and parameters to be investigated. The process stage that determines the scope of this characterization is Risk Assessment 1. Insufficient process data and knowledge have the greatest negative impact: unknown effects on product quality make unit operations and parameters potentially critical, requiring their inclusion in the PC study for risk mitigation. In the end, the PC study often examines many more parameters than really needed. This is time consuming as all these experiments require protocols and reports with quality assurance (QA) approval. The same is true for each individual unit operation in scope. Each one requires an SDM, which needs qualification. As a result, if process knowledge is limited when Risk Assessment 1 is prepared, PC-studies can mushroom in size and time.
The solution here is self-evident: Increase process knowledge before starting to define the scope of the PC study experiments, with our recommended approach being to implement a dedicated process understanding study to fill knowledge gaps before conducting Risk Assessment 1. As shown in Figure 6, these studies can be executed very early during process development, or during tech transfer where the process comes to Lonza via another CDMO.
As these studies look at key process impacts rather than defining GMP manufacturing parameters, they can potentially be carried out with less administration and cost effort.
Figure 6
Timeline saving by bringing forward Risk Assessment 1 and SDMQ, and performing process
understanding studies prior to Risk Assessment 1.
The knowledge gained from the process understanding studies is fed into the Risk Assessment 1 cause-and-effect analysis, so that the scope of the PC study can be focused on only the potentially critical process parameters. Having a good understanding of the process before defining the scope of the PC study can save up to several months. Whilst we consider this a low-risk approach, conducting process understanding experiments requires extensive knowledge and experience of microbial processes to anticipate potential issues. It also requires the right equipment and capacity to support many parallel lab activities. Finally, from the perspective of our customers, it requires an "invest early, save later" mentality, as time savings of several months will only be realized later during PC. Thus, the final strategy needs to be tailored to the specific project and the consideration of the associated risks.
Summary: Accelerated Approaches To Commercial Filing
The evolving complexity of biopharmaceutical programs, particularly those involving microbial expression systems with its high diversity of molecules produced necessitates tailored strategic and nuanced approaches to many aspects of CMC, including the final commercial filing.
An experience-founded strategy such as the 5-step methodology, as described in this paper, exemplifies a proactive approach, balancing time efficiency with risk management. This methodology allows for significant time savings by aligning Customer/CDMO strategies early, optimizing timelines through careful planning, and executing tasks in a manner that mitigates risks inherent to parallel processes.
The three case studies presented highlight practical examples of how this methodology can be applied to accelerate commercial application filings. By advancing key processes such as Risk Assessment, Scale-Down Model Qualification, and Process Understanding, substantial potential for reducing timelines can be achieved without compromising on quality or compliance. However, these strategies also require a deep understanding of microbial processes and a willingness to invest early for long-term gains.
In conclusion, while microbial processes are complex, they don’t have to be challenging. The right combination of strategic planning, innovative methodologies and expert partnership paves the way for faster, more efficient dossier preparation and ultimately quicker time-to-market for new therapies.
Accelerate Your Path To Commercial Filing
Partner with Lonza to streamline CMC strategies for microbial processes and reduce time to market. Talk to a CMC expert today.
