OR WAIT null SECS
Process analytical testing for biopharmaceuticals requires enhanced methods due to complex bioprocesses.
One of the challenges with biopharmaceutical production is the need for enhanced process analytical testing techniques. Manufacturing biologics, such as monoclonal antibodies and therapeutic proteins, inherently relies on complex and sophisticated cell-based processes that require close monitoring, particularly during scale-up for commercial production. Sensitive testing and sophisticated data analysis are important for ensuring product quality.
The launch in 2004 of FDA’s guidance on process analytical technology (PAT) set the regulatory framework for biopharmaceutical processes (1). The objectives include increasing process robustness and process understanding and improving yields by way of less rejections and reprocessing to enable batch release in real-time (2).
When a biomanufacturing process has been developed, process characterization studies are typically conducted to define the ranges for various control parameters. These studies involve a number of small-scale bioreactor cultivations.
Of the data generated during these studies, only that which is needed to deï¬ne the acceptable ranges for the control parameters are processed in more detail. Evaluating specific sets of performance data (e.g., metabolite profiles) with a toolbox of advanced statistical methods can be used to forecast the product quality and quantity of mammalian cell culture processes (2).
Though described as a “holistic” approach, the PAT concept is often reduced to the specific aspects of real-time monitoring and multivariate modeling. The many new applications of sophisticated on-line analyzers have since been evaluated, targeting the enhancement of bioprocess monitoring capabilities (2).
Some monitoring techniques typically involve infrared spectroscopy, dielectric spectroscopy, and 2D ï¬uorescence. Other exotic analyzers, such as proton transfer reaction mass spectrometry or electronic noses, have also found their way into bioprocessing.
To correlate the complex signals of most on-line analyzers with meaningful parameters for cultivation, such as viable cell density or metabolite concentrations, the signals of these analyzers must be modeled by multivariate statistics, according to the authors of a study (2).
One way to innovate bioprocesses is to use new analyzers that can thus generate new data sets, but using data sets of routine at-line or off-line analytics by applying more sophisticated data analysis methods is an aspect of PAT that is often neglected, the authors said.
In that study (2), the use of performance-based modeling (PBM) allowed for the prediction in daily intervals of the ï¬nal product concentration and 12 quality attributes (QAs). This approach gave the best forecast for product concentration in an early phase of the process. Some glycan isoforms were also predicted--with accuracy--several days before the bioreactor was harvested.
PBM thus demonstrated its capability to predict the endpoint of the manufacturing process early on by only using commonly available data, even though it was not possible to predict all QAs with the desired accuracy.
“Knowing the product quality prior to the harvest allows the manufacturer to take countermeasures in case the forecasted quality or quantity deviates from what is expected. This would be a big step towards real-time release, an important element of [FDA’s] PAT initiative,” said the authors of the study.
One method explored to increase the efficiency of the most commonly used downstream biopharmaceutical separation process, tangential flow microfiltration (MF), and decrease the risk of MF scale-up involved an integrated approach of using at-line PAT and mass balance-based modeling. MF is a type of tangential flow filtration (TFF) that is known to be a robust and versatile separation technique used in biopharmaceutical manufacturing. MF processes have been traditionally used in the harvest and primary recovery of biopharmaceuticals (3). However, successful scale-up is an important challenge in the implementation of microfiltration processes.
A group of researchers used chromatography-based PAT to improve the consistency of an MF step where filter fouling due to deposition of insoluble material caused a bottleneck in the manufacturing process for a therapeutic protein (3). A reverse-phase ultra-high-performance liquid chromatography (RP-UHPLC) assay was developed to provide at-line monitoring of protein concentration. The method successfully resulted in the highlighting of specific areas where the manufacturing process could be improved by showing areas of divergence from a mass balance-based model.
Following illumination of the specific areas for improvement, process controls were appropriately adjusted, resulting in better operability and significantly increased yield (3). The approach presented in the study is applicable to reducing the risk during scale-up filtration processes in general and is considered suitable for feed-forward and feed-back process control, according to the authors of the study.
The researchers showed that leveraging UHPLC technology to reduce the run time of liquid chromatography-based methods increased the sampling frequency, which enabled more rapid feedback. “Because PAT systems are highly automated, the cost and time impact associated with the increased sampling is minimal. As shown in this report, feed-back control during MF processing could enhance detectability of filter fouling to allow processing to be paused for filter regeneration. Feed-back control could also enable adjustments of operating parameters to mitigate further fouling,” the report authors said.
They concluded that PAT is an attractive option for improving consistency of MF processes and enabling more efficient scale-up. Though MF is a cost-effective and robust separation technique in the downstream purification of a therapeutic protein, full scale implementation has been hindered by the empirical, trial-and-error nature of scale-up. With the application of a UHPLC-based PAT method to lessen the scale-up risk, the MF process was optimized, leading to improved yields that were sustained over time, and which also aligned with historical laboratory-scale practice (3).
1. FDA, Guidance for Industry, PAT--A Framework for Innovative Pharmaceutical Development, Manufacturing, and Quality Assurance (Rockville, MD, Sep. 2004).
2. T. Schmidberger et al., “Progress Toward Forecasting Product Quality and Quantity of Mammalian Cell Culture Processes by Performance-Based Modeling,” Biotechnol. Progress online, DOI: 10.1002/btpr.2105 (June 14, 2015).
3. D.S. Watson et al., “At-Line Process Analytical Technology (PAT) for More Efficient Scale Up of Biopharmaceutical Microfiltration Unit Operations,” Biotechnol. Progress online, DOI: 10.1002/btpr.2193, Nov. 17, 2015.
Volume 30, Number 10
When referring to this article, please cite as F. Mirasol, “The Challenges of PAT in the Scale-Up of Biologics Production,” BioPharm International 30 (10) 2017.